content
stringlengths 7
2.61M
|
|---|
package com.gamecraft.web.rest.util;
import java.io.BufferedWriter;
public class IrcBotListener {
public static void sendString(BufferedWriter bw, String str) {
try {
bw.write(str + "\r\n");
bw.flush();
}
catch (Exception e) {
System.out.println("Exception: "+e);
}
}
}
|
After opening at 73.85 per USD, the rupee made a cautious recovery of 18 paise in the early morning trade on Tuesday, on fresh selling of the American currency by banks and exporters.
The rupee touched an intra-day high of 73.85 per US dollar on Tuesday, Bloomberg data showed..
Extending its free-fall, Indian rupee touched a fresh low of Rs 74.26 against the US dollar in intra-currency trades on Tuesday, an increase of 18 paise from its previous close of 74.07, on the back of elevated prices of crude oil, amid weak domestic stock markets and foreign capital outflows. The rupee has fallen 2% in October and 16% in 2018.
After opening at 73.85 per USD on Tuesday, the rupee made a cautious recovery of 18 paise in the early morning trade, on fresh selling of the American currency by banks and exporters. However, it, later on, turned weak and declined to trade at a fresh lifetime low of 74.27 against the US dollar. It touched an intra-day high of 73.85 per US dollar on Tuesday, Bloomberg data showed.
On Monday, the domestic currency declined by 30 paise to close at a record low of 74.07 against the US dollar.
Meanwhile, Brent crude breached the $84 a barrel-mark again and the US dollar strengthened against its global peers.
Stock exchanges – Sensex and Nifty – traded rangebound in the afternoon session on Tuesday, with Tata Motors shares plunging nearly 20% to seven-year low levels. The BSE Sensex fell 189 points to an intra-day low of 34,285.06 points, while the NSE Nifty slipped to a low of 10,279.35 points intra-day.
On the other side, Asian shares hit 17-month lows on Tuesday as China allowed its currency to slip past a psychological bulwark amid sharp losses in domestic share markets, a shift that pressured other emerging currencies to depreciate to stay competitive, said a Reuters report.
Also, the International Monetary Fund has also predicted a growth rate of 7.3% for India in the current year of 2018 and that of 7.4% in 2019. In 2017, India had recorded a growth rate of 6.7%. |
package com.twu.biblioteca.factory;
import com.twu.biblioteca.model.Book;
import com.twu.biblioteca.model.Movie;
import com.twu.biblioteca.model.interfaces.Rentable;
import java.time.Year;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.List;
public class RentableFactory {
public static List<Rentable> books() {
return new ArrayList<>(Arrays.asList(new Book("TDD By Example", true, "<NAME>", Year.parse("2002")),
new Book("Head First Java", true, "<NAME> and <NAME>", Year.parse("2003")),
new Book("Rails", true, "R. R.", Year.parse("2003")),
new Book("Python", true, "P. P.", Year.parse("2004")),
new Book("Dumplings", true, "D. D.", Year.parse("2005")),
new Book("Agile", true, "A. A.", Year.parse("2006")),
new Book("English Muffins", true, "E. M.", Year.parse("2007")),
new Book("Coffee", true, "C. C.", Year.parse("2008"))));
}
public static List<Rentable> movies() {
return Arrays.asList(new Movie("Les Miserables", Year.parse("2007"), "Me", 10, true),
new Movie("The Shining", Year.parse("2008"), "You", 9, true),
new Movie("Help!", Year.parse("2009"), "H.", 8, true),
new Movie("The WallStreet Wolf", Year.parse("2010"), "<NAME>.", 7, true),
new Movie("Fight Club", Year.parse("2011"), "F. C.", 6, true),
new Movie("How To Train Your Dragon", Year.parse("2012"), "<NAME>.", 5, true),
new Movie("A Series Of Unfortunate Events", Year.parse("2013"), "<NAME>.", 4, true));
}
}
|
<filename>data-structures/src/test/java/pl/slemjet/stacks/MyStackTest.java
package pl.slemjet.stacks;
import org.assertj.core.api.Assertions;
import org.junit.jupiter.api.Test;
class MyStackTest {
@Test
void test() {
//given
MyStack<String> stack = new MyStack<>();
stack.push("Discord");
stack.push("Udemy");
stack.push("Google");
//when
System.out.println(stack);
Assertions.assertThat(stack.isEmpty()).isFalse();
String first = stack.pop();
System.out.println(stack);
Assertions.assertThat(first).isEqualTo("Google");
Assertions.assertThat(stack.pop()).isEqualTo("Udemy");
System.out.println(stack);
Assertions.assertThat(stack.pop()).isEqualTo("Discord");
System.out.println(stack);
Assertions.assertThat(stack.isEmpty()).isTrue();
stack.push("Hello");
System.out.println(stack);
Assertions.assertThat(stack.isEmpty()).isFalse();
}
} |
iPad Surprise: Lower TV Episode Prices in iTunes?
Any hope we had that iPad rumors would end with the actual announcement were quickly dashed, of course, but some are better than others: Sources tell The Wall Street Journal and Financial Times that Apple is in talks with studios to drop the price of a TV show episode download to $1 to coincide with the release of the iPad at the end of next month – half what standard definition shows now cost on iTunes, and one-third of current HD versions.
Those of us who expected the device to be more about content than hardware were a tad disappointed with the unveiling last month. But the good thing about a two-month wait is that deals can close, just in time. Content creators who were not part of the launch, and those which had a mere three weeks to decipher the iPad SDK, now have a quiet period to get their acts together.
Which brings us back to TV, anytime, anywhere, and the iPad, which sports a screen larger than any mobile phone and is smaller and lighter than any netbook.
Slashing the price of television shows by 50 percent (as Apple has already done for titles including Wonder Showzen, Celebrity Paranormal Project and Children's Hospital, according to WSJ) won't be enough to convince someone to buy an iPad. But deals like this, taken in aggregate, could make the actual iPad much more alluring as the casual "read-only" device Apple has positioned it to be.
The iPad's screen would make these cheaper television shows look great in a portable setting, and it stands to reason that a price drop this big would increase the sales of television shows, to an extent. Just one problem: Nobody over the age of 10 wants to buy television shows on a regular basis. Aside from one's absolute favorites, there's really no reason to re-watch a television episode more than once.
Apple CEO Steve Jobs has said for years that people want to buy music and rent videos, and not vice versa. If he's right, this reported price drop represents a band-aid rather than a full solution for Apple's television strategy. According to The Wall Street Journal, Apple initiated these talks around the idea of a subscription television service, and while those talks have stalled, Apple still hopes to offer one.
Pressure from television and movie studios has forced Apple "to avoid linking new TV services to its Apple TV device" in the past, according to the FT's sources, because if Apple threatens their bread-and-butter cable and satellite businesses, it could see its television and film contracts unrenewed. However, the iPad, unlike Apple TV, does not compete directly with cable and satellite boxes. |
/* compiled from: Schedulers */
static final class C13778a {
/* renamed from: a */
static final C13805u f41928a = new C13708b();
} |
International application WO2005/007185 describes attempts to stabilize protein pharmaceuticals without the addition of the often used stabilizer human serum albumin (HAS). The stabilizing solution employed in WO2005/007185 instead comprises (i) a surface-active substance that is preferably a non-ionic detergent, i.e. a surfactant and (ii) a mixture of at least two amino acids, wherein the at least two amino acids are either glutamate and glutamine or aspartate and asparagine. The sole example testing the stabilizing effect of amino acids on their own (table 2) shows that in the absence of the surfactant polysorbat 80, no or only insubstantial stabilization—for a limited amount of time—of the protein solution was found.
In the international application WO 2008/000780, a spray dried powder containing protein is stabilized and has an advantageous aerodynamic behavior when at least 30% or at least 40% phenylalanine are included. Due to the addition of phenylalanine in the powder, the cohesive and adhesive properties of the powder are altered to reduce the interactions between the particles. By rendering the surface of the powder particles more hydrophobic, the aerodynamic properties of the powder are improved, thus rendering it more suitable for pulmonary application. The induction of folding or prevention of unfolding of proteins by a solution containing amino acids is not envisaged in WO 2008/000780.
European Patent application EP 1789019 describes spray dried powders of protein drugs for pulmonary application that are stabilized by the addition of novel oligosaccharide mixtures. International application WO 2010/151703 discloses a pharmaceutical composition for increasing the stability, reducing aggregation or reducing immunogenicity of a peptide or polypeptide, comprising at least one alkylglycoside.
In summary, it is crucial that the formulation scientist has a thorough knowledge of several factors: how to optimize the physical and chemical stability of the active ingredient; how, rationally, to include specific excipients in the formulation; how to obtain the optimum conditions for stability; how to prevent stability issues during up-scaling; and, finally, how to design a formulation that is suitable for the intended route of administration, that is, one that allows the absorption barrier to be overcome. The choice of excipients is often based on previous experience and on which excipients have been approved by authorities. Excipients are generally chosen (i) to ensure a successful end point for a preparation process, for example, allowing a dry powder to be obtained; (ii) to ensure that a liquid formulation remains at constant pH value; or (iii) to stabilize the protein against a certain production induced effect, for example adsorption. However, what may be useful for one protein can have detrimental effects on another.
The large size of proteins, their compositional variety and amphiphatic characteristics constitute specific behavior such as folding, conformational stability, and unfolding/denaturation. The structural differences among different proteins are so significant that generalization of universal stabilization strategies has not been successful to date. Despite the fact that a lot of research is invested into providing suitable methods to prevent unfolding and/or ensure proper folding of proteins during the exposure to the different forms of stresses, there are still problems with e.g. aggregation or increased unspecific immunogenicity due to protein misfolding. Even more, to date, the basic properties of a protein usually need to be examined prior to the development of a stable protein formulation. These properties include protein purity, pI as well as solubility and stability at different pH and in different buffer systems. With these data, protein formulation issues are usually addressed.
While it is desirable to keep the number of functional excipients as low as possible, for example in order to reduce instability resulting from solid state interactions, the vast amount of potential destabilizing effects often necessitates the addition of several excipients. Furthermore, for each protein of interest, cumbersome investigations are necessary to identify the respectively suitable excipients. Consequently, there is still a need to provide improved means to fold proteins or to prevent their unfolding during the exposure to the different forms of stresses upon the production and storage process that overcome these problems and are suitable for the majority of proteins and peptides, eliminating the need for individual investigations.
This need is addressed by the provision of the embodiments characterised in the claims. |
// Copyright (c) 2015-2020 <NAME>
// Use of this source code is governed by the MIT license that can be found in the LICENSE file.
#ifndef missing_index_error_hpp
#define missing_index_error_hpp
#include <string>
#include <boost/filesystem/path.hpp>
#include <boost/optional.hpp>
#include "user_error.hpp"
namespace octopus {
/**
A MissingIndexError should be thrown when an index associated with a user specified file
does not exist.
*/
class MissingIndexError : public UserError
{
public:
using Path = boost::filesystem::path;
MissingIndexError() = delete;
// When no associated index could be located
MissingIndexError(Path associate, std::string type);
// When an index was given, but is non-existent
MissingIndexError(Path associate, Path given_index, std::string type);
virtual ~MissingIndexError() override = default;
private:
virtual std::string do_why() const override;
virtual std::string do_help() const override;
Path associate_;
boost::optional<Path> given_index_;
std::string type_;
};
} // namespace octopus
#endif
|
<reponame>Overnap/overnap.com
import styled from "@emotion/styled"
import { graphql } from "gatsby"
import React from "react"
import Layout from "../components/Layout"
import SEO from "../components/SEO"
import Tag from "../components/Tag"
import Title from "../components/Title"
import { TagsQuery } from "../graphqlTypes"
const Line = styled.div`
margin: 11px 0px;
display: flex;
align-items: center;
`
interface Props {
data: TagsQuery
}
const Tags = ({ data }: Props) => {
const groups = data.allMarkdownRemark.group
if (!groups) {
return <div>Error!</div>
}
groups.sort((a, b) => { return b.totalCount - a.totalCount })
return (
<>
<SEO title='Tags' />
<Layout>
<Title>Tags</Title>
{groups.map(tag => (
<Line>
<Tag tag={tag.fieldValue!} />
- {tag.totalCount} posts
</Line>
))}
</Layout>
</>
)
}
export default Tags
export const pageQuery = graphql`
query tags {
allMarkdownRemark {
group(field: frontmatter___tags) {
fieldValue
totalCount
}
}
}
`
|
Andrew Richardson-USA TODAY Spor
UPDATE 3, 4:16 p.m. ET:
Marc Ratner, UFC Vice President of Regulatory Affairs, confirmed with Bleacher Report that Eye's fight against Alexis Davis will indeed move forward on February 22.
Separate sources also told Bleacher Report that the substance Eye failed for was not performance-enhancing, but rather a blood-thinning medication used to treat a long-term issue stemming from an accident when she was hit by a drunk driver at 16 years old.
UPDATE 2, 3:43 p.m. ET:
Bleacher Report has filed an official request with the Texas Secretary of State to obtain access to the Eye administrative order. However, HIPAA privacy laws may prevent the release of the information.
We're also still waiting on official word from the UFC on Eye's status for her UFC 170 bout against Alexis Davis.
UPDATE:
A TDLR official told Bleacher Report that Eye's suspension is considered "probationary in nature" and that she is still allowed to fight at UFC 170 as long as she complies with the terms of the administrative decision. We have reached out to TDLR officials and hope to provide clarification shortly.
UFC bantamweight Jessica Eye has been suspended for one year (effective January 22, 2014) and fined $1,875 for failing a drug test for a banned substance following her UFC 166 win over Sarah Kaufman.
Sources close to the Texas Department of Licensing and Regulation informed Bleacher Report of the result on Monday morning.
Last week, the result of the fight was quietly changed from a win for Eye to a no decision. The TDLR made the administrative decision on January 22, which will also serve as the first date of her suspension.
Eye was scheduled to face Alexis Davis at UFC 170 on February 22.
Andrew Richardson-USA TODAY Spor
Sources close to her camp told Bleacher Report that Eye plans on continuing with the fight and will file a dispute with the Texas commission. Officials with the UFC's public relations department told Bleacher Report that they are looking into the situation and will have an update on Eye's UFC 170 status shortly.
At press time, there is no word what substance Eye failed for or why it took nearly three months for the commission to overturn the decision and issue the suspension.
Eye's victory over Kaufman was a controversial one. Kaufman has been vocal in her belief that she defeated Eye and noted on Twitter last week that she was glad to have the result overturned to something other than a loss.
We'll continue to follow this story and provide updates at the top as we receive them. |
An Arizona woman who allegedly confessed to her boyfriend that she has sex with dogs and fantasizes about incest has been charged with crimes against nature, police said.
Brittany Angelique Sonnier, 20, is accused of committing "vaginal and oral sex with two canines" in Lake Havasu City, according to police Sgt. Joe Harrold.
"We received a report ... alleging bestiality on the part of Brittany Sonnier. Our investigations bureau investigated the allegations and [then] forwarded a criminal complaint to the county attorney's office," Harrold told The Huffington Post.
Sonnier was arrested on November 21 and charged with bestiality. Those charges were later changed to two counts of "crime against nature."
BRITTANY SONNIER PHOTOS: (Story Continues Below)
PHOTO GALLERY Brittany Angelique Sonnier
According to a copy of the police report provided to HuffPost, authorities were tipped off in September by a former boyfriend of Sonnier's about her alleged illegal four-legged activities.
The ex-boyfriend, the report states, told police he had started dating Sonnier in March or April. The ex, whose name has been redacted from the police report, said Sonnier seemed, "like a nice, sweet and innocent girl."
Sonnier's ex told police things went well at first between the two and said he enjoyed spending time with both her and her 11-month-old son. However, during mid-summer the relationship changed, and their sex life began to slow down. It was around this time that Sonnier allegedly approached him and asked if they could "talk."
"They sat down and Brittany used his phone to look some things up," the police report states. "He expected Brittany to ask if they could have a threesome or if he wanted to be a swinging couple. He would have been fine with either of those things — he is very open."
The photos, however, indicated different interests, police said.
"Instead, Brittany started to show him pictures of people having sex with animals," the report states. "She told him that she was interested and into having sex with dogs. She went on to tell him that she has been having sex with their family dogs since she was 13 years old."
The report states one of the dogs is about 13 and the other is just a couple years old. They are not identified by breed, but are described as "medium size" dogs in the police report.
READ THE POLICE REPORT: (Story Continues Below)
Brittany Sonnier
The police report further alleges that Sonnier told her boyfriend that she has had "vaginal and oral sex with the dogs" and has trained the animals so they "don't act strange" around other people.
Sonnier also allegedly said she was into incest and pulled up images on the phone of a family of cartoon characters having sex, police said.
The ex-boyfriend told police he was "disturbed and freaked out" and said he told Sonnier "she was disgusting." The two broke up shortly thereafter.
Sonnier's ex told police he first approached her father and informed him of her alleged bizarre sexual preferences. Sonnier's dad, according to the police report, allegedly told him that his daughter had in the past admitted to him that "she has been having sex with the family dogs." Her dad also allegedly said he is in the process of trying to gain custody of his grandson.
The ex-boyfriend told police he suspects Sonnier is pregnant with his child, but said she has refused to take a pregnancy test. He said that, if she is pregnant, he wants her to get help.
Further allegations by her ex-boyfriend in the police report read: She is a sex addict and he thinks she will be an unfit parent.
He suspects Brittany started to have sex with animals when she was younger because she couldn't find a partner.
He is worried about the child Brittany might be carrying.
Neither Sonnier, her ex-boyfriend nor her father responded to calls and emails requesting comment from The Huffington Post.
According to a Facebook page attributed to Sonnier, she is a 2010 graduate of South View High School in Hope Mills, North Carolina. Under "favorite quotations" Aurdey Hepburn is listed, with the following quote, "I was born with an enormous need for affection, and a terrible need to give it." |
import random
import math
import matplotlib.pyplot as plt
import matplotlib.image as mpimg
import colorsys
import copy
def visualization(machines, jobs, algo):
# Declaring a figure "gnt"
fig, gnt = plt.subplots()
# Setting labels for x-axis and y-axis
gnt.set_xlabel('Processing Time')
gnt.set_ylabel('Machine')
yticks = []
ylabels = []
for i in range(len(machines)):
yt = 15 * (i + 1)
yl = i + 1
ylabels.append(str(yl))
yticks.append(yt)
color = []
for i in range(len(jobs)):
h, s, l = random.random(), 0.5 + random.random() / 2.0, 0.4 + random.random() / 5.0
r, g, b = [int(256 * i) for i in colorsys.hls_to_rgb(h, l, s)]
c = '#%02x%02x%02x' % (r, g, b)
color.append(c)
# print(color)
# color = ["#"+''.join([random.choice('0123456789ABCDEF') for j in range(6)]) for i in range(len(jobs))]
# Setting ticks on y-axis
gnt.set_yticks(yticks)
# Labelling tickes of y-axis
gnt.set_yticklabels(ylabels)
# print(yticks, ylabels)
# Setting graph attribute
gnt.grid(True)
# Declaring a bar in schedule
previous = 0
# for i in range (len(machines)):
# for j in range (len(machines[i])):
# gnt.broken_barh([(previous, machines[i][j][0])], ((i+1)*10, 9), facecolors =(color[machines[i][j][1]-1]), edgecolor = "black")
# previous += machines[i][j][0]
# previous = 0
for i in range(len(machines)):
for j in range(len(machines[i])):
if machines[i][j][1] != 0:
gnt.broken_barh([(previous, machines[i][j][0])], ((i + 1) * 10, 9),
facecolors=(color[machines[i][j][1] - 1]), edgecolor="black")
previous += machines[i][j][0]
else:
if (float(machines[i][j][0])) != 0:
gnt.broken_barh([(previous, (float(machines[i][j][0])))], ((i + 1) * 10, 9), facecolors='white',
edgecolor="black")
previous += (float(machines[i][j][0]))
previous = 0
plt.yticks([])
fig.set_size_inches(37, 21)
plt.title(algo)
plt.show()
plt.savefig("{}.png".format(algo))
# mpimg.imsave("{}.png".format(algo), fig)
def evan(machines, jobs, c):
machines[0].append([jobs[0][0], 1])
high = 0
low = 1
for index2, item in enumerate(jobs):
if index2 == 0:
continue
if makespan_machines(machines[high]) + item[0] - makespan_machines(machines[low]) <= makespan_machines(
machines[low]):
machines[high].append([item[0], index2 + 1])
elif (item[0] / 2 + c > item[0]) or min(makespan_machines(machines[low]) + item[0],
makespan_machines(machines[high])) >= max(
makespan_machines(machines[low]) + item[0], makespan_machines(machines[high])) - min(
makespan_machines(machines[low]) + item[0], makespan_machines(machines[high])):
machines[low].append([item[0], index2 + 1])
else:
machines[low].append([str(makespan_machines(machines[high]) - makespan_machines(machines[low])), 0])
machines[low].append([item[0]/2+c, index2+1])
machines[high].append([item[0] / 2 + c, index2+1])
if makespan_machines(machines[0]) > makespan_machines(machines[1]):
high = 0
low = 1
else:
high = 1
low = 0
def evan_greedy(machines, jobs, c):
low = 0
high = 1
for index2, item in enumerate(jobs):
if item[0]/2 + c < makespan_machines(machines[low]) + item[0]:
machines[low].append([str(makespan_machines(machines[high]) - makespan_machines(machines[low])), 0])
machines[low].append([item[0] / 2 + c, index2+1])
machines[high].append([item[0] / 2 + c, index2+1])
else:
machines[low].append([item[0]/2+c, index2+1])
if makespan_machines(machines[0]) > makespan_machines(machines[1]):
high = 0
low = 1
else:
high = 1
low = 0
def evan_76(machines, jobs, c):
machines[0].append([jobs[0][0], 1])
high = 0
low = 1
job_sum = jobs[0][0]
r4 = 0
for index2, item in enumerate(jobs):
if index2 == 0:
continue
job_sum += item[0]
if makespan_machines(machines[high]) + item[0] <= 7/12 * job_sum:
machines[high].append([item[0], index2+1])
elif max(makespan_machines(machines[low])+item[0], makespan_machines(machines[high])) <= 7/12 * job_sum:
machines[low].append([item[0], index2+1])
elif makespan_machines(machines[high]) + item[0]/2 + c <= 7/12 * job_sum:
machines[low].append([str(makespan_machines(machines[high]) - makespan_machines(machines[low])), 0])
machines[low].append([item[0] / 2 + c, index2 + 1])
machines[high].append([item[0] / 2 + c, index2 + 1])
else:
r4 += 1
machines[low].append([item[0], index2+1])
if makespan_machines(machines[0]) > makespan_machines(machines[1]):
high = 0
low = 1
else:
high = 1
low = 0
# print(makespan(machines)*2/sum([makespan_machines(machines[0])+makespan_machines(machines[1])]))
return r4
def LS(machines, jobs):
min = machines[0][0]
index = 0
for index2, item in enumerate(jobs):
print(machines)
min = machines[index][0]
for i in range(0, len(machines)):
if machines[i][0] < min:
min = machines[i][0]
index = i
insert_job(machines, item[0], index)
machines[index].append([item[0], index2 + 1])
LS_makespan = makespan(machines)
for i in range(len(machines)):
machines[i] = machines[i][1:]
return LS_makespan
def SET(machines, jobs, c):
m = len(machines) - 1
for index, item in enumerate(jobs):
kj = 0
k = min(m, math.sqrt(item[0] / c))
if k == m:
kj = m
elif (item[0] / (math.floor(k)) + (math.floor(k) - 1) * c) <= (
item[0] / (math.ceil(k)) + (math.ceil(k) - 1) * c):
kj = math.floor(k)
else:
kj = math.ceil(k)
machines.sort(key=lambda machines: machines[0])
sj = machines[kj - 1][0]
for j in range(kj):
# if machines[j][0] == 0:
# machines[j].append([sj + (item[0]/(kj)+(kj-1)*c), index+1])
# else:
machines[j].append([str(sj - machines[j][0]), 0])
machines[j].append([(item[0] / (kj) + (kj - 1) * c), index + 1])
machines[j][0] = sj + (item[0] / (kj) + (kj - 1) * c)
SET_makespan = makespan(machines)
for i in range(len(machines)):
machines[i] = machines[i][1:]
return SET_makespan
def LPT(machines, jobs):
jobs.sort(key=lambda jobs: jobs[0], reverse=True)
min = machines[0][0]
index = 0
for index2, item in enumerate(jobs):
min = machines[index][0]
for i in range(0, len(machines)):
if machines[i][0] < min:
min = machines[i][0]
index = i
insert_job(machines, item[0], index)
machines[index].append([item[0], index2 + 1])
LPT_makespan = makespan(machines)
for i in range(len(machines)):
machines[i] = machines[i][1:]
return LPT_makespan
def makespan_machines(machine):
return sum([float(machine[i][0]) for i in range(len(machine))])
def makespan(machine_list):
return max([makespan_machines(machine_list[i]) for i in range(len(machine_list))])
def insert_job(machine_list, job, machine_index):
machine_list[machine_index][0] += job
def main(m, nj):
machines = []
jobs = []
for i in range(m):
machines.append([0])
for i in range(nj):
jobs.append([random.randint(1, 10)])
machines_evan = [[], []]
# machines2 = copy.deepcopy(machines)
# machines3 = copy.deepcopy(machines)
#
# makespan = SET(machines, jobs, 1)
# makespan2 = LS(machines2, jobs)
# makespan3 = LPT(machines3, jobs)
#
# visualization(machines, jobs, "SET Algorithm")
# visualization(machines2, jobs, "LS Algorithm")
# visualization(machines3, jobs, "LPT Algorithm")
# print(makespan)
# for i in machines:
# print(makespan_machines(i), i)
# print(jobs)
# LS(machines, jobs)
# for i in machines:
# print(i)
evan_76(machines_evan, jobs, 3)
print(jobs)
for i in machines_evan:
print(i)
print(makespan(machines_evan)*2/sum([jobs[i][0] for i in range(len(jobs))]))
# visualization(machines_evan, jobs, "Evan algo")
def main_stimulation_2machines():
maxi = 0
max_jobs = []
c_max = 0
jobs_range = 0
no_job_max = 0
no_r4 = 0
max_r4 = 0
for j in range(50000):
nj = random.randint(1000,2000)
jobs = []
jr = random.randint(1,500)
for i in range(nj):
jobs.append([random.randint(1, jr)])
machines_evan = [[], []]
cr = random.randint(2000,3000)
no_r4 = evan_76(machines_evan, jobs, cr)
k = makespan(machines_evan)*2/sum([jobs[i][0] for i in range(len(jobs))])
if k > maxi:
maxi = k
max_jobs = copy.deepcopy(jobs)
c_max = cr
jobs_range = jr
no_job_max = nj
max_r4 = no_r4
print(maxi, no_r4)
print(maxi, max_r4)
print(max_jobs)
print(c_max)
print(jobs_range)
print(no_job_max)
# main_stimulation_2machines()
# c = 4
# epsilon = 0.01
# l = []
# for i in range (300):
# if i == 0:
# l.append([12*c-epsilon])
# else:
# l.append([12*(6**i)*c-epsilon])
# m = [[],[]]
# print(evan_76(m, l, c))
# print(l)
# for i in (m):
# print(i)
# avg = sum(l[i][0] for i in range(len(l)))
# print(len(l))
# print(makespan(m), avg)
# print(makespan(m)*2/avg)
def find_counter(c):
epsilon = 0.0001*c
job = [[], []]
job[0].append(12*c-epsilon)
li = 0
si = 1
for i in range (1000):
# print("Bound:", 7/5 * (sum(job[0])+sum(job[1])) - 12/5 * sum(job[si]), 12*sum(job[li]) + 12*c - 7*(sum(job[0])+sum(job[1])) - epsilon)
k = 12 * sum(job[li]) + 12 * c - 7 * (sum(job[0]) + sum(job[1])) - epsilon
while epsilon/(12*sum(job[li]) + 12*c - 7*(sum(job[0])+sum(job[1])) - epsilon) < 10**(-10):
epsilon *= 10
if math.isnan(k):
break
job[si].append(12*sum(job[li]) + 12*c - 7*(sum(job[0])+sum(job[1])) - epsilon)
if sum(job[0]) > sum(job[1]):
li = 0
si = 1
else:
si = 0
li = 1
# for i in (job):
# print(i)
fin_job = []
for i in range(len(job[1])):
for j in range(len(job)):
fin_job.append([job[j][i]])
return fin_job
max_approx = 0
for c in range(1,50):
l = find_counter(c)
m = [[],[]]
print(evan_76(m, l, c))
# print(len(l))
for i in (m):
print(i)
avg = max(sum([l[i][0] for i in range(len(l))])/2, max([l[i][0] for i in range(len(l))]))
# print(len(l))
# print(makespan(m), avg)
# print(makespan(m)/avg)
if makespan(m)/avg > max_approx:
max_approx = makespan(m)/avg
print(max_approx) |
/**
* A Method that keeps only the blue value of the image
* and sets the Red and Green parameters to 0
*/
public void keepOnlyBlue() {
Pixel[][] pixels = this.getPixels2D();
for(Pixel[] row: pixels) {
for(Pixel col: row) {
col.setRed(0);
col.setGreen(0);
}
}
} |
Effects of centrally administered angiotensin II on salt and water excretion. Adult male rats (n = 34) acclimated to metabolism cages received 5 l artificial cerebrospinal fluid (CSF) with or without 0.5 g angiotensin intraventricularly (IVT). Urinary volume, Potassium excretion sodium and potassium were followed for 3 h. With a suitable recovery time between experiments every rat was randomly exposed to each of 3 test procedures and thus participated in both control and experimental conditions. The procedures consisted of administration of CSF IVT after gentle bladder massage and of 5 ml water by stomach tube CSF-PO group; same, except angiotensin was injected IVT ANG-PO group; same as the ANG-PO group except the stomach tube insertion was a sham procedure; the rats were allowed to drink 5 ml water in response to angiotensin ANG-SPON group. The angiotensin-induced drinking of 5 ml water took 510 min. Except for the 5 ml allowed ANG-SPON rats, water was withheld from all groups during the 3-hour experimental period, and spontaneously voided urine was collected every 15 min. The results of these experiments showed that centrally administered angiotensin produced significant (p |
from __future__ import division
from decimal import *
from Bio import SeqIO
import subprocess
import scipy
import sys
from scipy import stats
import numpy as np
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt
import time
from matplotlib import rcParams
# This file is part of 3D TMH Complexity.
#
# 3D TMH Complexity is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# 3D TMH Complexity is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with 3D TMH Complexity. If not, see <http://www.gnu.org/licenses/>.
print("This file is part of 3D TMH Complexity.")
print("3D TMH Complexity is free software: you can redistribute it and/or modify")
print("it under the terms of the GNU General Public License as published by")
print("the Free Software Foundation, either version 3 of the License, or")
print("(at your option) any later version.")
print("3D TMH Complexity is distributed in the hope that it will be useful,")
print("but WITHOUT ANY WARRANTY; without even the implied warranty of")
print("MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the")
print("GNU General Public License for more details.")
print("You should have received a copy of the GNU General Public License")
print("along with 3D TMH Complexity. If not, see <http://www.gnu.org/licenses/>.")
print("\n\n\nUSAGE:python3 [Script] [file] [TMH number for graph]\nEXAMPLE:python3 Uniprot_to_graphs_and_stats.py file.txt 7 ")
# To do:
# Separate hydrophobicity calculation - more efficient and flexible
# Length restrictions and statistics for final results
# Input files should be obtained in text format downloaded from Uniprot
# and moved to the same directory as this script.
input_filenames = [
str(sys.argv[1])
]
# Parameters for tmh allowances
minimum_tmd_length = 16
maximum_tmd_length = 38
feature_type = "TRANSMEM"
alternative_feature = "INTRAMEM"
# Several dictionaries of hydrophobicity values are needed to verify findings.
white_and_wimley_hydrophobicity = {
'type': ' White and Wimley',
'A': 0.33,
'C': 0.22,
'D': 2.41,
'E': 1.61,
'F': -0.58,
'G': 0.12,
'H': 1.37,
'I': -0.81,
'K': 1.81,
'L': -0.69,
'M': -0.44,
'N': 0.43,
'P': -0.31,
'Q': 0.19,
'R': 1.00,
'S': 0.33,
'T': 0.11,
'V': -0.53,
'W': -0.24,
'Y': 0.23,
}
eisenberg_hydrophobicity = {
'type': "Eisenberg",
'A': 0.620,
'C': 0.290,
'D': -0.900,
'E': -0.740,
'F': 1.190,
'G': 0.480,
'H': -0.400,
'I': 1.380,
'K': -1.500,
'L': 1.060,
'M': 0.640,
'N': -0.900,
'P': 0.120,
'Q': -0.850,
'R': -2.530,
'S': -0.180,
'T': -0.050,
'V': 1.080,
'W': 0.810,
'Y': 0.260,
}
kyte_doolittle_hydrophobicity = {
'type': "Kyte & Doolittle",
'A' : 1.800,
'C' : 2.500,
'D' : -3.500,
'E' : -3.500,
'F' : 2.800,
'G' : -0.400,
'H' : -3.200,
'I' : 4.500,
'K' : -3.900,
'L' : 3.800,
'M' : 1.900,
'N' : -3.500,
'P' : -1.600,
'Q' : -3.500,
'R' : -4.500,
'S' : -0.800,
'T' : -0.700,
'V' : 4.200,
'W' : -0.900,
'Y' : -1.300,
}
### Hydrophobicity calculation code###
def hydrophobicity_calculation(sequence, tmh_locations):
'''
Returns a list of lists, and a list of the hydrophocity scales used in the list of lists.
Each child lists contains a further list of hydrophobicity values
ordered by tmh number. The parent list is used to split the child lists up according to
the different hydrophobicity scales.
'''
list_of_hydrophobicity_dictionaries = [
kyte_doolittle_hydrophobicity, eisenberg_hydrophobicity, white_and_wimley_hydrophobicity]
list_of_tmhs_from_different_hyrodobicity_dictionaries = []
list_of_dictionary_types = []
# tmh locations exists as a string in the format: start,end start,end
tmh_locations = tmh_locations.split(" ")
for hydrophobicity_type in list_of_hydrophobicity_dictionaries:
list_of_dictionary_types.append(hydrophobicity_type.get("type"))
list_of_hydrophobicity_by_tmh_number = []
for n, i in enumerate(tmh_locations):
location = i.split(",")
if len(location) == 2:
# Some sequence positions given are reported as <0 or >N. These cases
# will be treated as length restrictions.
if '<' in str(location):
list_of_hydrophobicity_by_tmh_number.append("null")
elif '>' in str(location):
list_of_hydrophobicity_by_tmh_number.append("null")
else:
start = int(location[0])
end = int(location[1])
# location start and end are human readable locations,
# whereas the slices start counting at 0.
tmh_sequence = sequence[start - 1:end - 1]
# Length restriction and 'X' residues are counted together.
if len(tmh_sequence) < maximum_tmd_length and len(tmh_sequence) > minimum_tmd_length and 'X' not in tmh_sequence:
hydrophobicity_values_in_tmh = []
for residue in tmh_sequence:
hydrophobicity_values_in_tmh.append(
hydrophobicity_type.get(residue))
tmh_hydrophobicity = np.mean(
hydrophobicity_values_in_tmh)
list_of_hydrophobicity_by_tmh_number.append(
tmh_hydrophobicity)
else:
list_of_hydrophobicity_by_tmh_number.append("null")
list_of_tmhs_from_different_hyrodobicity_dictionaries.append(
list_of_hydrophobicity_by_tmh_number)
return(list_of_tmhs_from_different_hyrodobicity_dictionaries, list_of_dictionary_types)
# The bahadur statistic allows statistical comparison of P-values.
def bahadur_statistic(p_value, sample_size):
bahadur_value = (abs(np.log(p_value))) / sample_size
return(bahadur_value)
# Violin plots for datasets.
rcParams.update({'figure.autolayout': True})
def violin_plot(dataset, name, input_file):
'''
Gausian Kernel Density Estimation
Violin plots are similar to histograms and box plots in that they show
an abstract representation of the probability distribution of the
sample. Rather than showing counts of data points that fall into bins
or order statistics, violin plots use kernel density estimation (KDE) to
compute an empirical distribution of the sample. That computation
is controlled by several parameters. This example demonstrates how to
modify the number of points at which the KDE is evaluated (``points``)
and how to modify the band-width of the KDE (``bw_method``).
For more information on violin plots and KDE, the scikit-learn docs
have a great section: http://scikit-learn.org/stable/modules/density.html
'''
fig, axes = plt.subplots(nrows=1, ncols=1, figsize=(5, 4))
# plot violin plot
parts = axes.violinplot(dataset, showmeans=True,
showmedians=True, showextrema=False)
# axes[0].set_title('violin plot')
# Colour palette is colour blind friendly according to Wong B 2011 Points
# of view: Color blindness Nature Methods 8:441.
for pc in parts['bodies']:
pc.set_facecolor('#56b4e9')
pc.set_alpha(0.5)
pc.set_edgecolor('black')
pc.set_linewidth(0.2)
parts['cmedians'].set_color('black')
parts['cmeans'].set_color('#e69f00')
axes.yaxis.grid(False)
axes.set_xticks([y + 1 for y in range(len(dataset))])
axes.set_xlabel('TMH number')
axes.set_ylabel(name)
# add x-tick labels
plt.setp(axes, xticks=[y + 1 for y in range(len(dataset))],
xticklabels=[x + 1 for x in range(len(dataset))])
# plt.show()
#Save bitmap and vector images of figure, then flush figure.
date_time = time.strftime("%H_%M_%S__%d_%m_%Y")
file_name = str(date_time + "_" + input_file + "_" + name + ".pdf")
plt.savefig(file_name, dpi=700)
file_name = str(date_time + "_" + input_file + "_" + name + ".png")
plt.savefig(file_name, dpi=700)
plt.close(fig)
#### Length sorting for TMHs ####
def length_sorting(sequence, tmh_locations):
'''
Returns a list with values of complexities of TMHs.
'''
list_of_tmh_features = []
tmh_locations = tmh_locations.split(" ")
for n, i in enumerate(tmh_locations):
location = i.split(",")
if len(location) == 2:
if '<' in str(location):
list_of_tmh_features.append("null")
elif '>' in str(location):
list_of_tmh_features.append("null")
else:
start = int(location[0])
end = int(location[1])
# Lenght restrictions-these should obey parameters given at
# begining of script.
if abs(start - end) < minimum_tmd_length or abs(start - end) > maximum_tmd_length:
list_of_tmh_features.append(str("null"))
else:
list_of_tmh_features.append(abs(start - end))
return(list_of_tmh_features)
def tmsoc_calculation(sequence, tmh_locations):
'''
Returns a list with values of complexities of TMHs.
'''
list_of_tmh_features = []
# Generates the TMSOC input files from the source file.
with open("TMsegments.txt", 'w') as tm_segments_file:
tm_segments_file.write(tmh_locations)
with open("sequence.fasta", 'w') as fasta_file:
fasta_file.write(">placeholderheader\n")
fasta_file.write(str(sequence))
# Runs TMSOC.
perl_script_output = subprocess.check_output(
["perl", "TMSOC.pl", "sequence.fasta", "TMsegments.txt"])
# Processes the TMSOC input files. Normally the TMSOC output would be a
# multi-line file, not a string.
list_of_output_lines = str(perl_script_output.decode('UTF-8'))
list_of_output_lines = list_of_output_lines.replace("\n", ":")
list_of_output_lines = list_of_output_lines.split(":")
for line in list_of_output_lines:
line = str(line.replace(',', ";"))
line = line.split(";")
# First we can ignore the "junk" lines. This 7 is not referring to the
# number of TMHs, but the number of items in the TMSOC output.
if len(line) == 7:
tmh_sequence = line[0]
start_position = int(line[1])
end_position = int(line[2])
complexity_score = float(line[3])
hydrophobicity_score = float(line[4])
z_score = line[5]
complexity = line[6]
# Lenght restrictions-these should obey parameters given at
# begining of script.
if abs(start_position - end_position) < minimum_tmd_length or abs(start_position - end_position) > maximum_tmd_length:
complexity_score = str("null")
else:
list_of_tmh_features.append(complexity_score)
return(list_of_tmh_features)
# Before we conduct the analysis we must determine how many empty TMH
# lists to make. For example if the protein with the most TMHs in the
# dataset has 13 TMHs, then there will be 13 empty lists made in a list
# representing the 1-13 possible TMH locations.
for input_file in input_filenames:
max_tmd_to_print=int(sys.argv[2])
# These are the parameters used by the biopython Seq.IO module
filename = input_file
input_format = "swiss"
# We need to check against nearby features to prevent overlapping
# flanking regions. Note here we want to avoid clashing with INTRAMEM
# regions, since their flanking regions may be similar, however INTRAMEM regions
# should not be included in the logged transmembrane features since
# they may have extremely short transmembrane sequences which would
# disrupt very much so the alignment of the flanking regions.
# We iterate through each record, parsed by biopython to find the most
# number of TMDs. This avoids list index exceeded errors later on.
tmd_count = 0
for record in SeqIO.parse(filename, input_format):
this_record_tmd_count = 0
for i, f in enumerate(record.features):
if f.type == feature_type:
this_record_tmd_count = this_record_tmd_count + 1
if this_record_tmd_count > tmd_count:
tmd_count = this_record_tmd_count
if this_record_tmd_count > max_tmd_to_print:
print(record.id, " contained more than ", max_tmd_to_print, "TMHs.")
# Generate a list of empty lists, one for each tmh set.
print("Maximum tmh count in", input_file, "is", tmd_count)
list_of_complexity_scores_in_tmh = []
list_of_lengths_in_tmh = []
list_of_hydrophobicity_scores_in_tmh = [[]]
for n in range(tmd_count):
list_of_complexity_scores_in_tmh.append([])
for n in range(tmd_count):
list_of_lengths_in_tmh.append([])
# The hydrophobicity needs to be done separately since it's also dependent
# on the number of scales being used. ### ADDRESS THIS ISSUE. IT CAN BE
# OPTIMISED.###
# Now we can iterate through the records inserting complexity scores into
# the empty lists.
for record in SeqIO.parse(filename, input_format):
transmembrane_record = False
# Sequence fasta file
sequence = record.seq
# TMH positions file
tmh_positions = str("")
for i, f in enumerate(record.features):
if f.type == feature_type:
transmembrane_record = True
tmh_positions = tmh_positions + \
(str(f.location.start) + "," + str(f.location.end) + " ")
# Avoids adding uncleared scores to lists if no TM regions were in
# protein record.
if transmembrane_record == True:
# Adds the complexity score of a helix at (for example the 4th helix)
# to the complexity list of lists (for example in the 4th position)
record_complexity_scores = tmsoc_calculation(
sequence, tmh_positions)
for n, i in enumerate(record_complexity_scores):
# Null entries are added for TMHs that are not within length
# restrictions.
if i == "null":
pass
else:
list_of_complexity_scores_in_tmh[n].append(i)
# print(list_of_complexity_scores_in_tmh)
# Adds the lengths of a helix at (for example the 4th helix)
# to the complexity list of lists (for example in the 4th position)
record_lengths = length_sorting(sequence, tmh_positions)
for n, i in enumerate(record_lengths):
# Null entries are added for TMHs that are not within length
# restrictions.
if i == "null":
pass
else:
list_of_lengths_in_tmh[n].append(i)
# Adds to the hydrophobicity list of lists. The structure is different
# to complexity since there are 3 different hydrophobicity scales.
hydrophobicity_for_record = hydrophobicity_calculation(
sequence, tmh_positions)
for scale_number, hydrophobicity_scale in enumerate(range(len(hydrophobicity_for_record[1]))):
list_of_hydrophobicity_scores_in_tmh.append([])
for n in range(tmd_count):
list_of_hydrophobicity_scores_in_tmh[
scale_number].append([])
for tmh_number, i in enumerate(hydrophobicity_for_record[0][scale_number]):
if i == "null":
pass
else:
list_of_hydrophobicity_scores_in_tmh[
scale_number][tmh_number].append(i)
# Graphs
violin_plot(list_of_complexity_scores_in_tmh[
0:max_tmd_to_print], str("Complexity"), input_file)
violin_plot(list_of_lengths_in_tmh[0:max_tmd_to_print], str("Length"), input_file)
for scale_number, scales in enumerate(list_of_hydrophobicity_scores_in_tmh[0:len(hydrophobicity_for_record[1])]):
violin_plot(list_of_hydrophobicity_scores_in_tmh[scale_number][0:max_tmd_to_print], str(
hydrophobicity_for_record[1][scale_number] + " hydrophobiciity scale"), input_file)
stat_tests_list = [scipy.stats.kruskal, scipy.stats.ks_2samp]
print("\n\n", "###", input_file, "###\n")
for stat_tests in stat_tests_list:
# stats for complexity
for n, i in enumerate(list_of_complexity_scores_in_tmh):
# n is the index, so for human readable numbers we need to add 1. i.e
# the first helix is n=0, so we report it as n+1.
print("TMH ", n + 1)
# print(i)
print("Mean complexity:", np.mean(i), ", N:", len(i))
# list_of_complexity_scores_in_tmh = [
# x for x in list_of_hydrophobicity_scores_in_tmh if x != []]
# Only calculating stats if the sample size is large enough to be worth it.
sample_size_threshold = 10
if len(i) > sample_size_threshold:
if n + 1 < len(list_of_complexity_scores_in_tmh):
for x in range(len(list_of_complexity_scores_in_tmh)):
if x>n and len(list_of_complexity_scores_in_tmh[x]) > sample_size_threshold:
sample_size_for_bahadur = len(list_of_complexity_scores_in_tmh[x]) + len(i)
statistic_test_result=stat_tests(list_of_complexity_scores_in_tmh[n], list_of_complexity_scores_in_tmh[x])
print(sample_size_for_bahadur, statistic_test_result[1])
print("TMH ", n + 1, " to ", x + 1, ":", statistic_test_result, ", Bahadur slope=", bahadur_statistic(statistic_test_result[1], sample_size_for_bahadur))
print("\n")
# Stats for lengths
for n, i in enumerate(list_of_lengths_in_tmh):
print("TMH ", n + 1)
print("Mean length:", np.mean(i), ", N:", len(i))
if len(i) > 10:
if n + 1 < len(list_of_lengths_in_tmh):
print("TMH ", n + 1, " to ", n + 2, ":", stat_tests(
list_of_lengths_in_tmh[n], list_of_lengths_in_tmh[n + 1]))
print("\n")
# stats for hydrophobicity.
# Remove empty lists created previously as a one liner.
list_of_hydrophobicity_scores_in_tmh = [
x for x in list_of_hydrophobicity_scores_in_tmh if x != []]
for scale_number, scales in enumerate(list_of_hydrophobicity_scores_in_tmh):
print("Hydrophobicity scale:",
hydrophobicity_for_record[1][scale_number])
no_empty_tmh_list_of_hydrophobicity_scores_in_tmh = [
x for x in list_of_hydrophobicity_scores_in_tmh[scale_number] if x != []]
for n, i in enumerate(no_empty_tmh_list_of_hydrophobicity_scores_in_tmh):
# n is the index, so for human readable numbers we need to add 1. i.e
# the first helix is n=0, so we report it as n+1.
print("TMH ", n + 1)
print("Mean Hydrophobicity:", np.mean(i), ", N:", len(i))
if len(i) > 10:
if n + 1 < len(no_empty_tmh_list_of_hydrophobicity_scores_in_tmh):
print("TMH ", n + 1, " to ", n + 2, ":", stat_tests(
no_empty_tmh_list_of_hydrophobicity_scores_in_tmh[n], no_empty_tmh_list_of_hydrophobicity_scores_in_tmh[n + 1]))
print("\n")
|
2SPD-036Centralised propofol reconditioning procedure during COVID-19 Background and importanceBecause of the current pandemic, it was necessary to create an intensive care unit (ICU) in our hospital This meant an increase in the consumption of propofol and the associated supply problems It was necessary to develop a procedure to rationalise its use and administration Aim and objectivesTo describe the centralisation in the hospital pharmacy service of the reconditioning of propofol in bags to optimise its administration in the ICU during the COVID-19 pandemic Material and methodsThe ICU contacted the pharmacy service to express the need for higher volume presentations of propofol In response, a literature review was conducted to ascertain the possibility of reconditioning propofol in higher volume containers The stability of propofol in different primary packaging materials was reviewed to select the most appropriate The risk matrix for sterile preparations from the Guide to good practice in the preparation of medicines in hospital pharmacy services was applied to draw up the standard working procedure and to establish the processing conditions, stability of the preparation and storage conditions A centralised propofol reconditioning procedure (CPRP) was established in the pharmacy service: under sterile conditions, transfer the propofol into an ethylene-vinyl-acetate bag to obtain a final volume of 500 mL (10 mg/mL) (using a 0 22 m filter if the initial packaging is glass) It was sealed, labelled and packed in a photo protective bag The established stability was 7 days refrigerated or 30 hours at room temperature A descriptive retrospective study was carried out from its implementation (20 March 2020) to the date of closure of the ICU (5 May 2020) to determine the volume of reconditioned propofol and number of patients treated Data were collected from the electronic medical record and pharmacy programmes ResultsDuring this period, 258 propofol bags were produced Reconditioned propofol was dispensed to 16 patients (median age 59 years (range 4183);62 5% men) The median number of bags per patient was 13 5 (range 366) Conclusion and relevanceThe CPRP in the pharmacy service increases the safety of administration, allows preparation under aseptic conditions and enables the optimisation of available stock As it contains more volume, it facilitates the work and protects the nursing staff by reducing the frequency of contact with the patient References and/or acknowledgementsConflict of interestNo conflict of interest |
<gh_stars>0
import {
GithubClient,
GithubCommit,
GithubConfig,
GithubRepository
} from './Types';
import { Converters } from './Converters';
export class GithubRepositoryImpl implements GithubRepository {
constructor(private readonly githubClient: GithubClient) {}
async commits(githubConfig: GithubConfig): Promise<GithubCommit[]> {
const { repositoryName, orgName } = githubConfig;
const commitsSha: string[] = await this.githubClient.commits(githubConfig);
const commitsWithDetailsPromise: Promise<GithubCommit>[] = commitsSha.map(
commitSha => {
return this.getCommitDetails(repositoryName, orgName, commitSha);
}
);
return Promise.all(commitsWithDetailsPromise);
}
private async getCommitDetails(
repositoryName: string,
orgName: string,
sha: string
): Promise<GithubCommit> {
const commitDetailsResponse = await this.githubClient.getCommitDetails(
repositoryName,
orgName,
sha
);
return Converters.toGithubCommit(commitDetailsResponse);
}
}
|
/**
* Database verifications are used to verify some data is present in some database.
*
* Basic ATS verifications functionality is introduced at:
* https://axway.github.io/ats-framework/Common-test-verifications.html
*
* Database verifications are particularly introduced at:
* https://axway.github.io/ats-framework/Database-verifications.html
*
*/
public class DatabaseVerificationTests extends BaseTestClass {
// the ATS class used to interact with a database
private DbVerification dbVerification;
// we use it to do modify the DB data so our tests can work
private DatabaseOperations dbOperations;
// Here we keep all connection parameters
private TestBox dbServerBox;
// the simple table we work with
private static final String TABLE = "\"People\"";
/**
* Prior to each test we make sure we have the table we worked with
* is in same state
*/
@BeforeMethod
public void beforeMethod() {
// initialize all connection parameters
dbServerBox = new TestBox();
dbServerBox.setHost(configuration.getDatabaseHost());
dbServerBox.setDbType(configuration.getDatabaseType());
dbServerBox.setDbName(configuration.getDatabaseName());
dbServerBox.setDbUser(configuration.getDatabaseUser());
dbServerBox.setDbPass(configuration.getDatabasePassword());
// initialize this helper class
dbOperations = new DatabaseOperations(dbServerBox);
// cleanup the table we use and fill it with the needed data
dbOperations.delete(TABLE, "1=1");
dbOperations.insertValues(TABLE, new String[]{ "id", "firstName", "lastName", "age" },
new String[]{ "1", "Chuck", "Norris", "70" });
dbOperations.insertValues(TABLE, new String[]{ "id", "firstName", "lastName", "age" },
new String[]{ "2", "Jackie", "Chan", "64" });
// Initialize the class we will use in our tests
dbVerification = new DbVerification(dbServerBox, TABLE);
}
@Test( dependsOnMethods = "verifyDataExists" )
public void verifyTableExistsFromFirstTry() {
// just verify there a table with this name
// without paying attention on its content
dbVerification.verifyDbDataExists();
}
/**
* The verification will succeed on the very first try as the content is present.
*
* The verification will succeed only if there is at least one row which
* satisfies all required checks.
*/
@Test
public void verifyDataExistsFromFirstTry() {
// value must be contained or not
dbVerification.checkFieldValueContains("", "People.firstname", "Jackie");
dbVerification.checkFieldValueDoesNotContain("", "People.firstname", "Jacc");
// value must be exact match or not
dbVerification.checkFieldValueEquals("", "People.age", 64);
dbVerification.checkFieldValueDoesNotEqual("", "People.age", 65);
// value must match a regular expression or not
dbVerification.checkFieldValueRegex("", "People.lastname", "Cha.*");
dbVerification.checkFieldValueRegexDoesNotMatch("", "People.lastname", "Chh.*");
// start the verification process
dbVerification.verifyDbDataExists();
}
/**
* When started, the verification keeps failing because the searched person
* is not available.
*
* Then the searched person is available, but his age is wrong.
*
* And finally the age is right and it all matches.
*
* In this test a background thread manipulates the DB data.
*/
@Test
public void verifyDataExists() {
insertDbDataInBackground();
dbVerification.checkFieldValueEquals("", "People.firstname", "Will");
dbVerification.checkFieldValueEquals("", "People.lastname", "Smith");
dbVerification.checkFieldValueEquals("", "People.age", 50);
// start the verification process
// it will pass on the first time when the search data is present
dbVerification.verifyDbDataExists();
}
@Test
public void verifyDataDisappears() {
// insert a person
dbOperations.insertValues(TABLE, new String[]{ "id", "firstname", "lastname", "age" },
new String[]{ "3", "Will", "Smith", "50" });
// deleted that person, but in a few seconds
removeDbDataInBackground();
// wait until that person disappear from the database
dbVerification.checkFieldValueEquals("", "People.firstname", "Will");
// start the verification process
// it will pass on the first time when the search data is not present
dbVerification.verifyDbDataDoesNotExist();
}
/**
* Helper method used to insert data into DB from a background thread.
* This way the main thread is not blocked and the verification works as expected.
*/
private void insertDbDataInBackground() {
new Thread(new Runnable() {
@Override
public void run() {
try {
// insert the expected data, but with not expected age
Thread.sleep(2000);
dbOperations.insertValues(TABLE, new String[]{ "id", "firstName", "lastName", "age" },
new String[]{ "3", "Will", "Smith", "49" });
// now fix the age
Thread.sleep(2000);
dbOperations.updateValue(TABLE, "age", "50", "firstName", "Will");
Thread.sleep(20000);
} catch (InterruptedException e) {
e.printStackTrace();
}
}
}).start();
}
/**
* Helper method used to remove data from DB from a background thread.
* This way the main thread is not blocked and the verification works as expected.
*/
private void removeDbDataInBackground() {
new Thread(new Runnable() {
@Override
public void run() {
try {
// delete the expected data
Thread.sleep(3000);
dbOperations.delete(TABLE, "firstName='Will'");
} catch (InterruptedException e) {
}
}
}).start();
}
} |
import traceback
import tensorflow as tf
import matplotlib.pyplot as plt
import numpy as np
import forward
from helper import *
# 数据加载测试
def data_test():
try:
print('data_test begin')
d = forward.isometric_model()
d.load_from_file("../test_data/data_load_test.json")
print(d.count)
for i in range(d.count):
print(d['idx'][i])
d2 = forward.iso_to_geo(d)
d2.save_to_file('../test_data/data_save_test_geo.json')
except Exception as e:
traceback.print_exc()
log.error(repr(e))
finally:
log.debug("data_test finished")
# cuda测试
@timer
def cuda_test():
g = forward.forward_gpu()
g.init_cuda_device()
g.test_cuda_device()
print("cuda_test finished")
# 正演测试
def forward_test():
try:
# 正演类
f = forward.forward_gpu()
coef = forward.filter_coefficient()
geo = forward.geoelectric_model()
geo2 = forward.geoelectric_model()
iso = forward.isometric_model()
data = forward.forward_data()
# 各种数据加载
coef.load_cos_coef('../test_data/cos_xs.txt')
coef.load_hkl_coef('../test_data/hankel1.txt')
geo.load_from_file('../test_data/test_geo_model.json')
geo2.load_from_file('../test_data/test_geo_model2.json')
iso.load_from_file('../test_data/test_iso_model.json')
data.generate_time_stamp_by_count(-5, -2, 20)
# 测试绘制地电模型
fig = draw_resistivity(geo, geo2, forward.iso_to_geo(iso), last_height=300)
fig.show()
# 正演类加载数据
f.load_general_params(10, 100, 50)
f.load_filter_coef(coef)
f.load_geo_model(geo)
f.load_time_stamp(data)
# 正演开始
f.forward()
# 获得结果
m = f.get_result_late_m()
e = f.get_result_late_e()
m.name = 'late_m'
e.name = 'late_e'
# 绘制结果
fig = draw_forward_result(m, e)
fig.show()
add_noise(m, 0.1)
fig = draw_forward_result(m)
fig.show()
except Exception as e:
traceback.print_exc()
log.error(repr(e))
finally:
log.debug('forward_test finished')
if __name__ == '__main__':
data_test()
cuda_test()
forward_test()
input("Press Enter to continue...\n")
|
<reponame>asanoboy/keras
from __future__ import absolute_import
from .cifar import load_batch
from ..utils.data_utils import get_file
from .. import backend as K
import numpy as np
import os
def load_data(label_mode='fine'):
"""Loads CIFAR100 dataset.
# Arguments
label_mode: one of "fine", "coarse".
# Returns
Tuple of Numpy arrays: `(x_train, y_train), (x_test, y_test)`.
# Raises
ValueError: in case of invalid `label_mode`.
"""
if label_mode not in ['fine', 'coarse']:
raise ValueError('`label_mode` must be one of `"fine"`, `"coarse"`.')
dirname = 'cifar-100-python'
origin = 'https://www.cs.toronto.edu/~kriz/cifar-100-python.tar.gz'
path = get_file(dirname, origin=origin, untar=True)
fpath = os.path.join(path, 'train')
x_train, y_train = load_batch(fpath, label_key=label_mode + '_labels')
fpath = os.path.join(path, 'test')
x_test, y_test = load_batch(fpath, label_key=label_mode + '_labels')
y_train = np.reshape(y_train, (len(y_train), 1))
y_test = np.reshape(y_test, (len(y_test), 1))
if K.image_data_format() == 'channels_last':
x_train = x_train.transpose(0, 2, 3, 1)
x_test = x_test.transpose(0, 2, 3, 1)
return (x_train, y_train), (x_test, y_test)
|
JORDYN Woods is set to lift the lid on her tryst with Khloe Kardashian's boyfriend.
Jordyn Woods has become an outcast from the Keeping Up With The Kardashians family after it was revealed she’d spent the night and acted inappropriately with Khloe Kardashian’s boyfriend.
Earlier this week it was announced that Jordyn will be opening up for the first time about the rumoured tryst with Will Smith’s wife, Jada Pinkett-Smith.
A source close to the 21 year old’s camp spoke exclusively tok OK! online revealing what Jordyn is expected to say in the interview.
The insider said: “Jordyn will discuss the whole situation as openly as she can.
“It’s true that there had been non-disclosure agreements signed by her and her family regards talking about some aspects of the Kardashians.
The source continued: “I fully anticipate Jordyn will admit she did wrong, but she’s going to say she was inebriated, and wasn’t in a good place.
Jordyn’s interview will add much needed clarity to the situation after reports that she was all over Tristan and may have spent the night with him.
The pair are said to have "made out" at a party last week, leaving Khloe “devastated". |
from cv_bridge import CvBridge, CvBridgeError
from sensor_msgs.msg import Image
import rospy
import cv2
import numpy as np
_magic = [0.299, 0.587, 0.114]
_zero = [0, 0, 0]
_ident = [[1, 0, 0], [0, 1, 0], [0, 0, 1]]
true_anaglyph = ([_magic, _zero, _zero], [_zero, _zero, _magic])
gray_anaglyph = ([_magic, _zero, _zero], [_zero, _magic, _magic])
color_anaglyph = ([_ident[0], _zero, _zero], [_zero, _ident[1], _ident[2]])
half_color_anaglyph = ([_magic, _zero, _zero], [_zero, _ident[1], _ident[2]])
optimized_anaglyph = ([[0, 0.7, 0.3], _zero, _zero], [_zero, _ident[1], _ident[2]])
methods = [true_anaglyph, gray_anaglyph, color_anaglyph, half_color_anaglyph, optimized_anaglyph]
class GetImageClass:
def __init__(self,topicName):
self.topicName = '/stereo/' + topicName + '/image_color'
self.image = None
self.image_sub = rospy.Subscriber(self.topicName, Image, self.image_callback, queue_size=1) #/stereo/right/image_color
self.bridge = CvBridge()
def convertROSToCV(self, data):
try:
cv_image = self.bridge.imgmsg_to_cv2(data, 'bgr8')
return cv_image
except CvBridgeError, e:
print e
def image_callback(self, data):
self.image = self.convertROSToCV(data)
def getImage(self):
return self.image
class CreateAnaglyphImage:
def __init__(self):
self._magic = [0.299, 0.587, 0.114]
self._zero = [0, 0, 0]
self._ident = [[1, 0, 0], [0, 1, 0], [0, 0, 1]]
self.true_anaglyph = ([self._magic, self._zero, self._zero], [self._zero, self._zero, self._magic])
self.gray_anaglyph = ([self._magic, self._zero, self._zero], [self._zero, self._magic, self._magic])
self.color_anaglyph = ([self._ident[0], self._zero, self._zero], [self._zero, self._ident[1], self._ident[2]])
self.half_color_anaglyph = ([self._magic, self._zero, self._zero], [self._zero, self._ident[1], self._ident[2]])
self.optimized_anaglyph = ([[0, 0.7, 0.3], self._zero, self._zero], [self._zero, self._ident[1], self._ident[2]])
self.methods = [self.true_anaglyph, self.gray_anaglyph, self.color_anaglyph, self.half_color_anaglyph, self.optimized_anaglyph]
# http://bytes.com/topic/python/answers/486627-anaglyph-3d-stereo-imaging-pil-numpy
def anaglyph(self, image1, image2, method=true_anaglyph):
m1, m2 = [np.array(m).transpose() for m in method]
image1 = np.dot(image1, m1) # float64
image2 = np.dot(image2, m2) # int64
composite = cv2.add(np.asarray(image1, dtype="uint8"), np.asarray(image2, dtype="uint8"))
return composite
|
I ’D MADE IT 375 pages into Niall Ferguson’s newish first volume of a planned two-volume life of Henry Kissinger before receiving in the mail a copy of Greg Grandin’s review of same, in which the author of last year’s excellent Kissinger’s Shadow sums up Ferguson’s tome as follows: “The irony is that it has been Kissinger’s sharpest critics who have most appreciated his acute sense of self, who have treated him, however disapprovingly, as a fully dimensional individual with a churning, complex psyche. In contrast, Ferguson, tone deaf to Kissinger’s darker notes, condemns him to a literary fate worse than anything that Hitchens could have meted out: Kissinger, in this book, is boring.”
This is about as true a thing as has ever been written about any other thing, so much so that I feel both morally and professionally justified in simply abandoning this charmless book unfinished despite having promised to review it at the end of my last column (I would have figured out some other convenient justification for this regardless, but it’s always good to be able to show your work). Nor am I being insulting to Ferguson simply because I disagree with the pro-Kissinger stance he’s taken as the fellow’s authorized biographer and ideological admirer; two years ago I reviewed Kissinger’s own 1,200-page memoir, White House Years, which, though likewise betraying something of a pro-Kissinger stance, was also undeniably compelling and well-written. And while Kissinger is clever enough that one often needs to sort through a great deal of raw material in order to do a proper job of making fun of him, with Ferguson the threshold is somewhat … lower. Here, then, is my review of Ferguson’s 33-page introduction to Kissinger 1923-1968: The Idealist.
“A plainly unhinged woman writing as ‘Brice Taylor’ insists that, when she was a child, Kissinger turned her into a ‘mind-controlled slave,’ repeatedly making her eat her alphabet cereal in reverse order and taking her on the ‘It’s a Small World’ ride at Disneyland,” writes Niall Ferguson, Harvard’s Laurence A. Tisch Professor of History and a Hoover Institution senior fellow, who also scrutinizes Lyndon LaRouche’s claim that Kissinger is a British agent and David Icke’s assertion that he’s a reptilian shape-changer from the lower fourth dimension before concluding, “No rational people take such nonsense seriously. But the same cannot be said for the allegations made by conspiracy theorists of the left, who are a great deal more influential.” The conspiracy theorists of the left, it seems, include not only Oliver Stone but also Howard Zinn and Hunter S. Thompson.
Before things get totally out of hand, as they’re clearly about to, keep in mind that Ferguson was chosen by Kissinger to do this biography 10 years ago, which is to say that Ferguson had a decade to come up with some way of depicting the great bulk of anti-Kissinger sentiment as not only misguided but also malicious and at any rate beyond the pale of American political discourse as usually conducted, and that what follows is nonetheless the best that he could do.
Back to the text:
In his People’s History of the United States, Howard Zinn argues that Kissinger’s policies in Chile were intended at least in part to serve the economic interests of International Telephone and Telegraph. In place of evidence, such diatribes tend to offer gratuitous insult. According to Zinn, Kissinger “surrendered himself with ease to the princes of war and destruction.” In their Untold History of the United States, the film director Oliver Stone and Peter Kuznick refer to Kissinger as a “psychopath” (admittedly quoting Nixon). The doyen of “gonzo” journalism, Hunter S. Thompson, called him a “slippery little devil, a world-class hustler with a thick German accent and a very keen eye for weak spots at the top of the power structure” — adding, for good measure, “pervert.”
So Ferguson promises us influential left-wing conspiracy theorists, which we’re to understand are inherently silly things to be, as if one is not forced into theorizing about conspiracies when one studies a man who conspired to secretly carpet bomb Cambodia, and who did so under the aegis of a presidential administration in which was discussed the viability of assassinating a troublesome newspaper columnist by having LSD applied to his steering wheel.
Still, every allegation must be considered on its merits — something we are unable to do in the case of the single conspiracy theory Ferguson attributes to anyone by name, an allegation supposedly made by Howard Zinn, as Ferguson does not see fit to actually quote it for us. But he does find the space to quote Zinn deploying a disapproving metaphor to describe Kissinger’s decision to go to work for a man he himself had declared not long before to be unfit for the presidency. This, in Ferguson’s accounting, constitutes a “gratuitous insult” on Zinn’s part, whereas referring to another historian’s words as a “diatribe” without having the decency to even reproduce them is presumably not gratuitous at all.
I F ONE BOTHERS to check Ferguson’s footnotes, one comes across the first of several ethical oddities with which his introduction is dotted — for one finds that the “princes of war and destruction” remark is actually from another book written years after People’s History, and thus can hardly be said to constitute an especially good example of an insult offered “in place of evidence,” the place for evidence generally being in the vicinity of the allegation (but we’ll look at the actual evidence in a bit). We are not treated to any better examples down paragraph, where Ferguson quotes Oliver Stone and Peter Kuznik quoting Nixon insulting Kissinger — out of bounds, gentlemen, out of bounds! — without this time deigning even to summarize what allegations the two of them may have leveled at Kissinger to get them thrown in with the likes of Crazy Old Howard Zinn. Hunter S. Thompson makes an appearance as well, unencumbered by anything approaching a reason for being included in what was originally billed as a paragraph detailing unfounded conspiracy theories directed against Kissinger by influential leftists; by now Ferguson seems to have resorted merely to trying to prove that some on the left have insulted Kissinger, or at least quoted Nixon insulting him. To Ferguson’s credit, he does indeed prove this.
Ferguson himself may be aware that he’s promised more than he can deliver here; it may also have occurred to him that he’s just accused several people of failing to back up their allegations with evidence while he himself fails to back up this very allegation with evidence and that this sort of thing might be frowned upon in some circles, if not necessarily at the Hoover Institution. But rather than just deleting this paragraph due to one or more of these several distinct problems, which each on its own makes it worthy of deletion, Ferguson decides to just make it longer. “One left-of-center website recently accused Kissinger of having been somehow involved in the anthrax attacks of September 2001, when anthrax spores were mailed to various media and Senate offices, killing five people.”
Here we have finally been provided with something on the order of an implausible conspiracy theory emanating from the left. And we will be satisfied with this so long as we are willing to overlook the fact that Ferguson promised us that the bearers would be “influential” and yet here cannot bring himself to name the author or even the outlet, presumably in hopes that we won’t realize that both outlet and author are somewhat obscure and that this is merely the same manner of accusation that can be found on any number of minor websites about any number of powerful men.
If we check Ferguson’s endnotes again, we find that he’s referring to a piece by a certain Kevin Barrett titled “Arrest Kissinger for Both 9/11s.” If we check the URL he provides, which comes in the form of a blind bit.ly link-shortener, we find that it’s a dead link. Fair enough; perhaps it’s been taken down since then. If we simply Google the title and author (or have someone else do it for us because we’re in prison), we discover that the piece in question has appeared on a couple of sites, the most mainstream of which would seem to be presstv.ir, itself based on an Iranian host. If we go so far as to read the article, we find that the author does indeed accuse Kissinger of perpetrating the anthrax attacks. But he also accuses him of involvement in “the explosive demolition of the World Trade Center, and massacre of nearly 3,000 people in New York and Washington in 2001,” going on to denounce “Kissinger’s complicity in the coup d’état of September 11, 2001” and noting in passing that the former secretary of state was involved in “helping design the 9/11 shock-and-awe psychological warfare operation.”
One might ask why it is that Ferguson neglected to mention that this “left-of-center” website that’s supposedly mainstream enough to be worthy of inclusion in a paragraph with Zinn and Thompson actually went so far as to accuse Kissinger of involvement in 9/11 itself. After all, that would seem to be the smoking-gun proof that Kissinger really is subject to unsubstantiated allegations from influential leftists, an argument that Ferguson is plainly desperate to make. If we’re feeling gentlemanly, we might allow for the possibility that Ferguson is incapable of understanding what he’s reading — but then that would be something of a knock at Harvard, would it not? So wouldn’t it be even more polite to conclude, as is obvious anyway, that he left this out lest we realize that whatever site he’s taken pains not to name for us isn’t at all “influential” or even mainstream? Because, after all, Harvard?
A S A SORT OF professional courtesy to himself, Ferguson pretends that his case has now been made. “All this vitriol is at first sight puzzling,” he writes presently. A page later, after listing Kissinger’s various awards won and offices held and treaties negotiated, he invites us to ponder with him: “How, then, are we to explain the visceral hostility that the name Henry Kissinger arouses?” That there is to the contrary nothing puzzling about anything Ferguson has shown us and no degree of hostility to be found in connection with the name of this particular American political figure that cannot be found associated with dozens of others of similar prominence becomes even more, rather than less, evident to the extent that one’s been paying attention to Ferguson’s own examples. Hunter S. Thompson famously used similar language about everyone from Hubert Humphrey to his personal acquaintances. Oliver Stone is probably not best known for his reluctance to accuse public officials of involvement in criminal conspiracies (not that we’ve even been told what, if anything, he’s claimed about Kissinger, but whatever). And Howard Zinn has of course been a consistent critic of the American government’s amoral conduct abroad. Indeed, until his death a few years ago, Zinn was probably one of the nation’s most effective mobilizers of popular opposition to the ends-justifies-the-means-and-oops-we-fucked-up-the-ends-too foreign policy establishment that’s so perfectly represented not only by Kissinger, but by such quasi-intellectuals as Ferguson as well. Perhaps this is why Ferguson felt the need to lie about him.
For Zinn did not, in fact, argue that “Kissinger’s policies in Chile were intended at least in part to serve the economic interests of International Telephone and Telegraph,” as Ferguson claims he did, nor does he even imply it. What he actually wrote in People’s History, a copy of which I had sent to the prison from which I now currently serve as an unpaid fact-checker for Penguin, apparently, was this: “And in 1970, an ITT director, John McCone, who also had been head of the CIA, told Henry Kissinger, secretary of state, and Richard Helms, CIA director, that ITT was willing to give $1 million to help the U.S. government in its plans to overthrow the Allende government in Chile.” Elsewhere: “It was also learned from the investigation that the CIA — with the collusion of a secret Committee of Forty headed by Henry Kissinger — had worked to ‘destabilize’ the Chilean government headed by Salvadore Allende, a Marxist who had been elected president in one of the rare free elections in Latin America. ITT, with large interests in Chile, played a part in this operation.”
As these are the book’s only two references to ITT’s involvement in the Chile coup, and as Zinn does not in any way “argue” that those plans were originally composed or thereafter modified with any view to ITT’s economic interests whatsoever, and also taking into account that Ferguson refrained from actually quoting Zinn on this matter while having earlier given David Icke and his friends plenty of space in which to accuse Kissinger of being a shape-shifting lizard mage from the lower fourth dimension who forces children to eat cereal in an incorrect fashion, it’s difficult to avoid the conclusion that Ferguson has chosen to simply lie about another historian who, being dead, is not in a position to defend himself (not that I’m angry about it; on the contrary, this was my original excuse for not finishing the book).
The alternative explanation, again, is that Ferguson is incapable of understanding his sources. But, again, Harvard.
Harvard!
But what about the two assertions that Zinn actually does make? Are they provided without evidence, as Ferguson would have us believe? Not at all. Both of Zinn’s brief references to ITT and Chile, including his single reference to ITT and Chile and Kissinger, are clearly indicated in the text as being drawn from the various post-Watergate congressional investigations into the CIA and the Nixon administration; indeed, both of these Zinn quotes appear in passages that discuss the results of those investigations. The reader may have noticed, for instance, that one of those selections that Ferguson refrains from quoting begins, “It was also learned from the investigation that …” Zinn, obviously, is not “arguing” anything at all, much less putting forth some novel and outlandish “conspiracy theory”; as with the rest of the book, he’s drawing upon the public record — everything Zinn discusses pertaining to the overthrow of Allende, along with much else, can now be found among various online government archives.
At any rate, the Senate’s findings that ITT offered through McCone to assist in the overthrow of Allende didn’t entail any accusation to the effect that the overthrow itself was actually intended even in part to assist ITT; regardless of the extent to which any such offers were motivated by the firm’s economic interests, or ideology, or just the pure joy of overthrowing a democratically elected government, no one involved accused anyone on the planning side, much less Kissinger in particular, of actually tailoring the plot to assist the firm’s bottom line. The only person to have brought this up is Ferguson, in order to portray it as something made up by Zinn; for everyone else, the truth is sufficient.
So Ferguson has falsely accused Zinn of having made a supposedly outlandish claim about Kissinger, whereas in fact, Zinn was closely paraphrasing a Church Committee report published by the U.S. Senate, and implies that Zinn insulted Kissinger rather than providing the necessary evidence for his claim, whereas the “insult” was actually delivered in an entirely different book, and whereas of course no evidence is necessary because Zinn is merely relating an account of events derived from an official inquiry.
E VEN ASIDE FROM this instance of outright libel, which has at least the pragmatic justification of being not easily detectable by the sort of toy fascist, National Review-subscribing scum who would presumably make up the central audience for an authorized biography of Henry Kissinger as written by a Hoover Institution scholar and who would be unlikely to have copies of Howard Zinn books lying around with which to check up on Ferguson’s claims, this whole haphazard bid to portray Kissinger as being subject to outsized criticism relative to his actual conduct is also remarkable for how it occasionally collapses even without any need for research or in fact any particular knowledge whatsoever beyond the understanding that if X applies to A, B, and C, then X is not particular to B, and does not tell us anything about B by which we might differentiate it from A and C.
Ferguson himself notes, for instance, that David Icke’s surreal allegations encompass pretty much everyone of socio-political prominence; Icke’s “List of Famous Satanists,” Ferguson writes, “includes not only Kissinger but also the Astors, Bushes, Clintons, DuPonts, Habsburgs, Kennedys, Rockefellers, Rothschilds, and the entire British royal family — not to mention Tony Blair, Winston Churchill, Adolf Hitler, Mikhail Gorbachev, and Joseph Stalin. (The comedian Bob Hope also makes the list).” So why is it remarkable that Kissinger should be included? And what’s the point of bringing up Lyndon LaRouche’s allegation that Kissinger works for the British? Search LaRouche’s name on YouTube and you’ll find, among other things, a 1980s TV promo in which he denounces Walter Mondale as “not just a KGB agent in the ordinary sense” but also “wholly owned by the left wing of the Socialist International and the grain cartel interests.” If you’re wondering why I happen to have that memorized, the answer is that this was one of several amusing political clips I was in the habit of watching once a week or so prior to my arrest; the funny part is that if Mondale were indeed under someone’s control, the “grain cartel interests” is exactly the sort of lame-ass shit that he’d be fronting. Anyway, it’s none of your business.
Having finished doing whatever it is that he thinks he’s just done, Ferguson at last makes an effort to engage Kissinger’s critics on the complex issue of whether or not Kissinger bears any responsibility for his actions. He now lurches into an overview of Christopher Hitchens’s 2001 book The Trial of Henry Kissinger, in which Hitchens “went so far as to accuse Kissinger of ‘war crimes and crimes against humanity in Indochina, Chile, Argentina, Cyprus, East Timor, and several other places’ (in fact, the only other place discussed in his book is Bangladesh).” Apparently Hitchens didn’t think to just throw Hunter S. Thompson in there to round out his list, but then the old heretic apparently had worse problems than his well-known lack of imagination: “Hitchens was a gifted polemicist; his abilities as a historian are more open to question.” It’s the reverse with Ferguson, who’s undoubtedly an accomplished sorter-through of archives but who cannot seem to make even an exceedingly dishonest argument come out in his own favor.
But Ferguson isn’t done making dishonest arguments, and I’m not done making fun of them; we’ve really only covered three or four pages so far, after all. Next time we’ll take a look at how Ferguson handles Hitchens and certain other Kissinger critics. (SPOILER: He does it dishonestly.)
Harvard!
Awkward Questions of the Day, for the Hoover Institution and/or Harvard University to Ask Niall Ferguson About His Various False and Misleading Statements Taken From a Single Paragraph of His Introduction to His Mediocre Kissinger Biography: |
//
// ItemManagement.h
// TOPIOSSdk
// 商品模块插件必须实现的业务协议，作为应用跳转或者消息跳转的实现支持
//
// Created by fangweng on 12-11-20.
//
//
#import <Foundation/Foundation.h>
//以下接口uid是系统必选传递参数
@protocol JDY_ItemManagement <NSObject>
//跳转到商品详情功能模块，iid是商品id（必选）chatNick（可选）如果从旺旺聊天窗口跳转，则会带上聊天对象的nick
//authString在第一次从无线卖家平台跳转的时候带有的授权信息（如果用户和应用授权有效的情况下会带有）
-(void) itemDetail:(NSString *)uid iid:(NSString *)iid chatNick:(NSString *)chatNick authString:(NSString *)authString params:(NSDictionary *)params;
//跳转到商品列表，ItemStatus参看JDY_ProtocolConstants中的定义（可选）
//authString在第一次从无线卖家平台跳转的时候带有的授权信息（如果用户和应用授权有效的情况下会带有）
//chatNick（可选）如果从旺旺聊天窗口跳转，则会带上聊天对象的nick
-(void) itemList:(NSString *)uid itemStatus:(NSString *)itemStatus chatNick:(NSString *)chatNick authString:(NSString *)authString params:(NSDictionary *)params;
@end
|
import {DispatcherEvent} from "./DispatcherEvent";
export class Debug {
public skeleton: boolean = false;
public collision: boolean = false;
public stats: boolean = false;
}
|
Daniel Sturridge and Simon Mignolet are the stars for Liverpool as the Reds open season with victory over Stoke.
New Liverpool goalkeeper Simon Mignolet saved a late penalty to earn his side a deserved 1-0 win over Stoke City in a frenetic opening match of the Premier League season at Anfield on Saturday.
Striker Daniel Sturridge put Liverpool ahead in the first half with a firm shot from 20 metres and Mignolet plunged to his right to keep out Jonathan Walters's spot-kick a minute from time.
Brazilian playmaker Philippe Coutinho pulled the strings for Liverpool and Stoke goalkeeper Asmir Begovic made several fine saves before Sturridge struck after 37 minutes, drilling a low left-foot shot into the corner of the net.
Midfielder Jordan Henderson hit the post in the second half as Liverpool poured forward in search of another goal but Begovic continued to defy the hosts who are bidding to win the league title for the first time since 1990.
Stoke threatened occasionally on the break and they were awarded a penalty when Liverpool defender Daniel Agger needlessly handled a long Charlie Adam free kick.
But Mignolet, who had endured a nervous debut, palmed away a weak effort by Walters to spark wild celebrations around the ground.
Five matches start at 1400 GMT - Arsenal host Aston Villa, Norwich City meet Everton, Sunderland face Fulham, West Bromwich Albion take on Southampton and promoted Cardiff City travel to West Ham United.
Champions Manchester United begin the defence of their title at Swansea City in the late kickoff. |
Chicago (AFP) – Chicagoans are not too happy with Donald Trump’s portrayal of their home town.
Elected leaders in the third largest US city are trying to show their displeasure by taking down an honorary “Trump Plaza” street sign in front of the businessman’s Chicago skyscraper.
In the first presidential debate, on September 26, the Republican nominee brought up the thousands of shootings recorded in Chicago and asked, “Where is this? Is this a war-torn country? What are we doing?”
There have in fact been 3,324 shooting victims in the city so far this year, according to data compiled by the Chicago Tribune, and 568 murders.
The city, President Barack Obama’s home town, has the worst violent crime statistics of any major American metropolis.
But it is also home to vibrant tourism, shimmering skyscrapers — including Trump’s own glass and steel tower housing a hotel and condominiums — and one of the most striking waterfronts in the country.
Local officials have bristled at comparisons to war zones.
“His decision to drag us into this campaign and… paint a very distorted caricature of Chicago is a mistake,” said Brendan Reilly, a city council member and the architect of the sign removal effort, which now has the support of Mayor Rahm Emanuel, a prominent Democrat and former chief of staff to Obama.
“We’ll put the sign back up when he releases his tax returns,” Emanuel quipped at a news conference this week, ribbing at Trump’s refusal to disclose those documents.
Trump’s campaign shot back that Reilly “should spend his time focused on economic growth, educational choice, and safer streets for Chicago children,” Chicago-based Trump surrogate Steve Cortes said, according to the Sun-Times newspaper.
It would take at least a month for the city council to vote on the idea. The presidential election may well be over by the time the sign’s fate is decided.
Early this year Trump offended another world-class city, saying that parts of Paris seemed to be “outside the law.” Mayor Anne Hidalgo later branded the candidate as “so stupid.” |
Modeling aerial refueling operations OF THE THESIS Modeling Aerial Refueling Operations by Allen B. McCoy III Doctor of Science in Systems Science and Mathematics Washington University in St. Louis, 2010 Research Advisor: Professor Ervin Y. Rodin Aerial Refueling (AR) is the act of offloading fuel from one aircraft (the tanker) to another aircraft (the receiver) in mid flight. Meetings between tanker and receiver aircraft are referred to as AR events and are scheduled to: escort one or more receivers across a large body of water; refuel one or more receivers; or train receiver pilots, tanker pilots, and boom operators. In order to efficiently execute the Aerial Refueling Mission, the Air Mobility Command (AMC) of the United States Air Force (USAF) depends on computer models to help it make tanker basing decisions, plan tanker sorties, schedule aircraft, develop new organizational doctrines, and influence policy. We have worked on three projects that have helped AMC improve its modeling and decision making capabilities. Optimal Flight Planning: Currently Air Mobility simulation and optimization software packages depend on algorithms which iterate over three dimensional fuel flow tables to compute aircraft fuel consumption under changing flight conditions. When a high degree of fidelity is required, these algorithms use a large amount of |
Markedness and Second Language Acquisition In this paper, various definitions of markedness are discussed, including the difference in the assumptions underlying psychological and linguistic approaches to markedness. It is proposed that if one adopts a definition derived from theories of language learnability, then the second language learner's prior linguistic experience may predispose him or her towards transferring marked structures from the first language to the second, contrary to usual assumptions in the literature that suggest that second language learners will avoid marked forms. To test this hypothesis, adult and child learners of French as a second language were tested using grammaticality judgment tasks on two marked structures, preposition stranding and the double object construction, which are grammatical in English but ungrammatical in French, to see if they would accept French versions of these structures. It was found that the second language learners did not accept preposition stranding in French but did accept the double object construction, suggesting that transfer takes place only with one of the two marked structures. In addition, the children took tests on these structures in their native language to see if they perceived them as in any sense psycholinguistically marked. Results show that they do not treat marked and unmarked structures differently in the native language. It is suggested that the concept of markedness may cover a range of phenomena that need to be further clarified and investigated. |
<gh_stars>0
#include <stdio.h>
int main()
{
int a=1;
while(a<1){ //primero compara y luego corre
printf( "Iteracion numero %d\n",a );
a++;
}printf( "Fin de programa, valor actual de %d",a );
return 0;
} |
def ontologyIncludes(self):
imports = self.graph.query("""
SELECT distinct ?import_file
WHERE {?s owl:imports ?import_file.}
ORDER BY (?import_file)
""")
print("It has %s import files ..." % len(imports))
for result_row in imports:
file = result_row.import_file.rsplit('/',1)[1]
try:
if os.path.isfile( "../imports/" + file):
self.graph.parse("../imports/" + file)
else:
print ('WARNING:' + "../imports/" + file + " could not be loaded! Does its ontology include purl have a corresponding local file? \n")
except rdflib.exceptions.ParserError as e:
print (file + " needs to be in RDF OWL format!") |
Relationship between the 5'UTR of vascular endothelial growth factor polymorphism and retinopathy of prematurity in Chinese premature newborns. OBJECTIVE To investigate the relationship between vascular endothelial growth factor (VEGF) gene polymorphisms and Retinopathy of prematurity (ROP) development in preterm infants of China Han ethnic population. METHODS VEGF gene promoter polymorphisms (-165C/T and -141A/C) were studied in 54 neonates with ROP but not requiring treatment (regressive ROP group), 48 neonates with ROP that requires cryotherapy/photocoagulation (severe ROP group), and in a control group of 62 preterm infants without ROP. Genotyping for VEGF gene promoter was performed by polymerase chain reaction (PCR) and gene sequencing. RESULTS In this study, -165C/T genotype was found to be associated with severe ROP (P=0.012<0.05), but not with regressive ROP. For the dominant genetic model, subjects carrying the -165T allele had a decreased risk of ROP compared to those non-carrying the -165C allele in both regressive ROP regressive group and severe ROP group, and especially in severe ROP group (in regressive ROP group: OR=2.069, 95% CI=0.986-4.340; in severe ROP group: OR=3.677, 95% CI=1.339-10.099). For the allelic model, comparing the A allele to the C allele, the -165C/T site also showed a decreased risk of ROP in both regressive ROP group and severe ROP group (in regressive ROP group: OR=2.395, 95% CI=1.112-5.157; in severe ROP group: OR=4.258, 95% CI=1.518-11.945). In -141A/C genotypes and alleles were found not to be associated with severe ROP or regressive ROP. Linkage disequilibrium analysis revealed -165C/T and -141A/C were a strong linkage disequilibrium, and haplotype analysis revealed that T-165C-141 haplotype was associated with resistance to severe ROP but C-165A-141 haplotype was associated with risk to severe ROP. No statistically significant differences were observed in regressive ROP group. CONCLUSION VEGF -165C/T SNP is associated with ROP especial in severe ROP. VEGF -141A/C SNP is not associated with ROP. T-165C-141 and C-165A-141 haplotypes may have a role in severe ROP. |
After law enforcement shot and killed a half-asleep woman who was allegedly suicidal and then attempted to bribe her family, the victim’s children are speaking out and suing their local police department.
On March 25, 2015, law enforcement officials from Gardner, Kansas, responded to a call from Andrew Musto about 53-year-old Deanne Choate, who reportedly had been drinking, was in possession of a gun, and was suicidal, as reported by Courthouse News.
When officers Robert Huff, Justin Mohoney, and Jeff Breneman arrived at the home, they arrested and removed Musto, Deanne’s boyfriend from the house. They then headed upstairs to Deanne's bedroom, where the 53-year-old was found sleeping naked. After asking her to move quickly and locate the gun, she grabbed a hooded sweatshirt and found the weapon. The officers shot her dead soon after.
“They say she was pointing a pistol at them,” Michael Weddington, Deanne’s son, told The Daily Beast.
"Deanne was not threatening in any way as she complied with officers' instructions in providing a handgun located between the mattress and headboard of the bed," said Michelle Choate, Deanne’s daughter, who is the primary filer of the wrongful death lawsuit. "The gun would have easily been located by officers if they had searched and 'cleared' the room as reported."
Neighbors also witnessed the incident and the terrible aftermath.
Deanne’s family insists that the situation could have ended peacefully, as she had never acted belligerently.
“They could have called me and said, ‘Your mom is apparently suicidal and we can’t get in contact with her; she’s not coming out of her room. I was no more than a 10-minute drive away,” her son said.
With the lawsuit, the family wants to bring attention and reform to the police department so a similar situation doesn’t happen to another family. |
class ValidateOrganization:
"""validates the organizations added to the database
Returns:
[boolean] -- [returns true for valid fields and false for invalid
fields]
"""
def __init__(self, data):
self.data = data
def validate_organization_name(self):
"""validates the organization name
Returns:
[True] -- [returns true for valid organization name else false]
"""
try:
if self.data['organization_name'] == "" or \
not isinstance(self.data['organization_name'], str):
return False
else:
return True
except KeyError:
return False
|
ASH Position Paper: Treatment of Hypertension in Patients With DiabetesAn Update This report updates concepts on hypertension management in patients with diabetes. It focuses on clinical outcomes literature published within the last 3years and incorporates these observations into modifications of established guidelines. While the fundamentals of treatment and goal blood pressures remain unchanged, approaches to specific patientrelated issues has changed. This update focuses on questions such as what to do when a patient has an elevated potassium level when therapy is initiated and whether combinations of agents that block the reninangiotensin system still be used. In addition, there are updates from trials, just published and in press, that focus on related management issues influencing cardiovascular outcomes in persons with diabetes. Last, an updated algorithm is provided that incorporates many of the new findings and is suggested as a starting point to achieve blood pressure goals. |
If you’ve visited Disneyland, you may have seen a small plane fly overhead at one point. The OC is full of rich-ass people, might be a Newport Beach golfer, no big deal, right? Except, as it turns out, the Anaheim police department had access to military-grade dragnet phone spying equipment, the kind that can suck up your phone’s information from an airplane along with thousands of others.
Anaheim isn’t a big city. With a population of 336,265, it is substantially smaller than Wichita, Kansas and Mesa, Arizona. This is how entrenched America’s berserk surveillance culture is: A suburban city’s police department had access to spying tools called dirtboxes, the same spying tools sought by the Air Force. Insane militarization isn’t just for big-city cops.
Advertisement
Anaheim Police requested a Homeland Security grant to fund its dirtbox purchase, it claimed that “every city in Orange County has benefited from the DRT device.” Unless the police lied on this request, this means that Anaheim’s secret spy tool had a wide-enough range to reach people in other cities. It also means that the millions of tourists passing through Disneyland would’ve been within reach.
The FBI is so ferociously tight-lipped about its use of surveillance equipment that it has advised police to dismiss criminal cases rather than expose information about Stingrays, which are spying devices that mimic cell phone towers, allowing owners to intercept information from cell phones. “Dirtbox” devices are similar to Stingrays, except they can go in airplanes, which makes them even more potent since it increases the number of phones they can spy on. And as Ars Technica points out, some dirtboxes can break encryption on hundreds of cell phones at once. They are extremely powerful dragnet spying devices.
Advertisement
The FBI’s attitude towards Stingrays and dirtboxes has been mimicked by law enforcement. I say Anaheim was “caught” with these tools because, like the FBI, it did everything it could to keep them secret. The American Civil Liberties Union released the documents yesterday, but only after fighting a court battle to obtain them.
California passed a law requiring warrants for police use of dirtboxes, Stingrays, and other electronic surveillance tools—but it only went into effect January 1, 2016. It’s not clear if Anaheim ever sought warrants previously. Considering it labeled the dirtboxes a “Covert Purchase,” and the level of secrecy shown in emails about the devices, it is clear that the police told as few people as possible.
Advertisement
Screenshot from ACLU documents
We already knew that Chicago and Los Angeles police used dirtboxes for years, along with the FBI, but now it’s evident that use of these devices is even more widespread than previously thought. The good news is that laws are finally catching up with the police. The terrifying news is that we need new laws to protect our privacy from the people supposedly protecting us.
[Ars Technica via ACLU]
Image: AP |
WASHINGTON (Reuters) - NBC, Fox News and Facebook pulled an ad by President Donald Trump’s campaign that critics had labeled racist as a bitter election fight for control of the U.S. Congress headed on Monday for an unpredictable finish.
Tuesday’s elections, widely seen as a referendum on Trump, have been portrayed by both Republicans and Democrats as critical for the direction of the country. At stake is control of both chambers of Congress, and with it the ability to block or promote Trump’s agenda, as well as 36 governor’s offices.
A surge in early voting, fueled by a focus on Trump’s pugilistic, norms-breaking presidency by supporters of both parties, could signal the highest turnout in 50 years for a midterm U.S. election, when the White House is not on the line.
The 30-second ad, which was sponsored by Trump’s 2020 re-election campaign and which debuted online last week, featured courtroom video of an illegal immigrant from Mexico convicted in the 2014 killings of two police officers, juxtaposed with scenes of migrants headed through Mexico.
Critics, including members of Trump’s own party, had condemned the spot as racially divisive.
Fox News Channel, which Trump has repeatedly named his favorite broadcaster, also said it would no longer run the spot. Fox News, a unit of Twenty-First Century Fox Inc, said it had made the decision after a review but did not elaborate.
Facebook Inc said it would no longer allow paid promotions of the ad, although it would allow users to share the ad on their own pages.
Trump batted away reporters’ questions about the networks’ decision to drop the ad.
“You’re telling me something I don’t know about. We have a lot of ads, and they certainly are effective based on the numbers we’re seeing,” Trump said as he departed Joint Base Andrews in Maryland for a rally in Cleveland.
After Ohio, Trump headed to campaign against vulnerable Democratic U.S. senators in Indiana and Missouri at the end of a six-day pre-election sweep that has featured heated rhetoric about immigration and repeated warnings about a caravan of Central American migrants moving through Mexico toward the U.S. border.
Opinion polls and election forecasters favor Democrats to pick up the minimum of 23 seats they need on Tuesday to capture a majority in the U.S. House of Representatives, which would enable them to stymie Trump’s legislative agenda and investigate his administration.
But 64 of the 435 House races remain competitive, according to a Reuters analysis of the three main nonpartisan U.S. forecasters, and control of the Senate is likely to come down to a half dozen close contests in Arizona, Nevada, Missouri, North Dakota, Indiana and Florida.
Democrats also are threatening to recapture governor’s offices in several battleground states such as Michigan, Wisconsin, Ohio and Pennsylvania, a potential help for the party in those states in the 2020 presidential race.
Democratic former President Barack Obama delivered doughnuts to campaign volunteers in a House district in suburban Virginia, where Democrat Jennifer Wexton, a state senator, is challenging Republican incumbent Barbara Comstock in a fiercely contested race.
Obama said the country’s character and its commitment to decency and equality were on the ballot on Tuesday.
About 40 million early votes – including absentee, vote-by-mail and in-person ballots – will likely be cast by Election Day, according to Michael McDonald, a professor at the University of Florida who tracks the figures. In the last such congressional elections in 2014, there were 27.5 million early votes.
McDonald estimated that 45 percent of registered voters would cast ballots, which would be the highest for a midterm election in 50 years.
“The atypical thing that we’re seeing is high early vote activity in states without competitive elections or no statewide elections,” McDonald said in a phone interview. |
Impact of Fabric Moisture Transport Properties on Physiological Responses when Wearing Protective Clothing This purpose of this study was to investigate the impact of fabric moisture transport properties (MTP) on physiological responses when wearing protective clothing. Ten healthy subjects wore two kinds of personal protective equipment (PPE) ensembles and exercised on a treadmill, worked on a computer, and moved a mannequin in an environment that simulated where health carers work. PPE1 consisted of cotton underwear and 100% polyethylene outerwear. PPE2 consisted of cotton underwear with moisture management function and outerwear made of waterproof breathable fabric. The results showed that there were significantly higher cumulative one-way transport capacity, liquid moisture management capacity, and wetting time in PPE2 than in PPE1 underwear. There was significantly higher water vapor permeability (WVP) in PPE2 than in PPE1 outerwear. Deep ear canal temperature, mean skin temperature, and chest wall skin and clothing microclimates (temperature and humidity) were significantly lower with PPE2 than PPE1. The level of plasma oxygen saturation was significantly higher with PPE 2 than PPE1. In the present study, due to the MTP of the fabrics, liquid sweat transferred from the skin surface to the opposite surface quickly and speeded up the processes of evaporation and heat dissipation. It was concluded that the fabric's MTP, when incorporated into protective clothing, is the main physiological mechanism for reduced heat stress. |
/**
* Wrapper to manage Hadoop flows for ThirdEye.
* <h1>Config</h1>
* <table>
* <tr>
* <th>Property</th>
* <th>Description</th>
* </tr>
* <tr>
* <td>thirdeye.flow</td>
* <td>One of {@link com.linkedin.thirdeye.bootstrap.ThirdEyeJob.FlowSpec}</td>
* </tr>
* <tr>
* <td>thirdeye.flow.schedule</td>
* <td>A string describing the flow schedule (used to tag segments)</td>
* </tr>
* <tr>
* <td>thirdeye.phase</td>
* <td>One of {@link com.linkedin.thirdeye.bootstrap.ThirdEyeJob.PhaseSpec}</td>
* </tr>
* <tr>
* <td>thirdeye.root</td>
* <td>Root directory on HDFS, under which all collection data is stored</td>
* </tr>
* <tr>
* <td>thirdeye.collection</td>
* <td>Collection name (data stored at ${thirdeye.root}/${thirdeye.collection}</td>
* </tr>
* <tr>
* <td>thirdeye.server.uri</td>
* <td>URI prefix for thirdeye server (e.g. http://some-machine:10283)</td>
* </tr>
* <tr>
* <td>thirdeye.time.path</td>
* <td>A path to a properties file on HDFS containing thirdeye.time.min, thirdeye.time.max</td>
* </tr>
* <tr>
* <td>thirdeye.time.min</td>
* <td>Manually override thirdeye.time.min from thirdeye.time.path</td>
* </tr>
* <tr>
* <td>thirdeye.time.max</td>
* <td>Manually override thirdeye.time.max from thirdeye.time.path</td>
* </tr>
* <tr>
* <td>thirdeye.dimension.index.ref</td>
* <td>Manually override thirdeye.dimension.index.dir to use, by default it will use the latest
* directory under thirdeye.root/DIMENSION_INDEX</td>
* </tr>
* *
* <tr>
* <td>thirdeye.cleanup.skip</td>
* <td>Disable cleanup of old folders if this is set to true</td>
* </tr>
* *
* <tr>
* <td>thirdeye.cleanup.daysago</td>
* <td>Number of days ago to look while cleaning up old folders generated by hadoop workflow</td>
* </tr>
* *
* <tr>
* <td>thirdeye.skip.missing</td>
* <td>Filter all input paths in input.paths which do not exist if set to true</td>
* </tr>
* *
* <tr>
* <td>thirdeye.folder.permission</td>
* <td>The permissions to be set on thirdeye output folders</td>
* </tr>
* </table>
*/
public class ThirdEyeJob {
private static final Logger LOGGER = LoggerFactory.getLogger(ThirdEyeJob.class);
private static final String ENCODING = "UTF-8";
private static final String USAGE = "usage: phase_name job.properties";
private static final String AVRO_SCHEMA = "schema.avsc";
private static final String PINOT_JSON_SCHEMA = "schema.json";
private static final String TREE_FILE_FORMAT = ".bin";
private static final String DEFAULT_CLEANUP_DAYS_AGO = "7";
private static final String DEFAULT_CLEANUP_SKIP = "false";
private static final String DEFAULT_SKIP_MISSING = "true";
private static final String DEFAULT_FOLDER_PERMISSION = "755";
private static final String DEFAULT_THIRDEYE_COMPACTION = "false";
private static final String DEFAULT_THIRDEYE_PRESERVE_TIME_COMPACTION = "false";
private static final String FOLDER_PERMISSION_REGEX = "0?[1-7]{3}";
private static final String DEFAULT_POLL_ENABLE = "false";
private static final String DEFAULT_TOPK = "false";
private static final String DEFAULT_NUM_PARTITIONS = "1";
private static final String DEFAULT_POLL_FREQUENCY = "60000";
private static final String DEFAULT_POLL_TIMEOUT = "3600000";
private static final String INPUT_PATHS_JOINER = ",";
private static final String DATA_FOLDER_JOINER = "_";
private static final String TEMPORARY_FOLDER = "_temporary";
private static final String DEFAULT_CONVERTER_CLASS = ThirdEyeAvroUtils.class.getName();
protected enum FlowSpec {
DIMENSION_INDEX,
METRIC_INDEX
}
private enum FlowSchedule {
HOURLY,
DAILY,
MONTHLY
}
private enum PhaseSpec {
JOIN {
@Override
Class<?> getKlazz() {
return JoinPhaseJob.class;
}
@Override
String getDescription() {
return "Joins multiple data sets based on join key";
}
@Override
Properties getJobProperties(Properties inputConfig, String root, String collection,
FlowSpec flowSpec, DateTime minTime, DateTime maxTime, String inputPaths) {
return inputConfig;
}
},
TRANSFORM {
@Override
Class<?> getKlazz() {
return TransformPhaseJob.class;
}
@Override
String getDescription() {
return "Transforms avro record";
}
@Override
Properties getJobProperties(Properties inputConfig, String root, String collection,
FlowSpec flowSpec, DateTime minTime, DateTime maxTime, String inputPaths) {
return inputConfig;
}
},
WAIT {
@Override
Class<?> getKlazz() {
return null;
}
@Override
String getDescription() {
return "Polls a pre-determined amount of time for the existence of input paths";
}
@Override
Properties getJobProperties(Properties inputConfig, String root, String collection,
FlowSpec flowSpec, DateTime minTime, DateTime maxTime, String inputPaths)
throws Exception {
return null;
}
},
ANALYSIS {
@Override
Class<?> getKlazz() {
return AnalysisPhaseJob.class;
}
@Override
String getDescription() {
return "Analyzes input Avro data to compute information necessary for job";
}
@Override
Properties getJobProperties(Properties inputConfig, String root, String collection,
FlowSpec flowSpec, DateTime minTime, DateTime maxTime, String inputPaths)
throws Exception {
Properties config = new Properties();
config.setProperty(AnalysisJobConstants.ANALYSIS_INPUT_AVRO_SCHEMA.toString(),
getSchemaPath(root, collection));
config.setProperty(AnalysisJobConstants.ANALYSIS_CONFIG_PATH.toString(),
getConfigPath(root, collection));
config.setProperty(AnalysisJobConstants.ANALYSIS_INPUT_PATH.toString(), inputPaths);
config.setProperty(AnalysisJobConstants.ANALYSIS_OUTPUT_PATH.toString(),
getMetricIndexDir(root, collection, flowSpec, minTime, maxTime) + File.separator
+ ANALYSIS.getName());
return config;
}
},
AGGREGATION {
@Override
Class<?> getKlazz() {
return AggregatePhaseJob.class;
}
@Override
String getDescription() {
return "Aggregates input data";
}
@Override
Properties getJobProperties(Properties inputConfig, String root, String collection,
FlowSpec flowSpec, DateTime minTime, DateTime maxTime, String inputPaths)
throws Exception {
Properties config = new Properties();
config.setProperty(AggregationJobConstants.AGG_INPUT_AVRO_SCHEMA.toString(),
getSchemaPath(root, collection));
config.setProperty(AggregationJobConstants.AGG_CONFIG_PATH.toString(),
getConfigPath(root, collection));
boolean skipMissing = Boolean.parseBoolean(inputConfig.getProperty(
ThirdEyeJobConstants.THIRDEYE_SKIP_MISSING.getName(), DEFAULT_SKIP_MISSING));
if (skipMissing) {
inputPaths = getFilteredInputPaths(inputPaths);
}
config.setProperty(AggregationJobConstants.AGG_INPUT_PATH.toString(), inputPaths);
config.setProperty(AggregationJobConstants.AGG_OUTPUT_PATH.toString(),
getMetricIndexDir(root, collection, flowSpec, minTime, maxTime) + File.separator
+ AGGREGATION.getName());
config.setProperty(AggregationJobConstants.AGG_DIMENSION_STATS_PATH.toString(),
getMetricIndexDir(root, collection, flowSpec, minTime, maxTime) + File.separator
+ StarTreeConstants.DIMENSION_STATS_FOLDER);
config.setProperty(AggregationJobConstants.AGG_PRESERVE_TIME_COMPACTION.toString(),
inputConfig.getProperty(
ThirdEyeJobConstants.THIRDEYE_PRESERVE_TIME_COMPACTION.getName(),
DEFAULT_THIRDEYE_PRESERVE_TIME_COMPACTION));
config.setProperty(AggregationJobConstants.AGG_METRIC_SUMS_PATH.toString(),
getMetricIndexDir(root, collection, flowSpec, minTime, maxTime) + File.separator
+ StarTreeConstants.METRIC_SUMS_FOLDER + File.separator
+ StarTreeConstants.METRIC_SUMS_FILE);
String converterClass = inputConfig.getProperty(ThirdEyeJobConstants.THIRDEYE_INPUT_CONVERTER_CLASS.getName(), DEFAULT_CONVERTER_CLASS);
if (converterClass.isEmpty()) {
converterClass = DEFAULT_CONVERTER_CLASS;
}
config.setProperty(AggregationJobConstants.AGG_CONVERTER_CLASS.toString(), converterClass);
return config;
}
},
SEGMENT_CREATION {
@Override
Class<?> getKlazz() {
return SegmentCreationPhaseJob.class;
}
@Override
String getDescription() {
return "Generates pinot segments";
}
@Override
Properties getJobProperties(Properties inputConfig, String root, String collection,
FlowSpec flowSpec, DateTime minTime, DateTime maxTime, String inputPaths)
throws Exception {
Properties config = new Properties();
config.setProperty(SegmentCreationPhaseConstants.SEGMENT_CREATION_SCHEMA_PATH.toString(),
getPinotSchemaPath(root, collection));
config.setProperty(SegmentCreationPhaseConstants.SEGMENT_CREATION_CONFIG_PATH.toString(),
getConfigPath(root, collection));
String inputPath = getMetricIndexDir(root, collection, flowSpec, minTime, maxTime) + File.separator;
FileSystem fs = FileSystem.get(new Configuration());
if (fs.exists(new Path(inputPath + TOPK_ROLLUP_PHASE3.getName()))) {
inputPath = inputPath + TOPK_ROLLUP_PHASE3.getName();
} else if (fs.exists(new Path(inputPath + ROLLUP_PHASE4.getName()))) {
inputPath = inputPath + ROLLUP_PHASE4.getName();
} else {
inputPath = inputPath + AGGREGATION.getName();
}
config.setProperty(SegmentCreationPhaseConstants.SEGMENT_CREATION_INPUT_PATH.toString(), inputPath);
config.setProperty(SegmentCreationPhaseConstants.SEGMENT_CREATION_OUTPUT_PATH.toString(),
getMetricIndexDir(root, collection, flowSpec, minTime, maxTime) + File.separator + SEGMENT_CREATION.getName());
config.setProperty(SegmentCreationPhaseConstants.SEGMENT_CREATION_SEGMENT_TABLE_NAME.toString(),
collection);
config.setProperty(SegmentCreationPhaseConstants.SEGMENT_CREATION_WALLCLOCK_START_TIME.toString(),
String.valueOf(minTime.getMillis()));
config.setProperty(SegmentCreationPhaseConstants.SEGMENT_CREATION_WALLCLOCK_END_TIME.toString(),
String.valueOf(maxTime.getMillis()));
String schedule = inputConfig.getProperty(ThirdEyeJobConstants.THIRDEYE_FLOW_SCHEDULE.getName());
config.setProperty(SegmentCreationPhaseConstants.SEGMENT_CREATION_SCHEDULE.toString(), schedule);
return config;
}
},
SEGMENT_PUSH {
@Override
Class<?> getKlazz() {
return SegmentPushPhase.class;
}
@Override
String getDescription() {
return "Pushes pinot segments to pinot controller";
}
@Override
Properties getJobProperties(Properties inputConfig, String root, String collection,
FlowSpec flowSpec, DateTime minTime, DateTime maxTime, String inputPaths)
throws Exception {
Properties config = new Properties();
config.setProperty(SegmentPushPhaseConstants.SEGMENT_PUSH_INPUT_PATH.toString(),
getMetricIndexDir(root, collection, flowSpec, minTime, maxTime) + File.separator + SEGMENT_CREATION.getName());
config.setProperty(SegmentPushPhaseConstants.SEGMENT_PUSH_CONTROLLER_HOSTS.toString(),
inputConfig.getProperty(ThirdEyeJobConstants.THIRDEYE_PINOT_CONTROLLER_HOSTS.getName()));
config.setProperty(SegmentPushPhaseConstants.SEGMENT_PUSH_CONTROLLER_PORT.toString(),
inputConfig.getProperty(ThirdEyeJobConstants.THIRDEYE_PINOT_CONTROLLER_PORT.getName()));
config.setProperty(SegmentPushPhaseConstants.SEGMENT_PUSH_TABLENAME.toString(),
inputConfig.getProperty(ThirdEyeJobConstants.THIRDEYE_COLLECTION.getName()));
return config;
}
},
TOPK_ROLLUP_PHASE1 {
@Override
Class<?> getKlazz() {
return TopKRollupPhaseOneJob.class;
}
@Override
String getDescription() {
return "Topk rollup phase1";
}
@Override
Properties getJobProperties(Properties inputConfig, String root, String collection,
FlowSpec flowSpec, DateTime minTime, DateTime maxTime, String inputPaths)
throws Exception {
Properties config = new Properties();
config.setProperty(TopKRollupPhaseOneConstants.TOPK_ROLLUP_PHASE1_CONFIG_PATH.toString(),
getConfigPath(root, collection));
config.setProperty(TopKRollupPhaseOneConstants.TOPK_ROLLUP_PHASE1_INPUT_PATH.toString(),
getMetricIndexDir(root, collection, flowSpec, minTime, maxTime) + File.separator
+ AGGREGATION.getName());
config.setProperty(TopKRollupPhaseOneConstants.TOPK_ROLLUP_PHASE1_OUTPUT_PATH.toString(),
getMetricIndexDir(root, collection, flowSpec, minTime, maxTime) + File.separator
+ TOPK_ROLLUP_PHASE1.getName());
config.setProperty(
TopKRollupPhaseOneConstants.TOPK_ROLLUP_PHASE1_METRIC_SUMS_PATH.toString(),
getMetricIndexDir(root, collection, flowSpec, minTime, maxTime) + File.separator
+ StarTreeConstants.METRIC_SUMS_FOLDER + File.separator
+ StarTreeConstants.METRIC_SUMS_FILE);
return config;
}
},
TOPK_ROLLUP_PHASE2 {
@Override
Class<?> getKlazz() {
return TopKRollupPhaseTwoJob.class;
}
@Override
String getDescription() {
return "Topk rollup phase2";
}
@Override
Properties getJobProperties(Properties inputConfig, String root, String collection,
FlowSpec flowSpec, DateTime minTime, DateTime maxTime, String inputPaths)
throws Exception {
Properties config = new Properties();
config.setProperty(TopKRollupPhaseTwoConstants.TOPK_ROLLUP_PHASE2_CONFIG_PATH.toString(),
getConfigPath(root, collection));
config.setProperty(TopKRollupPhaseTwoConstants.TOPK_ROLLUP_PHASE2_INPUT_PATH.toString(),
getMetricIndexDir(root, collection, flowSpec, minTime, maxTime) + File.separator
+ TOPK_ROLLUP_PHASE1.getName());
config.setProperty(TopKRollupPhaseTwoConstants.TOPK_ROLLUP_PHASE2_OUTPUT_PATH.toString(),
getMetricIndexDir(root, collection, flowSpec, minTime, maxTime) + File.separator
+ TOPK_ROLLUP_PHASE2.getName());
return config;
}
},
TOPK_ROLLUP_PHASE3 {
@Override
Class<?> getKlazz() {
return TopKRollupPhaseThreeJob.class;
}
@Override
String getDescription() {
return "Topk rollup phase3";
}
@Override
Properties getJobProperties(Properties inputConfig, String root, String collection,
FlowSpec flowSpec, DateTime minTime, DateTime maxTime, String inputPaths)
throws Exception {
Properties config = new Properties();
config.setProperty(TopKRollupPhaseThreeConstants.TOPK_ROLLUP_PHASE3_CONFIG_PATH.toString(),
getConfigPath(root, collection));
config.setProperty(TopKRollupPhaseThreeConstants.TOPK_ROLLUP_PHASE3_INPUT_PATH.toString(),
getMetricIndexDir(root, collection, flowSpec, minTime, maxTime) + File.separator
+ AGGREGATION.getName());
config.setProperty(TopKRollupPhaseThreeConstants.TOPK_ROLLUP_PHASE3_OUTPUT_PATH.toString(),
getMetricIndexDir(root, collection, flowSpec, minTime, maxTime) + File.separator
+ TOPK_ROLLUP_PHASE3.getName());
config.setProperty(TopKRollupPhaseThreeConstants.TOPK_DIMENSIONS_PATH.toString(),
getMetricIndexDir(root, collection, flowSpec, minTime, maxTime) + File.separator
+ TOPK_ROLLUP_PHASE2.getName());
return config;
}
},
ROLLUP_PHASE1 {
@Override
Class<?> getKlazz() {
return RollupPhaseOneJob.class;
}
@Override
String getDescription() {
return "Splits input data into above / below threshold using function";
}
@Override
Properties getJobProperties(Properties inputConfig, String root, String collection,
FlowSpec flowSpec, DateTime minTime, DateTime maxTime, String inputPaths)
throws Exception {
Properties config = new Properties();
config.setProperty(RollupPhaseOneConstants.ROLLUP_PHASE1_CONFIG_PATH.toString(),
getConfigPath(root, collection));
boolean compaction = Boolean.parseBoolean(inputConfig.getProperty(
ThirdEyeJobConstants.THIRDEYE_COMPACTION.getName(), DEFAULT_THIRDEYE_COMPACTION));
if (compaction) {
config.setProperty(RollupPhaseOneConstants.ROLLUP_PHASE1_INPUT_PATH.toString(),
inputPaths);
} else {
config.setProperty(RollupPhaseOneConstants.ROLLUP_PHASE1_INPUT_PATH.toString(),
getMetricIndexDir(root, collection, flowSpec, minTime, maxTime) + File.separator
+ AGGREGATION.getName());
}
config.setProperty(RollupPhaseOneConstants.ROLLUP_PHASE1_OUTPUT_PATH.toString(),
getMetricIndexDir(root, collection, flowSpec, minTime, maxTime) + File.separator
+ ROLLUP_PHASE1.getName());
return config;
}
},
ROLLUP_PHASE2 {
@Override
Class<?> getKlazz() {
return RollupPhaseTwoJob.class;
}
@Override
String getDescription() {
return "Aggregates all possible combinations of raw dimension combination below threshold";
}
@Override
Properties getJobProperties(Properties inputConfig, String root, String collection,
FlowSpec flowSpec, DateTime minTime, DateTime maxTime, String inputPaths)
throws Exception {
Properties config = new Properties();
config.setProperty(RollupPhaseTwoConstants.ROLLUP_PHASE2_CONFIG_PATH.toString(),
getConfigPath(root, collection));
config.setProperty(RollupPhaseTwoConstants.ROLLUP_PHASE2_INPUT_PATH.toString(),
getMetricIndexDir(root, collection, flowSpec, minTime, maxTime) + File.separator
+ ROLLUP_PHASE1.getName() + File.separator + "belowThreshold");
config.setProperty(RollupPhaseTwoConstants.ROLLUP_PHASE2_OUTPUT_PATH.toString(),
getMetricIndexDir(root, collection, flowSpec, minTime, maxTime) + File.separator
+ ROLLUP_PHASE2.getName());
config.setProperty(RollupPhaseTwoConstants.ROLLUP_PHASE2_ANALYSIS_PATH.toString(),
getMetricIndexDir(root, collection, flowSpec, minTime, maxTime) + File.separator
+ StarTreeConstants.DIMENSION_STATS_FOLDER);
return config;
}
},
ROLLUP_PHASE3 {
@Override
Class<?> getKlazz() {
return RollupPhaseThreeJob.class;
}
@Override
String getDescription() {
return "Selects the rolled-up dimension key for each raw dimension combination";
}
@Override
Properties getJobProperties(Properties inputConfig, String root, String collection,
FlowSpec flowSpec, DateTime minTime, DateTime maxTime, String inputPaths)
throws Exception {
Properties config = new Properties();
config.setProperty(RollupPhaseThreeConstants.ROLLUP_PHASE3_CONFIG_PATH.toString(),
getConfigPath(root, collection));
config.setProperty(RollupPhaseThreeConstants.ROLLUP_PHASE3_INPUT_PATH.toString(),
getMetricIndexDir(root, collection, flowSpec, minTime, maxTime) + File.separator
+ ROLLUP_PHASE2.getName());
config.setProperty(RollupPhaseThreeConstants.ROLLUP_PHASE3_OUTPUT_PATH.toString(),
getMetricIndexDir(root, collection, flowSpec, minTime, maxTime) + File.separator
+ ROLLUP_PHASE3.getName());
return config;
}
},
ROLLUP_PHASE4 {
@Override
Class<?> getKlazz() {
return RollupPhaseFourJob.class;
}
@Override
String getDescription() {
return "Sums metric time series by the rolled-up dimension key";
}
@Override
Properties getJobProperties(Properties inputConfig, String root, String collection,
FlowSpec flowSpec, DateTime minTime, DateTime maxTime, String inputPaths)
throws Exception {
Properties config = new Properties();
config.setProperty(RollupPhaseFourConstants.ROLLUP_PHASE4_CONFIG_PATH.toString(),
getConfigPath(root, collection));
config.setProperty(RollupPhaseFourConstants.ROLLUP_PHASE4_INPUT_PATH.toString(),
getMetricIndexDir(root, collection, flowSpec, minTime, maxTime) + File.separator
+ ROLLUP_PHASE3.getName() + ","
+ getMetricIndexDir(root, collection, flowSpec, minTime, maxTime) + File.separator
+ ROLLUP_PHASE1.getName() + File.separator + "aboveThreshold");
config.setProperty(RollupPhaseFourConstants.ROLLUP_PHASE4_OUTPUT_PATH.toString(),
getMetricIndexDir(root, collection, flowSpec, minTime, maxTime) + File.separator
+ ROLLUP_PHASE4.getName());
return config;
}
},
PARTITION {
@Override
Class<?> getKlazz() {
return PartitionPhaseJob.class;
}
@Override
String getDescription() {
return "Partitions rollup output into configured number of partitions";
}
@Override
Properties getJobProperties(Properties inputConfig, String root, String collection,
FlowSpec flowSpec, DateTime minTime, DateTime maxTime, String inputPaths)
throws Exception {
Properties config = new Properties();
config.setProperty(PartitionPhaseConstants.PARTITION_PHASE_CONFIG_PATH.toString(),
getConfigPath(root, collection));
boolean isTopK = Boolean.parseBoolean(
inputConfig.getProperty(ThirdEyeJobConstants.THIRDEYE_TOPK.getName(), DEFAULT_TOPK));
if (isTopK) {
config.setProperty(PartitionPhaseConstants.PARTITION_PHASE_INPUT_PATH.toString(),
getMetricIndexDir(root, collection, flowSpec, minTime, maxTime) + File.separator
+ TOPK_ROLLUP_PHASE3.getName());
} else {
config.setProperty(PartitionPhaseConstants.PARTITION_PHASE_INPUT_PATH.toString(),
getMetricIndexDir(root, collection, flowSpec, minTime, maxTime) + File.separator
+ ROLLUP_PHASE4.getName());
}
config.setProperty(PartitionPhaseConstants.PARTITION_PHASE_OUTPUT_PATH.toString(),
getMetricIndexDir(root, collection, flowSpec, minTime, maxTime) + File.separator
+ PARTITION.getName());
config.setProperty(PartitionPhaseConstants.PARTITION_PHASE_NUM_PARTITIONS.toString(),
inputConfig.getProperty(ThirdEyeJobConstants.THIRDEYE_NUM_PARTITIONS.getName(),
DEFAULT_NUM_PARTITIONS));
return config;
}
},
STARTREE_GENERATION {
@Override
Class<?> getKlazz() {
return StarTreeGenerationJob.class;
}
@Override
String getDescription() {
return "Builds star tree index structure using rolled up dimension combinations and those above threshold";
}
@Override
Properties getJobProperties(Properties inputConfig, String root, String collection,
FlowSpec flowSpec, DateTime minTime, DateTime maxTime, String inputPaths)
throws Exception {
Properties config = new Properties();
config.setProperty(StarTreeGenerationConstants.STAR_TREE_GEN_CONFIG_PATH.toString(),
getConfigPath(root, collection));
String inputPath = getMetricIndexDir(root, collection, flowSpec, minTime, maxTime) + File.separator;
FileSystem fs = FileSystem.get(new Configuration());
if (fs.exists(new Path(inputPath + TOPK_ROLLUP_PHASE3.getName()))) {
inputPath = inputPath + TOPK_ROLLUP_PHASE3.getName();
} else if (fs.exists(new Path(inputPath + ROLLUP_PHASE4.getName()))) {
inputPath = inputPath + ROLLUP_PHASE4.getName();
} else {
inputPath = inputPath + AGGREGATION.getName();
}
config.setProperty(StarTreeGenerationConstants.STAR_TREE_GEN_INPUT_PATH.toString(), inputPath);
config.setProperty(StarTreeGenerationConstants.STAR_TREE_GEN_OUTPUT_PATH.toString(),
getMetricIndexDir(root, collection, flowSpec, minTime, maxTime) + File.separator
+ STARTREE_GENERATION.getName());
return config;
}
},
STARTREE_BOOTSTRAP_PHASE1 {
@Override
Class<?> getKlazz() {
return StarTreeBootstrapPhaseOneJob.class;
}
@Override
String getDescription() {
return "Sums raw Avro time-series data by dimension key";
}
@Override
Properties getJobProperties(Properties inputConfig, String root, String collection,
FlowSpec flowSpec, DateTime minTime, DateTime maxTime, String inputPaths)
throws Exception {
Properties config = new Properties();
String dimensionIndexDirRef =
inputConfig.getProperty(ThirdEyeJobConstants.THIRDEYE_DIMENSION_INDEX_REF.getName());
config.setProperty(
StarTreeBootstrapPhaseOneConstants.STAR_TREE_BOOTSTRAP_CONFIG_PATH.toString(),
getConfigPath(root, collection));
config.setProperty(
StarTreeBootstrapPhaseOneConstants.STAR_TREE_BOOTSTRAP_INPUT_AVRO_SCHEMA.toString(),
getSchemaPath(root, collection));
config.setProperty(
StarTreeBootstrapPhaseTwoConstants.STAR_TREE_GENERATION_OUTPUT_PATH.toString(),
getDimensionIndexDir(root, collection, dimensionIndexDirRef) + File.separator
+ STARTREE_GENERATION.getName());
boolean skipMissing = Boolean.parseBoolean(inputConfig.getProperty(
ThirdEyeJobConstants.THIRDEYE_SKIP_MISSING.getName(), DEFAULT_SKIP_MISSING));
if (skipMissing) {
inputPaths = getFilteredInputPaths(inputPaths);
}
config.setProperty(
StarTreeBootstrapPhaseOneConstants.STAR_TREE_BOOTSTRAP_INPUT_PATH.toString(),
inputPaths);
config.setProperty(
StarTreeBootstrapPhaseOneConstants.STAR_TREE_BOOTSTRAP_OUTPUT_PATH.toString(),
getMetricIndexDir(root, collection, flowSpec, minTime, maxTime) + File.separator
+ STARTREE_BOOTSTRAP_PHASE1.getName());
config.setProperty(
StarTreeBootstrapPhaseOneConstants.STAR_TREE_BOOTSTRAP_COMPACTION.toString(),
inputConfig.getProperty(ThirdEyeJobConstants.THIRDEYE_COMPACTION.getName(),
DEFAULT_THIRDEYE_COMPACTION));
String converterClass = inputConfig.getProperty(ThirdEyeJobConstants.THIRDEYE_INPUT_CONVERTER_CLASS.getName(),
DEFAULT_CONVERTER_CLASS);
if (converterClass.isEmpty()) {
converterClass = DEFAULT_CONVERTER_CLASS;
}
config.setProperty(
StarTreeBootstrapPhaseOneConstants.STAR_TREE_BOOTSTRAP_CONVERTER_CLASS.toString(), converterClass);
return config;
}
},
STARTREE_BOOTSTRAP_PHASE2 {
@Override
Class<?> getKlazz() {
return StarTreeBootstrapPhaseTwoJob.class;
}
@Override
String getDescription() {
return "Groups records by star tree leaf node and creates leaf buffers";
}
@Override
Properties getJobProperties(Properties inputConfig, String root, String collection,
FlowSpec flowSpec, DateTime minTime, DateTime maxTime, String inputPaths)
throws Exception {
Properties config = new Properties();
String dimensionIndexDirRef =
inputConfig.getProperty(ThirdEyeJobConstants.THIRDEYE_DIMENSION_INDEX_REF.getName());
config.setProperty(
StarTreeBootstrapPhaseTwoConstants.STAR_TREE_BOOTSTRAP_PHASE2_CONFIG_PATH.toString(),
getConfigPath(root, collection));
config.setProperty(
StarTreeBootstrapPhaseTwoConstants.STAR_TREE_GENERATION_OUTPUT_PATH.toString(),
getDimensionIndexDir(root, collection, dimensionIndexDirRef) + File.separator
+ STARTREE_GENERATION.getName());
config.setProperty(
StarTreeBootstrapPhaseTwoConstants.STAR_TREE_BOOTSTRAP_PHASE2_INPUT_PATH.toString(),
getMetricIndexDir(root, collection, flowSpec, minTime, maxTime) + File.separator
+ STARTREE_BOOTSTRAP_PHASE1.getName());
config.setProperty(
StarTreeBootstrapPhaseTwoConstants.STAR_TREE_BOOTSTRAP_PHASE2_OUTPUT_PATH.toString(),
getMetricIndexDir(root, collection, flowSpec, minTime, maxTime) + File.separator
+ STARTREE_BOOTSTRAP_PHASE2.getName());
return config;
}
},
SERVER_PACKAGE {
@Override
Class<?> getKlazz() {
return null; // unused
}
@Override
String getDescription() {
return "Packages data to upload to thirdeye.server.uri";
}
@Override
Properties getJobProperties(Properties inputConfig, String root, String collection,
FlowSpec flowSpec, DateTime minTime, DateTime maxTime, String inputPaths)
throws Exception {
return null; // unused
}
},
SERVER_UPLOAD {
@Override
Class<?> getKlazz() {
return null; // unused
}
@Override
String getDescription() {
return "Pushes packaged data to thirdeye.server.uri";
}
@Override
Properties getJobProperties(Properties inputConfig, String root, String collection,
FlowSpec flowSpec, DateTime minTime, DateTime maxTime, String inputPaths)
throws Exception {
return null; // unused
}
},
CLEANUP {
@Override
Class<?> getKlazz() {
return null;
}
@Override
String getDescription() {
return "Cleans up folders older than thirdeye.cleanup.daysago ago if "
+ "thirdeye.cleanup.skip is false";
}
@Override
Properties getJobProperties(Properties inputConfig, String root, String collection,
FlowSpec flowSpec, DateTime minTime, DateTime maxTime, String inputPaths)
throws Exception {
Properties config = new Properties();
String thirdeyeCleanupDaysAgo = inputConfig.getProperty(
ThirdEyeJobConstants.THIRDEYE_CLEANUP_DAYSAGO.getName(), DEFAULT_CLEANUP_DAYS_AGO);
config.setProperty(ThirdEyeJobConstants.THIRDEYE_CLEANUP_DAYSAGO.getName(),
thirdeyeCleanupDaysAgo);
String thirdeyeCleanupSkip = inputConfig.getProperty(
ThirdEyeJobConstants.THIRDEYE_CLEANUP_SKIP.getName(), DEFAULT_CLEANUP_SKIP);
config.setProperty(ThirdEyeJobConstants.THIRDEYE_CLEANUP_SKIP.getName(),
thirdeyeCleanupSkip);
String schedule =
inputConfig.getProperty(ThirdEyeJobConstants.THIRDEYE_FLOW_SCHEDULE.getName());
config.setProperty(ThirdEyeJobConstants.THIRDEYE_FLOW_SCHEDULE.getName(), schedule);
return config;
}
};
abstract Class<?> getKlazz();
abstract String getDescription();
abstract Properties getJobProperties(Properties inputConfig, String root, String collection,
FlowSpec flowSpec, DateTime minTime, DateTime maxTime, String inputPaths) throws Exception;
String getName() {
return this.name().toLowerCase();
}
String getAnalysisPath(String root, String collection) {
return getCollectionDir(root, collection) + File.separator + FlowSpec.DIMENSION_INDEX
+ File.separator + ANALYSIS.getName();
}
String getMetricIndexDir(String root, String collection, FlowSpec flowSpec, DateTime minTime,
DateTime maxTime) throws IOException {
return getCollectionDir(root, collection) + File.separator + flowSpec.name() + File.separator
+ "data_" + StarTreeConstants.DATE_TIME_FORMATTER.print(minTime) + "_"
+ StarTreeConstants.DATE_TIME_FORMATTER.print(maxTime);
}
String getDimensionIndexDir(String root, String collection, String dimensionIndexDirRef)
throws IOException {
return getCollectionDir(root, collection) + File.separator + FlowSpec.DIMENSION_INDEX
+ File.separator + dimensionIndexDirRef;
}
String getConfigPath(String root, String collection) {
return getCollectionDir(root, collection) + File.separator
+ StarTreeConstants.CONFIG_FILE_NAME;
}
String getSchemaPath(String root, String collection) {
return getCollectionDir(root, collection) + File.separator + AVRO_SCHEMA;
}
String getPinotSchemaPath(String root, String collection) {
return getCollectionDir(root, collection) + File.separator + PINOT_JSON_SCHEMA;
}
}
private static void usage() {
System.err.println(USAGE);
for (PhaseSpec phase : PhaseSpec.values()) {
System.err.printf("%-30s : %s\n", phase.getName(), phase.getDescription());
}
}
private static String getAndCheck(String name, Properties properties) {
String value = properties.getProperty(name);
if (value == null) {
throw new IllegalArgumentException("Must provide " + name);
}
return value;
}
private final String phaseName;
private final Properties inputConfig;
public ThirdEyeJob(String jobName, Properties config) {
String phaseFromConfig = config.getProperty(ThirdEyeJobConstants.THIRDEYE_PHASE.getName());
if (phaseFromConfig != null) {
this.phaseName = phaseFromConfig;
} else {
this.phaseName = jobName;
}
this.inputConfig = config;
}
private IndexMetadata mergeIndexMetadata(List<IndexMetadata> indexMetdataList,
Long startTimeMillis, Long endTimeMillis, String schedule) {
Long min = Long.MAX_VALUE;
Long minMillis = Long.MAX_VALUE;
Long max = 0L;
Long maxMillis = 0L;
Long startTime = Long.MAX_VALUE;
Long endTime = 0L;
String aggregationGranularity = "";
int bucketSize = 0;
for (IndexMetadata indexMetadata : indexMetdataList) {
if (aggregationGranularity.equals("")) {
aggregationGranularity = indexMetadata.getAggregationGranularity();
bucketSize = indexMetadata.getBucketSize();
}
if (indexMetadata.getMinDataTime() != null && indexMetadata.getMinDataTime() < min) {
min = indexMetadata.getMinDataTime();
minMillis = indexMetadata.getMinDataTimeMillis();
}
if (indexMetadata.getMaxDataTime() != null && indexMetadata.getMaxDataTime() > max) {
max = indexMetadata.getMaxDataTime();
maxMillis = indexMetadata.getMaxDataTimeMillis();
}
}
startTime =
TimeUnit.valueOf(aggregationGranularity).convert(startTimeMillis, TimeUnit.MILLISECONDS)
/ bucketSize;
endTime = TimeUnit.valueOf(aggregationGranularity).convert(endTimeMillis, TimeUnit.MILLISECONDS)
/ bucketSize;
IndexMetadata mergedIndexMetadata =
new IndexMetadata(min, max, minMillis, maxMillis, startTime, endTime, startTimeMillis,
endTimeMillis, schedule, aggregationGranularity, bucketSize, IndexFormat.VARIABLE_SIZE);
return mergedIndexMetadata;
}
private void writeMergedIndexMetadataServerPush(FileSystem fileSystem, Path metadataPath,
IndexMetadata mergedIndexMetadata) throws IOException {
OutputStream os = null;
try {
os = fileSystem.create(metadataPath);
Properties mergedProperties = mergedIndexMetadata.toProperties();
mergedProperties.store(os, "Merged index metadata properties");
} finally {
if (os != null)
os.close();
}
}
private static String getFilteredInputPaths(String inputPaths) throws IOException {
FileSystem fileSystem = FileSystem.get(new Configuration());
String[] inputs = inputPaths.split(INPUT_PATHS_JOINER);
List<String> filteredInputs = new ArrayList<String>();
for (String input : inputs) {
if (fileSystem.exists(new Path(input))) {
filteredInputs.add(input);
} else {
LOGGER.info("Skipping missing input path {}", input);
}
}
if (filteredInputs.isEmpty()) {
throw new IllegalArgumentException(
"Must provide valid existing " + ThirdEyeJobConstants.INPUT_PATHS.getName());
}
return Joiner.on(INPUT_PATHS_JOINER).join(filteredInputs);
}
public void serverPackage(FileSystem fileSystem, String root, String collection,
FlowSpec flowSpec, DateTime minTime, DateTime maxTime) throws Exception {
// Get config file to package
Path configPath = new Path(
root + File.separator + collection + File.separator + StarTreeConstants.CONFIG_FILE_NAME);
// Get data to package : overwrites existing data.tar.gz
String schedule =
inputConfig.getProperty(ThirdEyeJobConstants.THIRDEYE_FLOW_SCHEDULE.getName());
String metricIndexDir = PhaseSpec.STARTREE_BOOTSTRAP_PHASE2.getMetricIndexDir(root, collection,
flowSpec, minTime, maxTime);
String bootstrapPhase2Output =
metricIndexDir + File.separator + PhaseSpec.STARTREE_BOOTSTRAP_PHASE2.getName();
Path dataPath = new Path(bootstrapPhase2Output);
String outputTarGzFile = metricIndexDir + "/data.tar.gz";
Path outputTarGzFilePath = new Path(outputTarGzFile);
if (fileSystem.exists(outputTarGzFilePath)) {
fileSystem.delete(outputTarGzFilePath, false);
}
LOGGER.info("START: Creating output {} to upload to server ", outputTarGzFilePath.getName());
fileSystem.delete(new Path(dataPath, StarTreeConstants.METADATA_FILE_NAME), false);
TarGzBuilder builder = new TarGzBuilder(outputTarGzFile, fileSystem, fileSystem);
RemoteIterator<LocatedFileStatus> listFiles = fileSystem.listFiles(dataPath, false);
List<IndexMetadata> indexMetadataList = new ArrayList<IndexMetadata>();
// get merged metadata file
while (listFiles.hasNext()) {
Path path = listFiles.next().getPath();
IndexMetadata localIndexMetadata = builder.getMetadataObjectBootstrap(path);
if (localIndexMetadata != null) {
indexMetadataList.add(localIndexMetadata);
}
}
if (indexMetadataList.size() == 0) {
throw new IllegalStateException("No metadata files found");
}
Path metadataPath = new Path(bootstrapPhase2Output, StarTreeConstants.METADATA_FILE_NAME);
IndexMetadata mergedIndexMetadata =
mergeIndexMetadata(indexMetadataList, minTime.getMillis(), maxTime.getMillis(), schedule);
writeMergedIndexMetadataServerPush(fileSystem, metadataPath, mergedIndexMetadata);
listFiles = fileSystem.listFiles(dataPath, false);
while (listFiles.hasNext()) {
Path path = listFiles.next().getPath();
LOGGER.info("Adding {}, to {}", path, outputTarGzFile);
if (path.getName().equals(StarTreeConstants.TREE_FILE_NAME)
|| path.getName().equals(StarTreeConstants.METADATA_FILE_NAME)) {
builder.addFileEntry(path);
} else {
// its either dimensionStore.tar.gz or metricStore-x.tar.gz
builder.addTarGzFile(path);
}
}
builder.addFileEntry(configPath);
builder.finish();
if (fileSystem.exists(outputTarGzFilePath)) {
LOGGER.info("Successfully created {}.", outputTarGzFilePath);
} else {
throw new RuntimeException("Creation of" + outputTarGzFile + " failed");
}
}
public void serverUpload(FileSystem fileSystem, String root, String collection, FlowSpec flowSpec,
DateTime minTime, DateTime maxTime) throws Exception {
String thirdEyeServerUri =
inputConfig.getProperty(ThirdEyeJobConstants.THIRDEYE_SERVER_URI.getName());
if (thirdEyeServerUri == null) {
throw new IllegalArgumentException(
"Must provide " + ThirdEyeJobConstants.THIRDEYE_SERVER_URI.getName() + " in properties");
}
// Push config (may 409 but that's okay)
Path configPath = new Path(
root + File.separator + collection + File.separator + StarTreeConstants.CONFIG_FILE_NAME);
InputStream configData = fileSystem.open(configPath);
int responseCode = StarTreeJobUtils.pushConfig(configData, thirdEyeServerUri, collection);
configData.close();
LOGGER.info("Load {} #=> {}", configPath, responseCode);
// Push data
String schedule =
inputConfig.getProperty(ThirdEyeJobConstants.THIRDEYE_FLOW_SCHEDULE.getName());
String metricIndexDir = PhaseSpec.STARTREE_BOOTSTRAP_PHASE2.getMetricIndexDir(root, collection,
flowSpec, minTime, maxTime);
String outputTarGzFile = metricIndexDir + "/data.tar.gz";
Path outputTarGzFilePath = new Path(outputTarGzFile);
if (fileSystem.exists(outputTarGzFilePath)) {
LOGGER.info("Uploading {} of size:{} to ThirdEye Server: {}", outputTarGzFile,
fileSystem.getFileStatus(outputTarGzFilePath).getLen(), thirdEyeServerUri);
FSDataInputStream outputDataStream = fileSystem.open(outputTarGzFilePath);
responseCode = StarTreeJobUtils.pushData(outputDataStream, thirdEyeServerUri, collection,
minTime, maxTime, schedule);
LOGGER.info("Load {} #=> response code: {}", outputTarGzFile, responseCode);
} else {
throw new RuntimeException(
"Uploading of " + outputTarGzFile + " failed : package does not exist");
}
String thirdeyeFolderPermission = inputConfig.getProperty(
ThirdEyeJobConstants.THIRDEYE_FOLDER_PERMISSION.getName(), DEFAULT_FOLDER_PERMISSION);
if (!thirdeyeFolderPermission.matches(FOLDER_PERMISSION_REGEX)) {
throw new IllegalArgumentException(
"Invalid folder permission mode. Must provide octal " + FOLDER_PERMISSION_REGEX);
}
String dimensionIndexDir =
PhaseSpec.STARTREE_BOOTSTRAP_PHASE2.getDimensionIndexDir(root, collection,
Joiner.on(DATA_FOLDER_JOINER).join(StarTreeConstants.DATA_DIR_PREFIX,
StarTreeConstants.DATE_TIME_FORMATTER.print(minTime),
StarTreeConstants.DATE_TIME_FORMATTER.print(maxTime)));
FsShell shell = new FsShell(fileSystem.getConf());
LOGGER.info("Changing permission of {} to {}", dimensionIndexDir, thirdeyeFolderPermission);
shell.run(new String[] {
"-chmod", "-R", thirdeyeFolderPermission, dimensionIndexDir
});
LOGGER.info("Changing permission of {} to {}", metricIndexDir, thirdeyeFolderPermission);
shell.run(new String[] {
"-chmod", "-R", thirdeyeFolderPermission, metricIndexDir
});
}
private void purgeLowerGranularity(DateTime minTime, DateTime maxTime, String schedule,
Path dimensionIndexDir, Path metricIndexDir) throws IOException {
LOGGER.info("Created folder with higher granularity {} ",
new Path(dimensionIndexDir,
Joiner.on(DATA_FOLDER_JOINER).join(StarTreeConstants.DATA_DIR_PREFIX,
StarTreeConstants.DATE_TIME_FORMATTER.print(minTime),
StarTreeConstants.DATE_TIME_FORMATTER.print(maxTime))));
LOGGER.info("Created folder with higher granularity {} ",
new Path(metricIndexDir,
Joiner.on(DATA_FOLDER_JOINER).join(StarTreeConstants.DATA_DIR_PREFIX,
StarTreeConstants.DATE_TIME_FORMATTER.print(minTime),
StarTreeConstants.DATE_TIME_FORMATTER.print(maxTime))));
DateTime start = minTime;
DateTime end;
FileSystem fileSystem = FileSystem.get(new Configuration());
while (start.compareTo(maxTime) < 0) {
if (schedule.equals(FlowSchedule.DAILY.name())) {
end = start.plusHours(1);
} else if (schedule.equals(FlowSchedule.MONTHLY.name())) {
end = start.plusDays(1);
} else {
return;
}
Path folder = new Path(Joiner.on(DATA_FOLDER_JOINER).join(StarTreeConstants.DATA_DIR_PREFIX,
StarTreeConstants.DATE_TIME_FORMATTER.print(start),
StarTreeConstants.DATE_TIME_FORMATTER.print(end)));
Path dimensionFolder = new Path(dimensionIndexDir, folder);
Path metricFolder = new Path(metricIndexDir, folder);
if (fileSystem.exists(dimensionFolder)) {
LOGGER.info("Deleting folder {} with overlapping data and lower granularity",
dimensionFolder);
fileSystem.delete(dimensionFolder, true);
}
if (fileSystem.exists(metricFolder)) {
LOGGER.info("Deleting folder {} with overlapping data and lower granularity", metricFolder);
fileSystem.delete(metricFolder, true);
}
start = end;
}
}
@SuppressWarnings("unchecked")
public void run() throws Exception {
LOGGER.info("Input config:{}", inputConfig);
PhaseSpec phaseSpec;
try {
phaseSpec = PhaseSpec.valueOf(phaseName.toUpperCase());
} catch (Exception e) {
usage();
throw e;
}
/**
* This phase is optional for the pipeline
*/
if (PhaseSpec.JOIN.equals(phaseSpec)) {
JoinPhaseJob job = new JoinPhaseJob("Join Job", inputConfig);
job.run();
return;
}
if (PhaseSpec.TRANSFORM.equals(phaseSpec)) {
TransformPhaseJob job = new TransformPhaseJob("Transform Job", inputConfig);
job.run();
return;
}
String root = getAndCheck(ThirdEyeJobConstants.THIRDEYE_ROOT.getName(), inputConfig);
String collection =
getAndCheck(ThirdEyeJobConstants.THIRDEYE_COLLECTION.getName(), inputConfig);
String inputPaths = getAndCheck(ThirdEyeJobConstants.INPUT_PATHS.getName(), inputConfig);
FlowSpec flowSpec = null;
switch (phaseSpec) {
case WAIT:
case ANALYSIS:
case AGGREGATION:
case SEGMENT_CREATION:
case SEGMENT_PUSH:
case TOPK_ROLLUP_PHASE1:
case TOPK_ROLLUP_PHASE2:
case TOPK_ROLLUP_PHASE3:
case ROLLUP_PHASE1:
case ROLLUP_PHASE2:
case ROLLUP_PHASE3:
case ROLLUP_PHASE4:
case PARTITION:
case STARTREE_GENERATION:
flowSpec = FlowSpec.DIMENSION_INDEX;
break;
case STARTREE_BOOTSTRAP_PHASE1:
case STARTREE_BOOTSTRAP_PHASE2:
case SERVER_PACKAGE:
case SERVER_UPLOAD:
case CLEANUP:
flowSpec = FlowSpec.METRIC_INDEX;
break;
default:
break;
}
// Get min / max time
DateTime minTime;
DateTime maxTime;
String minTimeProp = inputConfig.getProperty(ThirdEyeJobConstants.THIRDEYE_TIME_MIN.getName());
String maxTimeProp = inputConfig.getProperty(ThirdEyeJobConstants.THIRDEYE_TIME_MAX.getName());
String timePathProp =
inputConfig.getProperty(ThirdEyeJobConstants.THIRDEYE_TIME_PATH.getName());
FileSystem fileSystem = FileSystem.get(new Configuration());
if (minTimeProp != null && maxTimeProp != null) // user provided, override
{
minTime = ISODateTimeFormat.dateTimeParser().parseDateTime(minTimeProp);
maxTime = ISODateTimeFormat.dateTimeParser().parseDateTime(maxTimeProp);
} else if (timePathProp != null) // use path managed by preparation jobs
{
InputStream inputStream = fileSystem.open(new Path(timePathProp));
Properties timePathProps = new Properties();
timePathProps.load(inputStream);
inputStream.close();
minTimeProp = timePathProps.getProperty(ThirdEyeJobConstants.THIRDEYE_TIME_MIN.getName());
maxTimeProp = timePathProps.getProperty(ThirdEyeJobConstants.THIRDEYE_TIME_MAX.getName());
minTime = ISODateTimeFormat.dateTimeParser().parseDateTime(minTimeProp);
maxTime = ISODateTimeFormat.dateTimeParser().parseDateTime(maxTimeProp);
} else {
throw new IllegalStateException(
"Must specify either " + ThirdEyeJobConstants.THIRDEYE_TIME_PATH.getName() + " or "
+ ThirdEyeJobConstants.THIRDEYE_TIME_MIN.getName() + " and "
+ ThirdEyeJobConstants.THIRDEYE_TIME_MAX.getName());
}
if (PhaseSpec.SERVER_PACKAGE.equals(phaseSpec)) {
serverPackage(fileSystem, root, collection, flowSpec, minTime, maxTime);
} else if (PhaseSpec.SERVER_UPLOAD.equals(phaseSpec)) {
serverUpload(fileSystem, root, collection, flowSpec, minTime, maxTime);
} else if (PhaseSpec.WAIT.equals(phaseSpec)) {
// Check if we need to poll for any input paths
boolean pollEnable = Boolean.parseBoolean(inputConfig
.getProperty(ThirdEyeJobConstants.THIRDEYE_POLL_ENABLE.getName(), DEFAULT_POLL_ENABLE));
if (pollEnable) {
long elapsedTime = 0;
long pollStart = (new DateTime()).getMillis();
long pollFrequency = Long.parseLong(inputConfig.getProperty(
ThirdEyeJobConstants.THIRDEYE_POLL_FREQUENCY.getName(), DEFAULT_POLL_FREQUENCY));
long pollTimeout = Long.parseLong(inputConfig.getProperty(
ThirdEyeJobConstants.THIRDEYE_POLL_TIMEOUT.getName(), DEFAULT_POLL_TIMEOUT));
String[] inputs = inputConfig.getProperty(ThirdEyeJobConstants.INPUT_PATHS.getName())
.split(INPUT_PATHS_JOINER);
List<String> missingInputs;
while (elapsedTime < pollTimeout) {
missingInputs = new ArrayList<String>();
for (String input : inputs) {
LOGGER.info("Checking path {}", input);
if (!fileSystem.exists(new Path(input))) {
missingInputs.add(input);
LOGGER.info("Missing input {}", input);
} else {
try {
fileSystem.getFileStatus(new Path(input));
if (fileSystem.exists(new Path(input, TEMPORARY_FOLDER))) {
missingInputs.add(input);
LOGGER.info("Data generation in progress");
} else {
LOGGER.info("Path available {}", input);
}
} catch (AccessControlException e) {
missingInputs.add(input);
LOGGER.warn("No access to path {}", input, e);
}
}
}
if (missingInputs.size() == 0) {
LOGGER.info("All paths available");
break;
}
LOGGER.info("Poll after {} milliseconds", pollFrequency);
TimeUnit.MILLISECONDS.sleep(pollFrequency);
elapsedTime = new DateTime().getMillis() - pollStart;
LOGGER.info("Time elapsed {} milliseconds", elapsedTime);
inputs = missingInputs.toArray(new String[missingInputs.size()]);
}
if (elapsedTime > pollTimeout) {
LOGGER.info("Timed out waiting for input");
}
}
String thirdeyeCheckCompletenessClass = inputConfig
.getProperty(ThirdEyeJobConstants.THIRDEYE_CHECK_COMPLETENESS_CLASS.getName());
if (thirdeyeCheckCompletenessClass != null && !thirdeyeCheckCompletenessClass.isEmpty()) {
LOGGER.info("Initializing class {}", thirdeyeCheckCompletenessClass);
Constructor<?> constructor = Class.forName(thirdeyeCheckCompletenessClass).getConstructor();
WaitUDF waitUdf = (WaitUDF) constructor.newInstance();
waitUdf.init(inputConfig);
boolean complete = waitUdf.checkCompleteness();
if (!complete) {
throw new RuntimeException("Input folder {} has not received all records");
}
}
} else if (PhaseSpec.CLEANUP.equals(phaseSpec)) {
boolean cleanupSkip = Boolean.parseBoolean(inputConfig
.getProperty(ThirdEyeJobConstants.THIRDEYE_CLEANUP_SKIP.getName(), DEFAULT_CLEANUP_SKIP));
LOGGER.info("cleanup skip {}", cleanupSkip);
if (cleanupSkip) {
LOGGER.info("Skipping cleanup");
} else {
int cleanupDaysAgo = Integer.valueOf(inputConfig.getProperty(
ThirdEyeJobConstants.THIRDEYE_CLEANUP_DAYSAGO.getName(), DEFAULT_CLEANUP_DAYS_AGO));
DateTime cleanupDaysAgoDate = (new DateTime()).minusDays(cleanupDaysAgo);
Path dimensionIndexDir = new Path(
getCollectionDir(root, collection) + File.separator + FlowSpec.DIMENSION_INDEX);
Path metricIndexDir =
new Path(getCollectionDir(root, collection) + File.separator + FlowSpec.METRIC_INDEX);
LOGGER.info("Cleaning up {} {} days ago from paths {} and {}", cleanupDaysAgo,
cleanupDaysAgoDate.getMillis(), dimensionIndexDir, metricIndexDir);
boolean preserveAggregation = Boolean.parseBoolean(inputConfig.getProperty(
ThirdEyeJobConstants.THIRDEYE_PRESERVE_TIME_COMPACTION.getName(),
DEFAULT_THIRDEYE_PRESERVE_TIME_COMPACTION));
// list folders in dimensionDir starting with data_
if (fileSystem.exists(dimensionIndexDir)) {
FileStatus[] fileStatus = fileSystem.listStatus(dimensionIndexDir, new PathFilter() {
@Override
public boolean accept(Path path) {
return path.getName().startsWith(StarTreeConstants.DATA_DIR_PREFIX);
}
});
for (FileStatus file : fileStatus) {
if (preserveAggregation) {
FileStatus[] phases = fileSystem.listStatus(file.getPath(), new PathFilter() {
@Override
public boolean accept(Path path) {
return !path.getName().equals(PhaseSpec.AGGREGATION.getName())
&& !path.getName().equals(PhaseSpec.SEGMENT_CREATION.getName());
}
});
for (FileStatus phase : phases) {
cleanupFolder(phase, cleanupDaysAgoDate, fileSystem);
}
} else {
FileStatus[] phases = fileSystem.listStatus(file.getPath(), new PathFilter() {
@Override
public boolean accept(Path path) {
return !path.getName().equals(PhaseSpec.SEGMENT_CREATION.getName());
}
});
for (FileStatus phase : phases) {
cleanupFolder(phase, cleanupDaysAgoDate, fileSystem);
}
cleanupFolder(file, cleanupDaysAgoDate, fileSystem);
}
}
}
// list folders in metricDir starting with data_
if (fileSystem.exists(metricIndexDir)) {
FileStatus[] fileStatus = fileSystem.listStatus(metricIndexDir, new PathFilter() {
@Override
public boolean accept(Path path) {
return path.getName().startsWith(StarTreeConstants.DATA_DIR_PREFIX);
}
});
for (FileStatus file : fileStatus) {
// get last modified date
DateTime lastModifiedDate = new DateTime(file.getModificationTime());
if (lastModifiedDate.isBefore(cleanupDaysAgoDate.getMillis())) {
Path startreeBootstrapPath1 = new Path(file.getPath(),
PhaseSpec.STARTREE_BOOTSTRAP_PHASE1.getName().toLowerCase());
if (fileSystem.exists(startreeBootstrapPath1)) {
LOGGER.info("Deleting {}", startreeBootstrapPath1);
fileSystem.delete(startreeBootstrapPath1, true);
}
Path startreeBootstrapPath2 = new Path(file.getPath(),
PhaseSpec.STARTREE_BOOTSTRAP_PHASE2.getName().toLowerCase());
if (fileSystem.exists(startreeBootstrapPath2)) {
LOGGER.info("Deleting {}", startreeBootstrapPath2);
fileSystem.delete(startreeBootstrapPath2, true);
}
}
}
}
String schedule =
inputConfig.getProperty(ThirdEyeJobConstants.THIRDEYE_FLOW_SCHEDULE.getName());
LOGGER.info("schedule : {} ", schedule);
purgeLowerGranularity(minTime, maxTime, schedule, dimensionIndexDir, metricIndexDir);
}
} else // Hadoop job
{
if (FlowSpec.METRIC_INDEX.equals(flowSpec)) {
String dimensionIndexRef =
inputConfig.getProperty(ThirdEyeJobConstants.THIRDEYE_DIMENSION_INDEX_REF.getName());
if (dimensionIndexRef == null) {
String msg = "dimensionIndexRef:" + dimensionIndexRef + ".Must provide "
+ ThirdEyeJobConstants.THIRDEYE_DIMENSION_INDEX_REF.getName() + " in properties";
LOGGER.error(msg);
throw new IllegalArgumentException(msg);
}
}
// Construct job properties
Properties jobProperties = phaseSpec.getJobProperties(inputConfig, root, collection, flowSpec,
minTime, maxTime, inputPaths);
String numReducers = inputConfig.getProperty(phaseSpec.getName() + ".num.reducers");
if (numReducers != null) {
jobProperties.put("num.reducers", numReducers);
}
// Instantiate the job
Constructor<Configured> constructor = (Constructor<Configured>) phaseSpec.getKlazz()
.getConstructor(String.class, Properties.class);
Configured instance = constructor.newInstance(phaseSpec.getName(), jobProperties);
setMapreduceConfig(instance.getConf());
// Run the job
Method runMethod = instance.getClass().getMethod("run");
Job job = (Job) runMethod.invoke(instance);
if (job != null) {
JobStatus status = job.getStatus();
if (status.getState() != JobStatus.State.SUCCEEDED) {
throw new RuntimeException(
"Job " + job.getJobName() + " failed to execute: Ran with config:" + jobProperties);
}
}
}
}
private void cleanupFolder(FileStatus fileStatus, DateTime cleanupDaysAgoDate,
FileSystem fileSystem) throws IOException {
DateTime lastModifiedDate = new DateTime(fileStatus.getModificationTime());
if (lastModifiedDate.isBefore(cleanupDaysAgoDate.getMillis())) {
LOGGER.info("Deleting {}", fileStatus.getPath());
fileSystem.delete(fileStatus.getPath(), true);
}
}
public static void main(String[] args) throws Exception {
if (args.length != 2) {
usage();
System.exit(1);
}
String phaseName = args[0];
Properties config = new Properties();
config.load(new FileInputStream(args[1]));
new ThirdEyeJob(phaseName, config).run();
}
private static String getCollectionDir(String root, String collection) {
return root == null ? collection : root + File.separator + collection;
}
private void setMapreduceConfig(Configuration configuration) {
String mapreduceConfig =
inputConfig.getProperty(ThirdEyeJobConstants.THIRDEYE_MR_CONF.getName());
if (mapreduceConfig != null && !mapreduceConfig.isEmpty()) {
String[] options = mapreduceConfig.split(",");
for (String option : options) {
String[] configs = option.split("=", 2);
if (configs.length == 2) {
LOGGER.info("Setting job configuration {} to {}", configs[0], configs[1]);
configuration.set(configs[0], configs[1]);
}
}
}
}
} |
The Vanishing Majority Gate Trading Power and Speed for Reliability In this paper we are going to explore low-level implementation issues for fault-tolerant adders based on multiplexing using majority gates (MAJ). We shall analyze the particular case of a 32-bit ripple carry adder (RCA), as well as different redundant designs using MAJ-3 (MAJ of fan-in 3) multiplexed RCAs: (i) with classical MAJ-3 gates in the restorative stages; (ii) with inverters driven by shortcircuited outputs at each restorative stage; and finally, (iii) only with short-circuited outputs at each restorative stage. From one solution to the next, the restorative MAJ-3 gates get simpler and simpler. These simplifications translate into different speeds and power consumptions; challenging aspects of future nanoelectronics. All these circuits have been designed and simulated in subthreshold. The speed and power will be reported and compared for designs in 0.18 m as well as 70 nm (using the Berkeley Predictive Technology Model). The results reveal interesting power-speed-reliability tradeoffs. In two of these designs, depending on the way the MAJ-3 function is implemented, defects translate into increased power, and suggest a (simple) way of detecting them. A detection circuit can trigger reconfiguration at a higher level, leading to a seamless transition from a fault-tolerant circuit to a defecttolerant system. The main advantage of such an approach would be that reconfiguration could be done on-line, i.e., while the circuit is still operating correctly. |
def is_blkdev_supported():
o = get_kernel_module()
if o:
return o.is_blkdev_supported()
else:
return False |
This year’s nominees show how visual effects have spread from summer blockbusters to genres as diverse as superheroes, different flavors of fantasy, more traditional sci-fi territory, and even the art-house film. For each nominee, there’s a moment that makes it worthy of an Oscar nomination. Here, the visual-effects supervisors on the nominated films break down the key challenges and talk about the sequence that clinched the nomination.
Tech breakthrough: The complexity and number of techniques used to create the digital creatures. “It’s a combination of lots of things to get a creature to that point”, says Letteri. “It’s muscles, it’s skin, it’s facial capture, it’s performance capture”. All those things had to come together to bring to convincing life six leading digital characters with dialogue.
Defining the aesthetic: “We were grounded in the Middle Earth we had established for The Lord of the Rings”, says Letteri. “For the landscapes and the environments, we wanted to extend that Tolkien-esque feeling, borrowing from what we had on the previous film, trying to keep the same look for Rivendell, for example, but kind of expanding it. Same thing with Gollum—we were trying to keep his same look, but bring him into a new dimension of what we could do 10 years on”.
Biggest challenge: The quantity of digital characters. “You’ve got dialogue, you’ve got personalities, you’ve got unique looks”, says Letteri. “You’ve got to have everything working: You’ve got to have the fur working, the eyes, the skin, the muscles, the performances—not only the capture but the animation side”.
The clincher: The confrontation between Martin Freeman’s Bilbo and Gollum, played via motion capture by Andy Serkis. “We all had a bit of nervousness going into creating (Gollum) because we had done him 10 years ago, and we spent so much time in the last 10 years really trying to delve into what makes a performance resonate with an audience”, says Letteri. “You’ve got here a nine-minute dialogue scene with a real character and a digital character, and it’s watchable in a way that keeps you engaged the whole way through”.
Tech breakthrough: Two of the major visual elements were done mostly with digital effects: The water and the tiger. “It was just pushing the bar for the realism of the tiger and the other animals involved, trying to blend water from a tank into CG water in stereo was a challenge”, says Westenhofer.
Defining the aesthetic: Westenhofer describes the look of the effects as “hyper-dreamlike reality”. “It’s a story being told by Pi, so there’s an element of his recollection and the human’s ability to exaggerate when they recollect”, he says. “That allows for a bit of stylization in the amount of color and detail”.
Biggest challenge: It’s a toss-up between the water and the animals. “Fourteen percent of the animals were real and the rest were digital, and we often cut back to back between them, so it forced our hand to make the matches as perfect as possible”, says Westenhofer. “Everything from the moment they set sail to when he lands on the beach, it’s a boy on a boat in front of a blue screen”.
The clincher: A shot where Pi pulls the tiger’s head into his lap and pets it. “We shot him on the boat in a gimbal, and he pulls a blue sock into his lap and he pets the blue sock. And we replaced that with our digital tiger, fitting in the animation to what he did. In stereo, it had to be perfectly precise to line up with everything, and then we had to animate the hair to respond to his hand as it moves back and forth”.
Tech breakthrough: The Hulk. “We leveraged on previous digital characters we had done, but really had to rebuild and improve the way our characters move, making it incredibly accurate in terms of the way the skeleton under his skin drives his muscles, which then drives his skin”, says White.
in New York City—some is Cleveland, where we did simpler set extensions, and then a significant portion was shot on a green-screen stage in New Mexico—those are things where we didn’t want the audience to even know there are visual effects”, says White.
Biggest challenge: The Hulk. “There’s a deep ravine to cross there, where it doesn’t look good for quite a long time, and it takes an incredible amount of artistry by the artists working on the shots to make it what it ultimately became”, says White.
The clincher: The climactic battle in New York. White says ILM spent about eight weeks shooting some 2,000 virtual background spheres—extremely high-resolution photographs—from streets and rooftops that were projected onto geography of the city as the basis for the digital city. To this was added the digital aliens and plates of the actors shot, as well as the details required to sell the scene as a full-on battle. “As we put our shots together of, say, Captain America talking to Black Widow, we really wanted to push it toward this feeling of being in the center of a battle. So in every shot we added additional smoke and dust and little embers going through the scene, just trying to really capture that feel of being in the middle of a disaster”.
Tech breakthrough: The specific look director Ridley Scott wanted for the alien creatures required redeveloping some commonly used tools. “We had to do a lot of work to really develop our subsurface scatter lighting technique to get that deep translucency that matched the prosthetics we were using live on set”, says Stammers.
Defining the aesthetic: The look of the alien landscape of LV-223 defined the look of the whole film and was something Scott was quite passionate about. “What we ended up with is this montage of two landscapes that he really liked. And then beyond that, we added additional mountains and sky that was very full of fast-moving clouds, and so you get a sense of constantly fast-moving layers of clouds and bad weather, (then) we could paint the landscape with fast-moving patches of sunlight”.
Biggest challenge: Stammers says the production only had three days to shoot all the references needed at Wadi Rum, Jordan, requiring an incredibly detailed plan. “We planned it out based on our Google Earth map of the location to the point where, for every take that we needed to shoot, we had a helicopter plan of altitude and GPS start and end point, so that we could go to each of the specific points and film the elements we needed in order to map out the terrain and texture it”.
The clincher: Everything came together in the shot of the Prometheus landing on LV-223. “We spent somewhere in the region of 300 or 400 days just on the texture work alone, just to get the level of detail we needed to sell the scale of it”, says Stammers. “All the elements come together in that one shot that we see throughout the rest of the film as well”.
Tech breakthrough: The extensive use of macrophotography in CG visual effects. “It’s very tricky to do macrophotography in a full CG shot, especially when you look at an animal or something close up like that, close up on the eye”, says Nicolas-Troyan. “That’s something that people don’t really realize when they see the movie, but if you pay attention you see there’s a lot of macro shots”.
Defining the aesthetic: Director Rupert Sanders set a distinct tone that required all the visual effects to be based in reality but juxtaposed with unusual situations or actions. “Everything is based on things that exist in the world”, says Nicolas-Troyan. “They might not be in the same place in the world, so we put them all together in this one spot, but they all do exist”.
Biggest challenge: Finding a way to make eight actors appear as dwarves on schedule and on budget. “We were always going to pick the right technique and the most efficient technique for the shot”, says Brennan. “That goes all the way from old-school in-camera tricks to using risers to vary the heights of people, working with prosthetics and costumes to make people appear a little bit different, all the way up to very complex effects like head and face replacements”.
The clincher: The pursuit through the Enchanted Forest, which encompassed all the techniques used in the movie. “Something like 70 percent or 80 percent of the animals that we created for the movie are in that scene, and they are everywhere”, says Nicolas-Troyan. “There’s birds, plants, and then within those scenes you have the dwarves, so we had to use pretty much all our techniques for the dwarves”. |
SEMANTIC ANALYSIS OF SCIENTIFIC DEFINITION CHILDREN WITH SPECIAL EDUCATIONAL NEEDS he article aims at investigation of approaches towards the scientific definition of notion children with special needs. The semantic analysis of scientific notions of needs, special educational needs, have revealed the essence and content of approaches of researchers to the interpretation of the notion children with special educational needs; information about inclusion and inclusive approach in the context of socialization has been systemized. It has been proved, that the notion people with special educational needs can be applied to those children, who need extra additional recourses in the process of being educated. This notion includes a great variety of children (gifted children, children with mental and physical disabilities, physically challenged children, homeless children, orphans). |
package app.dao.daoGenerique.impl;
import java.io.Serializable;
import java.util.ArrayList;
import java.util.Collection;
import java.util.LinkedHashSet;
import java.util.List;
import org.hibernate.Query;
import org.hibernate.Session;
import org.hibernate.SessionFactory;
import org.hibernate.Transaction;
import app.dao.daoGenerique.*;
import app.dao.daoGenerique.utils.*;
/**
* Classe de base pour tous les DAOs, elle implémente les méthodes CRUD
* génériques définit par le contrat GenericDAO<T>. Cette implémentation est
* basée sur Hibernate nativement
*
* @author <a href="mailto:<EMAIL>">T.BOUDAA Ecole Nationale des
* Sciences Appliquées </a>
*
* @param <T>
* le type d'objet métier manipulé
* @param <PK>
* le type utilisé pour l'indentifiant d'un objet métier
*/
public abstract class NativeHibernateGenericDAOImpl<T, PK extends Serializable> implements IgenericDao<T, PK> {
/** La classe BO manipulé par le DAO */
protected Class<T> boClass;
/** Utilisé par tous les DAOs */
//protected final Logger LOGGER;
/** la fabrique des session */
protected SessionFactory sf = SessionFactoryBuilderHibernate4.getSessionFactory();
public NativeHibernateGenericDAOImpl(Class<T> pClass) {
boClass = pClass;
//LOGGER = Logger.getLogger(boClass);
//LOGGER.debug("le dao de " + boClass + " a été initialisé");
}
/**
* Annule la transaction
*
* @param tx
* @param ex
* @throws Exception
*/
protected void handleDaoOpError(Transaction tx, RuntimeException ex) {
if (tx != null) {
tx.rollback();
}
// On trace l'erreur
//LOGGER.error("erreur é cause de l'exception " + ex);
// On remonte l'exception
throw new DataAccessLayerException(ex);
}
protected void closeSession(Session s) {
if (s != null && s.isOpen()) {
s.close();
}
}
protected Session getSession() {
return sf.getCurrentSession();
}
protected boolean anActiveTransactionExists(Session s) {
if (s != null && s.getTransaction() != null)
return s.getTransaction().isActive();
return false;
}
public PK create(T o) {
//LOGGER.debug("appel de la méthode create");
// On obtient la session en cours
Session s = getSession();
PK id = null;
// Si la couche BLL a démarré une transaction
if (anActiveTransactionExists(s)) {
//LOGGER.debug("le DAO utilise la transaction BLL");
// Dans ce cas c'est la couche BLL qui gere la session et la
// transaction
id = (PK) s.save(o);
} else {
//LOGGER.debug("le DAO initialise sa propre transaction");
Transaction tx = null;
try {
// On démarre une transaction localement
tx = s.beginTransaction();
id = (PK) s.save(o);
// On valide la transaction
tx.commit();
} catch (RuntimeException ex) {
// En cas de problèmes en annule la transaction
handleDaoOpError(tx, ex);
} finally {
// Fermer la session s'elle est encore ouverte
closeSession(s);
}
}
//LOGGER.debug("fin d'appel de la méthode save avec succés ");
return id;
}
public void update(T o) {
//LOGGER.debug("appel de la méthode save");
// On obtient la session en cours
Session s = getSession();
// Si la couche BLL a démarré une transaction
if (anActiveTransactionExists(s)) {
// Dans ce cas c'est la couche BLL qui gere la session et la
// transaction
s.update(o);
} else {
//LOGGER.debug("le DAO initialise sa propre transaction");
Transaction tx = null;
try {
// On démarre une transaction localement
tx = s.beginTransaction();
s.update(o);
// On valide la transaction
tx.commit();
} catch (RuntimeException ex) {
// En cas de problèmes en annule la transaction
handleDaoOpError(tx, ex);
} finally {
// Fermer la session s'elle est encore ouverte
closeSession(s);
}
}
//LOGGER.debug("fin d'appel de la méthode save avec succés ");
}
public List<T> getAll() throws EntityNotFoundException {
//LOGGER.debug("appel de la méthode save");
// On obtient la session en cours
Session s = getSession();
List<T> list = null;
// Si la couche BLL a démarré une transaction
if (anActiveTransactionExists(s)) {
// Dans ce cas c'est la couche BLL qui gere la session et la
// transaction
list = getSession().createCriteria(boClass).list();
} else {
//LOGGER.debug("le DAO initialise sa propre transaction");
Transaction tx = null;
try {
// On démarre une transaction localement
tx = s.beginTransaction();
list = getSession().createCriteria(boClass).list();
tx.commit();
} catch (RuntimeException ex) {
// En cas de problèmes en annule la transaction
handleDaoOpError(tx, ex);
} finally {
// Fermer la session s'elle est encore ouverte
closeSession(s);
}
}
if (list == null || list.size() == 0)
throw new EntityNotFoundException();
return list;
}
public List<T> getEntityByColValue(String pColName, String pColVal, String pClassName)
throws EntityNotFoundException {
//LOGGER.debug("appel de la méthode getByColValue");
// On obtient la session en cours
Session s = getSession();
List<T> list = new ArrayList<T>();
// Si la couche BLL a démarré une transaction
if (anActiveTransactionExists(s)) {
// Dans ce cas c'est la couche BLL qui gere la session et la
// transaction
Query q = s.createQuery("from " + pClassName + " where " + pColName + "=?");
q.setParameter(0, pColVal);
list = q.list();
} else {
//LOGGER.debug("le DAO initialise sa propre transaction");
Transaction tx = null;
try {
// On démarre une transaction localement
tx = s.beginTransaction();
Query q = s.createQuery("from " + pClassName + " where " + pColName + "=?");
q.setParameter(0, pColVal);
list = q.list();
tx.commit();
} catch (RuntimeException ex) {
// En cas de problèmes en annule la transaction
handleDaoOpError(tx, ex);
} finally {
// Fermer la session s'elle est encore ouverte
closeSession(s);
}
}
return list;
}
public void delete(PK pId) throws EntityNotFoundException {
//LOGGER.debug("appel de la méthode delete");
T obj = (T) findById(pId);
// On obtient la session en cours
Session s = getSession();
PK id = null;
// Si la couche BLL a démarré une transaction
if (anActiveTransactionExists(s)) {
// Dans ce cas c'est la couche BLL qui gere la session et la
// transaction
s.delete(obj);
} else {
//LOGGER.debug("le DAO initialise sa propre transaction");
Transaction tx = null;
try {
// On démarre une transaction localement
tx = s.beginTransaction();
s.delete(obj);
// On valide la transaction
tx.commit();
} catch (RuntimeException ex) {
// En cas de problèmes en annule la transaction
handleDaoOpError(tx, ex);
} finally {
// Fermer la session s'elle est encore ouverte
closeSession(s);
}
}
//LOGGER.debug("fin d'appel de la méthode delete avec succés ");
}
public List<T> getAllDistinct() throws EntityNotFoundException {
Collection<T> result = new LinkedHashSet<T>(getAll());
return new ArrayList<T>(result);
}
public T findById(PK pId) throws EntityNotFoundException {
//LOGGER.debug("appel de la méthode findById");
// On obtient la session en cours
Session s = getSession();
T obj = null;
// Si la couche BLL a démarré une transaction
if (anActiveTransactionExists(s)) {
// Dans ce cas c'est la couche BLL qui gere la session et la
// transaction
obj = (T) s.get(boClass, pId);
} else {
//LOGGER.debug("le DAO initialise sa propre transaction");
Transaction tx = null;
try {
// On démarre une transaction localement
tx = s.beginTransaction();
obj = (T) s.get(boClass, pId);
tx.commit();
} catch (RuntimeException ex) {
// En cas de problèmes en annule la transaction
handleDaoOpError(tx, ex);
} finally {
// Fermer la session s'elle est encore ouverte
closeSession(s);
}
}
if (obj == null)
throw new EntityNotFoundException();
return obj;
}
public boolean exists(PK pId) {
try {
findById(pId);
} catch (EntityNotFoundException ex) {
return false;
}
return true;
}
}
|
Comprehensive Methodology for the Evaluation of High-Resolution WRF Multiphysics Precipitation Simulations for Small, Topographically Complex Domains A stepwise evaluation method and a comprehensive scoring approach are proposed and applied to select a model setup and physics parameterizations of the Weather Research and Forecasting (WRF) model for high-resolution precipitation simulations. The ERA5 reanalysis data were dynamically downscaled to 1-km resolution for the topographically complex domain of the eastern Mediterranean island of Cyprus. The performance of the simulations was examined for three domain configurations, two model initialization frequencies and 18 combinations of atmospheric physics parameterizations (members). Two continuous scores, i.e., Bias and Mean Absolute Error (MAE) and two categorical scores, i.e., the Pierce Skill Score (PSS) and a new Extreme Event Score (EES) were used for the evaluation. The EES combines hits and frequency bias and it was compared with other commonly used verification scores. A composite scaled score (CSS) was used to identify the five best performing members. |
Natural supramolecular protein assemblies Supramolecular protein assemblies are an emerging area within the chemical sciences, which combine the topological structures of the field of supramolecular chemistry and the state-of-the-art chemical biology approaches to unravel the formation and function of protein assemblies. Recent chemical and biological studies on natural multimeric protein structures, including fibers, rings, tubes, catenanes, knots, and cages, have shown that the quaternary structures of proteins are a prerequisite for their highly specific biological functions. In this review, we illustrate that a striking structural diversity of protein assemblies is present in nature. Furthermore, we describe structurefunction relationship studies for selected classes of protein architectures, and we highlight the techniques that enable the characterisation of supramolecular protein structures. Introduction Proteins -biologically functional molecules -are involved in all fundamental processes in life. From transcription and translation to catalysis, metabolism, transport and structural integrity, proteins have evolved to be part of the sophisticated and highly efficient molecular machinery that controls the function of the cell. 1 Nature builds proteins by a bottom-up approach, in which the primary sequence of amino acids largely determines the proteins' tertiary three dimensional structure. Most of the characterised proteins, however, are organised in higher hierarchical quaternary structures, either by forming homo-oligomeric assemblies (i.e. proteins made by identical polypeptide chains) or hetero-oligomeric assemblies (i.e. proteins made by different polypeptide chains). 2 Protein homo-oligomerisation exhibits various advantages over the basic monomeric form of proteins, including functional control, allosteric regulation, higher-order complexity, and stability. Hetero-oligomerisation of these biomolecules has the advantage that the distinct functions of individual monomeric forms can be linked and also enables the formation of properly folded well-defined assemblies that would be hardly accessible when all protein subunits are covalently fused. A well-studied example of hetero-oligomeric proteins is hemoglobin, an a 2 b 2 tetrameric assembly that contains four haem prosthetic groups required for binding of molecular oxygen, which efficiently transports oxygen from lungs into tissues via the allosteric regulation mechanism. How can one visualise the architectures of protein assemblies? Over the years, many low-and high-resolution biophysical and bioanalytical techniques for the determination of protein structures have been developed. 2,3 The majority of the structures has been solved by X-ray crystallography, nuclear magnetic resonance (NMR) spectroscopy, and electron microscopy (EM). A recent examination of three dimensional protein structures showed that out of about twenty thousand protein structures that have been determined by X-ray crystallography, the majority was found to exist in monomeric forms (40%) and as homo-oligomeric because they are typically maintained via several weak noncovalent interactions, sometimes admixed with covalent bonding, as commonly found in small-molecule supramolecular structures. The purpose of this paper is to review supramolecular protein assemblies, and we will focus only on assemblies with topologies that have also been found in the fields of small-molecule supramolecular chemistry and macromolecular chemistry in the past decades, i.e. fibers, helices, tubes, rings, catenanes, knots, and cages. 4 In this tutorial review, we aim to illustrate the diversity of naturally-occurring supramolecular protein assemblies and describe recent chemical-biological studies that probe the structure-function relationship in the protein world. In the following sections, we will systematically describe each type of natural supramolecular protein assembly, in the order of increasing complexity. A few exemplary and concise reviews that focus on the design aspects of supramolecular protein assemblies, which are not discussed in this article, are highly recommended for readers with a greater interest in the area of supramolecular protein structures. 5,6 2. Linear proteins Elongated protein structures are prevalent in all forms of life. Several notable examples, including elastin and silk, have not been included in this review due to the restriction of space. Nonetheless, various more prominent structures, such as collagen, actin and amyloids will be briefly addressed in this section. Collagen Collagen is the most abundant protein in mammals, and its linear triple helical structure has been well-studied. 7 It is the main structural protein in animal connective tissue and a wide variety of collagen types have been characterised. The main identifying feature of collagen is its coiled structure in which three left-handed polyproline II-type (PPII) polypeptide strands form a right-handed triple helix (Fig. 1A). The tight coil structure requires each third amino acid to be a glycine residue, with the first and second residue being any residue, although proline and hydroxyproline are the most common ones. This variability in amino acid sequence results in the possibility for collagen to form both homomeric (i.e. three chemically equal polypeptide chains) and heteromeric triple helices (i.e. individual polypeptide chains in the triple helix are chemically different), with the latter being more prevalent. The collagen structure is stabilised via one single hydrogen bond per amino acid triplet, namely between glycine's amide NH and a neighbouring strands' backbone carbonyl group. Proline and hydroxyproline residues, located on the first and second position of the triplet, enable preorganisation of each individual strand in its PPII conformation, thus resulting in a thermodynamically more favourable collagen triple helix folding. The hydroxyproline residues on the second position are of major influence in collagen stability as the 4R-hydroxyl group imposes a C 4 -exo conformation on the pyrrolidine ring via the gauche effect, while prolines on the first position of the triplet usually display a C 4 -endo conformation, which is also believed to help stabilising the triple helical structure. 7 The collagen triple helix that is initially formed in vivo is referred to as procollagen. N-and C-proteinases cleave the procollagen at both termini in order to form tropocollagen (TC), which is flanked by short non-helical telopeptides. These TCs can self-assemble into higher-order structure, which has an even further stabilising effect on the triple helices. The telopeptides undergo lysyl oxidasecatalysed cross-linking and become covalently connected during fibril assembly, both within and between the microfibrils. This structure endows collagen with the unusual stability that is required for its function in tissues, such as cartilage or tendon. Actin Actin is one of the most abundant proteins in eukaryotic cells, comprising an estimated 15% of the total protein content in muscle cells. It is involved in a wide variety of cellular functions, such as maintaining the cytoskeleton or muscle contraction, where actin is part of the myofibril structure. Actin can be found as two distinct forms: monomeric globular (G-actin) and as a homopolymeric linear filament (F-actin). G-and F-actin structures are, although related, not identical. The crystal structure of G-actin has been well-defined; it is a 55 55 35 structure, which is folded into two a/b-domains. 8 Crystallisation of the filamentous structure has not been possible, therefore, the F-actin structure has been elucidated using X-ray diffraction data from concentrated F-actin solutions, and electron microscopy (EM), although it should be noted that these structures are still slightly speculative. These studies have shown F-actin to be a single left handed helical structure consisting of approximately 13 monomeric units per 6 turns (36 nm, Fig. 1B), with an extensive amount of longitudinal interactions mainly electrostatic and hydrophobic in nature, where several loop structures form intermolecular interactions. Lateral interactions are not as pronounced, however, and are governed by a structure dubbed a ''hydrophobic'' plug and a variety of hydrogen bonds. Amyloids Another interesting and biologically relevant structure is the amyloid fibril. 9 Although opposed to the other structures, which have been discussed, these fibrils are detrimental, rather than useful. Amyloid fibrils are rigid, highly structured protein polymers, which have been associated with a variety of diseases, including Alzheimer's and Huntington's disease. No apparent similarities in primary or higher order structures have yet been identified for amyloid associated proteins and it is postulated that any protein may be susceptible to amyloid formation. Even though a variety of proteins can form amyloid fibrils, the fibril structures are remarkably similar, generally being unbranched filaments several nanometers in diameter and up to micrometers in length. The fibrils are highly ordered and tightly packed structures, which are mainly composed of cross-b-sheets (Fig. 1C). It is the peptide backbone, which enables the formation of the rigid b-sheet structures that encompass the majority of the fibrils via inter-backbone hydrogen bonds. The space between the opposing b-sheet is mainly determined by the side chains of the, now non-native, proteins that make up the amyloid fibril. An interesting feature regarding amyloids is that increasing evidence points towards the notion that the amyloid state of a protein might actually be thermodynamically more stable than its natively folded counterpart. 9 This stability increases with protein concentration and it may be that the reason why proteins tend not to form amyloid structures under physiological conditions is due to kinetic barriers, which prevent amyloid formation. When this kinetic barrier is breached, however, toxic, selfpropagating oligomeric assemblies can form, thus leading to protein metastasis and disease progression. Ring proteins The function of numerous ring-shaped proteins (or toroidal proteins) is highly connected with manipulations on the doublestranded helical structure of DNA. Most notably, ring proteins are involved in fundamental biochemical processes in living organisms, including the enhancement of the processivity of DNA polymerases, unwinding of DNA to generate single stranded DNA, homologous DNA recombination, unwinding of supercoiled DNA and DNA transport (Fig. 2). 10 Although many ring proteins have an unrelated primary sequence of amino acids and have highly distinguished biological functions, the widespread appearance of this well-defined quaternary structure highlights its intrinsic ability to efficiently encode and repair the genetic material of prokaryotes, archaea and eukaryotes required for the organisms' viability. In this section, we will describe the role of several ringshaped DNA-binding proteins, with a focus on protein clamps, which are involved in bioprocessive catalysis, and thus resemble the very popular and extensively studied rotaxane structures in the field of supramolecular chemistry. DNA clamps Reproduction is essential to the survival of an organism and in order to effectuate this a large series of complex chemical events are needed of which the faithful copying of its DNA is one of the most important steps. The enzymes responsible for this copying process are the DNA polymerases, which by themselves are distributive enzymes, 11 meaning that they synthesise only a limited number of nucleotides of complementary DNA before they dissociate from the DNA template. In order to allow to this process to proceed for longer periods of time, i.e. to achieve multiple rounds of catalysis, the polymerase enzyme has to remain in contact with the DNA strand. During evolution nature has solved this problem by linking the polymerase to a ring-shaped protein, called clamp, which encircles the DNA template and freely slides along it. In this way the enzyme remains attached and the replication process continues to go on making it processive. The first crystal structure of such a ring-shaped protein, the sliding b clamp of Escherichia coli (E. coli) was reported in 1992. 12 It consists of two semi-circular protein subunits, which are aligned head-totail forming a ring with an inner diameter large enough to hold a duplex DNA chain ( Fig. 2A). Each subunit consists of three independent domains connected by long loops, providing the clamp with a pseudo-sixfold symmetry. The overall charge of the clamp is negative, but its inner surface, which is lined by a-helices, has a positive charge and interacts with the DNA chain via water molecules. The outer surface is covered by b-sheets. Association of the polymerase with the clamp considerably enhances the replication speed from approximately 20 nucleotides per second to 750 nucleotides per second, while the processivity increases from some 10 base pairs to more than 50 000 base pairs. The fact that the ring-shaped clamp protein is essential for DNA replication and hence for the survival of the organism follows from the fact that its circular architecture and function have been retained during evolution: it is found in various species ranging from bacteria and yeasts to Homo sapiens (H. sapiens). In eukaryotes the functional equivalent of the b-clamp is the proliferating cell nuclear antigen PCNA (Fig. 2B). Unlike the b-clamp, this circular protein is a homotrimer of head to tail aligned monomers each containing two domains, overall forming a six-domain ring. 13 The same circular trimeric structure is found in other species, such as the archaeal Pyrococcus furiosus and Sulfolobus solfataricus. Despite this similarity in circular shape there is little similarity (o10%) in the sequence of the amino acids of the constituting proteins. Apparently, these different sequences can lead to folding patterns that result in the same overall architecture. The stabilities of the clamps are regulated by intermolecular interactions between the subunit interfaces. Despite the fact that protein rings are under spring tension due to the bending of the subunits, they remain in most cases stable in solution and only open in the presence of a clamp loader. 13 Detailed mechanistic studies have shown that this opening and the subsequent loading of the clamp onto the DNA chain is an ATP-driven process, which is highly conserved during evolution. In the absence of ATP clamp loaders have only a weak interaction with the clamp. On binding ATP, the clamp loader undergoes a conformational change allowing a kind of ''hydrophobic plug'' to wedge into a hydrophobic pocket of the clamp forcing it to open (Fig. 3). This process is facilitated by the release of spring tension in the clamp. Once formed the clamp loader-open clamp complex recognises a special site (PT junction) on DNA, while adopting a conformation (notched screw cap arrangement) that matches the helical geometry of the DNA duplex. On hydrolysis of the bound ATP the clamp loader loses its affinity for the clamp-DNA complex and is ejected, after which the electrostatic interactions between the positively-charged inner surface of the clamp and the negativelycharged DNA lead to closure of the open sliding clamp around DNA. The replicative polymerase enzyme subsequently associates with this complex and the process of DNA replication starts. Helicases Interestingly, nature also makes use of ring-shaped protein architectures for separating the double helix of DNA. Hexameric helicases are highly conserved proteins that have been found in bacteriophages, viruses, bacteria and eukaryotes, and are involved in DNA replication, repair and recombination. 10 Bacteriophage T7 helicase, for example, has an inner diameter of B20 and encircles the single stranded DNA on the 5 0 strand during unwinding the DNA in the 5 0 -3 0 direction (Fig. 2C). Helicases also take part in replication of prokaryotes. Replication is initiated at a site on DNA called ori (origin of replication) and involves a series of proteins, i.e. an initiator protein, a helicase, a helicase loader, and a primase. The latter protein synthesises the short RNA fragment (primer) that serves as the starting point for DNA synthesis. The crystal structure of the helicase of Bacillus stearothermophilus in complex with the loader protein of Bacillus subtilis, and the helicase-binding domain of Bacillus stearothermophilus has been solved recently (Fig. 4). 14 The helicase has a ringshaped hexameric structure and interacts with the other proteins with a stoichiometry of one helicase ring binding to three loader protein dimers and to three helicase binding domains of the primase. The overall architecture is that of a three-layered ring system, displaying a height of 130, an outer diameter of 126, and an inner diameter of 50-55, which is large enough to allow a double-stranded or single-stranded DNA chain to pass. The study reveals that the helicase ring is not completely planar, but has a helical spring lockwasher structure, which is important for the translocation process. It, furthermore, suggests that the helicase undergoes a transformation from an open ring state to an openspiral state and finally to a closed spiral state during the process of loading to the ori-site on DNA. This eventually results in the dissociation of the loader protein ring from the complex and the start of the replication cycle. Nucleases DNA damage occurs frequently in cells and is a phenomenon that must be carefully managed in order to maintain the integrity of the genome. Nucleases are enzymes that are involved in DNA cleavage and as such participate in the processes that are used by the cell to repair its nucleic acid damages. Many of the nucleases have a ring-shaped structure. An example is the bacteriophage l exonuclease, which binds to double-stranded DNA and processively digests the 5 0 -strand into mononucleotides. 15 This nuclease activity is a metal-dependent process with a strong preference for Mg 2+ as the metal ion. The X-ray structure of the l-exonuclease revealed that this enzyme is a toroidal-shaped homotrimer possessing a central funnel-shaped channel for tracking along the DNA chain (Fig. 2D). This channel is at one side wide enough to allow double-stranded DNA to enter, but narrows at the other end, such that only single-stranded DNA can pass. Mechanistic studies have revealed that the cleavage of DNA proceeds via an S N 2 mechanism involving one or more metalbound water molecules that act as nucleophiles for the splitting of the phosphate backbone and as proton donors for the leaving groups. To date, several other ring proteins, which are associated with DNA and RNA, have been characterised. 10 For example, trp RNA-binding attenuation protein (TRAP) from Bacillus subtilis binds single stranded RNA and is involved in termination of transcription of the trp operon. Structural analyses revealed that TRAP forms an undecameric ring structure with a 25 diameter of the central hole (Fig. 2E). Non-denaturing mass spectrometric studies on TRAP showed that the 11-mer ring assembly also persists in the gas phase, in the absence of bulk water. 16 These studies demonstrated that the ring structure becomes more stable in the presence of tryptophan, and that binding of RNA substrate additionally stabilises TRAP. Another example of ring proteins is human RAD52, which is involved in homologous recombination of DNA; the full length protein forms a heptameric assembly, whereas the catalytically active N-terminal half exists in the undecameric form (Fig. 2F). Tubular proteins The simplest organisms such as viruses make use of tubular structures for a variety of purposes, including host infection and storage of their genetic material. The more complex eukaryotes use these structures to create cellular pores through which molecules can be actively shuttled in and out of a cell. In order to facilitate such a variety of functions, a strikingly large number of tubular structures has evolved. Here we take a closer look at a selection of such structures found in various organisms. For the purpose of this review, we define tubes as naturally-occurring elongated structures with a ''hollow'' core. Tobacco mosaic virus One of the best studied tubular protein structures is the tobacco mosaic virus (TMV) capsid. 17 In the late 19th century, the TMV virus was the first virus to be discovered by Beijerinck, when an infectious agent was identified by him as the cause for disease in tobacco plants. Its general structure had already been postulated as early as the 1950's by Rosalind Franklin, and consists of a helical coat structure encapsulating a single stranded RNA molecule. The TMV helical coat structure is composed of 2130 identical coat particles (CP), which self-assemble into a righthanded helix; 300 nm in length, 18 nm in diameter with a 4 nm inner channel (Fig. 5A). It is assumed that assembly starts when a two-layered disk (named 20S disk after its sedimentation coefficient) comprised of two rings of 17 CPs each recognises an RNA sequence. This recognition sequence facilitates a hydrogen-bonding pattern between the RNA bases and the RNA base binding sites of the 20S aggregate causing the RNA loop to fold. The RNA loop is then pulled through the TMV coat particle as additional short helix aggregates stack on top of the nucleation disc. TMV has a remarkably robust structure, displaying stability at temperatures as high as 90 1C, remaining stable in a pH range of 3.0-9.0 and it shows resilience to organic solvents such as acetone, ethanol, and water rich THF and DMSO. Interestingly, the TMV particles are also able to aggregate into larger units, such as fibers, which can be induced in vitro by increasing the particle and/or ion concentration. a-Hemolysin Another well-known tube-like protein is a-hemolysin, a homooligomeric heptameric protein complex secreted by Staphylococcus aureus. It forms a mushroom-shaped structure composed of a cap, rim and stem-like architecture (Fig. 5B). 18 The structure in total is 100 in length, with the stem being 52 and the cap 70 in height. Its width spans 100 at its maximum, whereas the solvent filled channel, which runs in the longitudinal direction of the pore has a diameter of 14-46 and the stem consisting of 14 b-strands, which possess a b-barrel structure, is 26 in diameter. The stem domain of a-hemolysin forms the transmembrane channel, whereas the cap and portions of the rim domain protrude from the phospholipid bilayer. The inside of the stem is composed of charged residues, alternated with several rings of hydrophobic residues. Its outside, on the other hand, is lined with hydrophobic amino acid residues in order to facilitate interactions with the phospholipid bilayer. Additionally, regions of the stem and rim domains are believed to interact with the head groups of the phospholipids, allowing for the binding of a-hemolysin to cell membranes. Recently, this protein complex has received a particular interest because of application as nanopore for DNA sequencing. Anthrax protective antigen pore One of the more aesthetic structures presented in this review is the anthrax protective antigen pore (PA pore); indeed, its structure has been compared to a flower (Fig. 5C). 19 Within this analogy its individual monomers consist of 4 domains representing the flower's corolla (domain 1, 3 and 4), calyx and stem (domain 2). The PA pore is a homo-heptamer consisting of 63 kDa subunits, which assemble into a tubular structure, 180 in height, 160 in width at its widest section and 27 diameter wide b-barrel with a predominantly hydrophobic outer surface. Upon endocytosis, the pore functions as a translocation channel through which the toxic lethal factor and edema factor enzymes can be unfolded, translocated and refolded, enabling the efficient delivery of the toxins into the cytosol. This translocation is mediated via the predominantly hydrophilic and negatively-charged inner surface. Although the opening of the pore is 30 in width, the f-clamp located below this opening is only 6 in diameter, and may only be able to accommodate the translocation of unfolded primary protein structures. Below the f-clamp lie an enlarged chamber and the b-barrel tube, which are large enough to hold secondary protein structures and may facilitate refolding of the translocated toxins. PhiX174 bacteriophage tail A more recently discovered protein tube is the PhiX174 bacteriophage tail structure. Viruses have developed a variety of mechanisms through which they can infect a host. One of these mechanisms is the formation of tube-like structures capable of penetrating the hosts' cell membrane allowing the genetic material of the virus to be injected into the host. The a-helical barrel structure allows the bacteriophage to infect cells using the PhiX174 DNA pilot protein H in order to form a DNA translocating channel. 20 The protein H has been shown to exist of 10 a-helical structures, which form a 170 long and 48 wide decameric coiled-coil structure (Fig. 5D). Because each monomer is kinked at residues Tyr193-Ala194-Gln195, the assembly can be divided into domains A and B, each with a specific twist and diameter. Domain A has a 22 diameter and a slight right-handed twist, whereas domain B has an inner diameter of 24 and a less steep left-handed twist. Due to the spatial organisation of the protein H coiled-coil tube, each monomer interacts with its two neighbouring monomers mainly via hydrogen bonds. The inside of the tube is predominantly negatively-charged, but also contains several glutamine, asparagine and arginine residues, thus presumably facilitating the transport of DNA across the bacterial membrane during infection. Hcp1 The Hcp1 protein that has been found in Pseudomonas aeruginosa is a circular homo-hexamer postulated to be involved in the bacterial type VI secretion system. 21 Its individual subunits are only 17.4 kDa in mass and assemble into a ring structure of 90 44 in dimension with a central hole of 40 in size. It was observed that these structures were able to further selfassemble into elongated tubular structures. This structure was artificially stabilised by introducing cysteine residues at residues Gly90 and Arg157, allowing for covalent linking of the individual ring structures (Fig. 5E). Optimisation of assembly conditions allowed for the generation of reasonably long (up to 0.1 mm) nanotubes which can be capped on both ends. The Hcp1 protein tube does illustrate that subtle mutations may lead to new protein topologies and suggest that these structures have interesting potential in the field of protein engineering. Such structures could serve as carriers for small molecules and may have an ability to be used as biologically inspired drug delivery systems. Catenane proteins Catenanes, one of the most widely studied mechanically-interlocked molecular topologies in chemistry, have been established as intriguing biomolecular architectures in nature since their appearance in the literature in 1960s. Pioneering work by Vinograd and coworkers demonstrated that catenated DNA structures exist inside living cells. On the other hand, only few examples of proteins that occur in interlocked catenane form in prokaryotes, archaea and eukaryotes have been reported in the past twenty years, most of them being discovered unexpectedly. The majority of characterised catenane proteins can be classified as catenanes. Two catenane protein subfamilies include the covalently maintained catenanes (i.e. there is a covalent linkage between monomers within each of the individual rings) and the non-covalently maintained catenanes (i.e. individual rings are assembled solely via non-covalent bonds). Herein, we describe examples of both types. Covalent catenanes 5.1.1 Bacteriophage HK97 capsid. The first direct evidence for the occurrence of protein catenation in nature appeared in 2000. High-resolution crystallographic analyses on the bacteriophage HK97 capsid illustrated that the latter possesses a catenane chainmail structure, formed from 60 hexameric and 12 pentameric rings; the hexamers are almost planar, whereas the pentamers are slightly concave, thus allowing the assembly to form a ball-like cage architecture (Fig. 6A). 22 The HK97 capsid catenane is maintained by 420 covalent isopeptide bonds between Lys169 and Asn356 side chains of the two neighbouring subunits. The Lys169-Asn356 isopeptide bond formation is autocatalytic and assisted by Glu363 that is located at the third subunit. Mutation studies showed that a substitution of Lys169 by Tyr resulted in the formation of a properly assembled capsid that lacks the cross-linkage. Similarly, the Glu363Ala variant also assembled without the cross-linkage and displayed a normal cage curvature. Overall, the lack of cross-linkage between Lys169 and Asn356 in the HK97 capsid still enables the formation of the so-called Head I cage, which cannot undergo the final step of maturation into the active Head II assembly. The cross-linked Head II assembly possesses an increased stability relative to the Head I counterpart against denaturation in the presence of guanidinium chloride, thus demonstrating that the natural chainmail structure exhibits an advantage over the highly similar structure that lacks the presence of the isopeptide bond. 5.1.2 Citric synthase and lysyl oxidase. Citric synthase from a thermophile Pyrobaculum aerophilum (PaCS) forms a catenane via the formation of two intramolecular disulfide bonds between Cys19 and Cys394 (Fig. 6B). 23 As in the case of HK97 capsid, the covalent connection between two monomers in a dimeric structure increased the stability of PaCS. Absence of the covalent bond within monomers in a double variant Cys19Ser/Cys394Ser caused a substantial decrease in melting temperature (by 10.5 1C) relative to the wild-type PaCS. Enzyme activity studies, furthermore, showed that both proteins exhibit similar reaction rates for the formation of citric acid at temperatures up to 90 1C, suggesting that the disulfide linkage is not solely responsible for the increased stability of the wild-type PaCS, but that other types of interactions provide additional stabilisation of the protein. Determination of the crystal structure of lysyl oxidase derived from Pichia pastoris yeast also revealed the presence of the catenane topology. Similarly to PaCS, the dimeric protein structure is maintained via two disulfide bridges between Cys45 and Cys756 within each of the two polypeptide chains (Fig. 6C). The functional advantage of the catenated lysyl oxidase remains to be established, but it is likely that it exhibits an increased thermal stability when compared to a putative noncross-linked protein. 5.2 Non-covalent catenanes 5.2.1 RecR. The discovery and characterisation of catenane proteins that are stabilised solely by non-covalent interactions has only recently been reported. RecR, an enzyme involved in the homologous recombinational DNA repair in prokaryotes, was the first characterised non-covalent catenane protein. 24 Dynamic light scattering analyses suggested that RecR exists as a tetrameric ring at low concentrations in solution, whereas at higher concentrations it forms a mechanically-interlocked octameric catenane architecture with both tetrameric rings intertwined through the central hole. In agreement with solution data that the catenated structure only persists at high concentration of protein, the determined RecR crystal structure confirmed the existence of the catenane form with dimensions of 120 80 60 (Fig. 7A). The central cavity of the tetrameric ring assembly has a diameter of 30-35, which is a similar dimension as observed in the DNA-binding clamps (discussed above). Although it is difficult to confirm the plausible function of the catenane/ring forms of RecR, it has been suggested based on the ability of both assemblies to undergo interconversion (i.e. to be able to open and close in solution) that the ring form might act as a structure-specific, nonsliding DNA clamp. The observed, presumably functionally-inactive, catenane form at high concentration of RecR might possess a regulatory role by controlling the DNA repair. 5.2.2 Peroxiredoxin III. The wild-type mitochondrial peroxiredoxin III (Prx III, also known as SP-22) from bovine primarily forms a dodecameric ring structure with an external diameter of 150 and a central hole with a diameter of 70. Its Cys168Ser SP-22 variant, interestingly, appears as a mixture of the ring and double-ring catenane assemblies both in solution and in the crystal state (Fig. 7B). 25 Both dodecameric rings in the mechanicallyinterlocked form are inclined at the angle of 551. Structural analyses of the catenane assembly suggested that the dimeric subunits in the individual dodecameric rings are stabilised predominantly via hydrophobic interactions (Leu41, Phe43, Phe45, Val73, Phe77, Leu103, Leu120), whereas the interactions between the two interlocked rings in the catenane form are primarily electrostatic in nature (hydrogen bonding Lys88-Thr104, hydrogen bonding Lys12-Tyr10, salt-bridge Glu67-Arg109). Based on the determined crystal structure, a suggested mechanism for the formation of the catenane assembly includes the initial formation of hydrophobic and polar dimer-dimer contacts, which allow a simultaneous constitution of two dodecameric rings around each other. As for many higher-order protein assemblies, the exact function of the catenane assembly of SP-22 is not clear, but it seems reasonable that the protein exists in the catenane form in the highly crowded environment present in mitochondria, which supposedly contains protein concentrations in the range of 100-200 mg mL 1. 5.2.3 Class Ia ribonucleotide reductase. E. coli class Ia ribonucleotide reductase (RNR), the enzyme that catalyses the conversion of ribonucleotides into deoxyribonucleotides that are subsequently used for the synthesis of DNA, was observed to exist in the (a 4 b 4 ) 2 catenane form in the presence of the inhibitor dATP and the crystallisation precipitant in solution, as well as in the crystal state (Fig. 7C). 26 Out of three postulated mechanisms for the formation of catenated RNR, a combination of SAXS and EM data suggested that the most likely scenario involves the opening of the a 4 b 4 ring, which embraces the second a 4 b 4 ring, and finally recloses. This hypothesis was supported by results showing that the a 4 b 4 ring assembly was formed predominantly, with no (a 4 b 4 ) 2 catenane structure observed, in the absence or below 10% of the precipitant (a PEG-based mixture), and that an increased amount of the precipitating agent gradually afforded the formation of the (a 4 b 4 ) 2 catenane at the cost of the a 4 b 4 ring form. It is possible that the enzymatically inactive (a 4 b 4 ) 2 catenane assembly provides an important regulatory function inside cells where the level of dATP inhibitor is high. Another step in the disassembly pathway of inactive (a 4 b 4 ) 2 catenane might proceed through an inactive a 4 b 4 ring and then to the enzymatically active a 2 b 2 dimeric assembly, which would directly control the amount and the rate of the formation of deoxynucleotide products. 5.2.4 CS 2 hydrolase. CS 2 hydrolase, a zinc-dependent enzyme from hyperthermophilic Acidianus A1-3 archaeon, which converts carbon disulfide into carbon dioxide and hydrogen sulfide, is a recent example of the class of catenane proteins. 27 CS 2 hydrolase exhibits a high degree of homology to b-carbonic anhydrase: two monomers associate to form a dimer with a dominated b-sheet core linked by a-helices. A combination of X-ray crystallography, analytical ultracentrifugation, SAXS, multiangle laser light scattering (MALLS), non-denaturing mass spectrometry and native PAGE gels revealed that purified CS 2 hydrolase exists, in the gas phase, in solution and in the crystal state, as a mixture of an octameric ring with a diameter of approximately 30 and a hexadecameric catenane in which two octameric oligomers are in compact perpendicular orientation (Fig. 7D). 27,28 Closer analysis of the crystal structure unveils that the basic dimeric core assembles to form an octameric ring through N-and C-terminal residues, suggesting that these interactions may be a determining factor for the stability of the catenane and the ring. Unambiguous experimental evidence for the existence of the catenane form of CS 2 hydrolase in solution has been provided recently. 28 A combination of size exclusion chromatography and non-denaturing mass spectrometry revealed that the ratio between catenane and ring forms remained unaltered in the concentration range of 0.1-10 mg mL 1, highlighting that the unique interlocked catenane structure persists in very dilute solutions in the absence of precipitating agents. Subsequently, transmission electron microscopic (TEM) studies on individual assemblies of CS 2 hydrolase provided visual evidence for the protein topologies in high resolution. TEM-based reconstructions of the catenane and ring assemblies of CS 2 hydrolase are in good agreement with their structures, as obtained by X-ray crystallography. 29 It should be noted that, in contrast to the catenane assembly observed in the highly crowded crystalline state, the TEM structures were solved at dilute protein concentrations and without potential crystal contacts between monomers/oligomers that may influence the spatial organisation of the protein assembly, thus additionally confirming the appearance of the catenane form of CS 2 hydrolase under a wide range of experimental conditions. The ability to separate and purify individually the catenane and ring forms of CS 2 hydrolase also allowed to comprehensively examine this special case of the topology-function relationship in the protein world. 30 Asymmetric flow field-flow fractionation coupled to multi-angle laser light scattering (AF4-MALLS) and native mass spectrometric studies on individual assemblies demonstrated that the catenane form is converted into the ring form at elevated temperatures and that this equilibrium is completely on the side of the ring. Compelling experimental data suggest that the ring assembly is the thermodynamically more stable form of the two assemblies, whereas the catenane is a kinetically trapped form. The catenane and ring forms of CS 2 hydrolase not only differ in their thermal stability, but also in the catalytic efficiency for the conversion of carbon disulfide into carbonyl sulfide and hydrogen sulfide. Enzyme kinetics studies showed that the hexadecameric catenane assembly is enzymatically more active than the octameric ring assembly, but that the ring form is more active than the catenane form when calculated as the enzyme efficiency per monomer. 30 It is possible that, in contrast to the fully accessible eight active sites of the ring assembly, the topologically more complex catenane assembly in which both interlocked rings are tightly packed, has partially hindered access for CS 2 substrate to some of the sixteen active sites, which in turn results in lowered k cat /K m values for the catenane relative to the ring. Knot proteins Another interesting and quite newly discovered type of protein structure is found in the class of knotted proteins. It has only recently come to light that proteins are able to display an intriguing knotted arrangement; that is, a tertiary structure which does not disentangle when the protein is pulled at both N-and C-termini at the same time. 31,32 The functional implications of knotted structures are not fully understood from both chemical and biological perspectives, but such structures do provide us with interesting models, which can be used to study the process of protein folding. Due to the complexity of identifying knots in proteins, however, traditional approaches such as inspection of structures obtained by X-ray crystallography are impractical, since all protein structures would need to be analysed manually. Especially for larger protein structures, the visual identification of knots becomes increasingly difficult; therefore, more computational approaches need to be employed in the discovery of such biomolecular structures. 33 The computational methods that have been developed can be divided into two main categories, namely mechanically and knot theorybased approaches, although neither of them is perfect. 32 In the past decade, various bioanalytical, biophysical, and bioinformatics tools have been developed in order to study knotted proteins in vitro and in silico. 31 For example, atomic force microscopy (AFM) can be employed to physically stretch and untangle isolated proteins. By employing in vitro transcriptiontranslation (IVTT) in cell-free expression systems, knotted proteins can be expressed in the absence of chaperonins, which gives the opportunity to examine the protein's intrinsic folding properties. In addition, recombinantly produced proteins can be chemically unfolded and refolded in order to investigate their folding pathways using urea-and/or pH-jump experiments. An approach has been devised in which denatured proteins are cyclised using disulfide bonds, thus allowing to determine whether proteins can exist in knotted conformations under denatured conditions. Supporting techniques, such as fluorescence spectroscopy, circular dichroism, and cofactor binding assays can be employed to determine correct (re-)folding of the knotted protein structures. 34 To date, four types of knots have been identified in various protein structures: namely trefoil knots (also referred to as 3 1 knot), figure-of-eight knots (4 1 ), a five-crossings knot (5 2 ) and a six-crossings knot (6 1 ). 31 Additionally, such knotted structures can be further subdivided into shallow and deep knot structures. Here the slightly arbitrary denotations shallow and deep refer to the number of amino acid residues that have to be cleaved from the proteins termini before the knotted structure untangles. Finally, knots display chirality, i.e. a certain type of handedness determined by the direction of each crossing of the protein backbone, although no apparent preference for knots to fold into a specific type of handedness has been identified. 3 1 knots The first protein that was identified as a knotted protein is the well-known carbonic anhydrase. 31 Carbonic anhydrase is an enzyme capable of converting carbon dioxide and water into bicarbonate and protons; this function is critical in maintaining the physiological bicarbonate buffer system in biological media. The protein is a trefoil knot, shallowly knotted at its C-terminus, where the thread is part of a b-sheet structure (Fig. 8A). Work in which atomic force microscopy (AFM) was employed in order to unfold single molecules of carbonic anhydrase has shown that when the protein is unfolded, by pulling at both termini, the knot tightens causing the protein to stretch over a shorter distance than would be expected for the unknotted linear structure. This implies that the knotted structure persists during this type of mechanical unfolding and would be supportive of the suggestion that knot structures may provide some kind of protection, increasing the protein's stability. The most studied knotted proteins, however, do not belong to the carbonic anhydrase fold family. Instead the Haemophilus influenzae YibK and E. coli YbeA proteins, which belong to the RNA methyltransferase family, have been most thoroughly investigated. 31 They were shown to possess a deep 3 1 a/b-knot structure. Despite having only 19% sequence similarity, both proteins do have common characteristics; YibK and YbeA are both homodimeric proteins of 160 and 155 residues, respectively, and both form the trefoil knot by threading their C-terminal residues. For YibK the final 40 residues are threaded through a knotting loop, whereas YbeA threads its final 35 residues (Fig. 8B). Both proteins fold via folding intermediates before finally dimerising, with YibK appearing to have a more complex folding pathway with three folding intermediates, whereas YbeA has only one intermediate. Interestingly, both proteins can be fully unfolded (although they retain a knotted ''primary'' structure) with urea after which they can readily be refolded into their native states without the assistance of chaperone proteins (Fig. 9). This observation indicates that the knotted structure does not impair proper protein folding, though it has been shown that the presence of bacterial GroEL-GroES (E) Stevedore's protein knot (6 1 ) in a-haloacid dehalogenase (PDB: 3BJX). Note the corresponding colour gradient from blue (N-terminus) to, cyan, green, yellow and red (C-terminus) between the protein structure and simplified visualisation. chaperonin does accelerate the folding process significantly, at least 20-fold at the experimental conditions used for the examination of the refolding process. 36 4 1, 5 2 and 6 1 knots Ketol-acid reductoisomerase (KARI) is the most deeply knotted protein identified to date, having a 4 1 knot, which can be retained until about 240 N-terminal and 60 C-terminal amino acids have been cleaved off (Fig. 8C). 31,37 It is a protein found in plants, such as spinach and rice, where it is a part of the branched chain amino acid synthesis pathway, as it is capable of catalysing the conversion of 2-aceto-2-hydroxybutyrate into (2R,3R)-2,3-dihydroxy-3-methylvalerate or 2-acetolactate into (2R)-2,3-dihydroxy-3-isovalerate. Via this pathway, the amino acids valine, leucine and isoleucine are synthesised, and all are classified as essential amino acids for humans. Another protein, ubiquitin hydrolase UCH-L3 contains an example of a 5 2 knotted structure (Fig. 8D). 38 It is a 26 kDa hydrolase believed to be involved in the maintenance of cellular ubiquitin concentrations and has been linked to diseases ranging from breast cancer to Parkinson's disease. Despite its complexity, this protein like YibK and YbeA, is able to fold properly into its native state without the aid of chaperones. It is proposed that the knotted topology may provide the ubiquitin hydrolase with a degree of resistance against degradation by the 26S proteasome. The most complex knot identified to date is the Stevedore's protein knot (6 1 ) in a-haloacid dehalogenase (DehI). 39 DehI is an enzyme, found in a Pseudomonas putida strain, capable of catalysing the cleavage of carbon-halogen bonds of halogenated organic compounds. The knot present in this protein is relatively deep, and over 65 residues on the N-terminus and 20 residues on the C-terminus can be deleted before the knot structure is abrogated (Fig. 8E). When the knotted structure is examined more closely, it is evident that it consists of two regions with similar structures and approximately 20% sequence identity, each B130 amino acid residues in length. Individually, these two regions are unknotted, it is only when taken together that a knotted structure is formed. Given the tutorial nature of this review, we are not able to discuss all aspects of knotted proteins here. Suffice to say that the scientific literature provides valuable information on other knotted, knot-like, and knot-related structures. For instance, the cysteine knot superfamily encompasses various polypeptide toxins, such as Kalata B1, which has anti-microbial activity. In addition, knot-related slipknots, including alkaline phosphatase and thymidine kinase, are strictly speaking not knotted structures, but can become knotted when one of the termini is deleted. 32 It is envisioned that future studies of knotted protein structures undoubtedly will greatly contribute to the scientific understanding of the fundamental biological principles behind protein folding. Protein cages Compartmentalisation is crucial to life by providing spatial control to biological processes and shielding compartmentalised compounds from the environment and, in turn protecting the environment from unstable and toxic intermediates. The importance of compartmentalisation is observed on different levels of complexity, ranging from the cellular level, as well as the level of organelles, down to the molecular level of protein compartments. These protein compartments have been found both in bacteria and in eukaryotic cells, but also viruses make use of a protein capsule to encapsulate their genetic material. These various types of protein cages form a very interesting class of supramolecular protein complexes with inspiring properties. Many of the known naturally-occurring protein cages have a spherical shape, although some more irregular shapes have also been observed. Most of these sphere-like protein capsules have the symmetry of icosahedrons. The materials science community has demonstrated particular interest in protein cages because of their potential applicability as drug delivery vehicles. This has also resulted in engineering of naturally-occurring capsules to optimise their properties for specific applications. In this section we will discuss a number of well-studied naturally-occurring protein cages, with a focus on homomultimeric protein cages. Ferritins Ferritin protein cages are ubiquitous in almost all forms of life, and can be divided into three subcategories: classical ferritins, bacterioferritins and DNA-binding proteins from starved cells (named Dps). Ferritins are members of ferroxidase family of enzymes and can sequester iron by concentrating it in their internal cavities for storage and detoxification. These protein cages assemble from four-helix bundle monomers (four a-helices, labeled A, B, C and D). Interestingly, two distinctly different ferritin architectures with different symmetries and oligomerisation states have been characterised in different organisms, maxiferritins and mini-ferritins ( Fig. 10A and B, respectively). Classical ferritins and bacterioferritins are maxi-ferritins and are found in animals, plants and bacteria and form capsules of 24 subunits (480 kDa) and can accommodate about 4500 iron ions, whereas the Dps proteins, that are present in bacteria, are mini-ferritins, consisting of 12 subunits (240 kDa). 40 Although the DNA and protein sequences of the different ferritins vary considerably, the secondary and tertiary structures have been found to be highly conserved. Superimposing the tertiary structures of the ferritin subunits from mini-and maxi-ferritin architectures also shows high degree of similarities (Fig. 10C). However, there also are several key differences; for example, in maxi-ferritin the ferroxidase active site is in the centre of the helix-bundle of each subunit, while the catalytic sites of mini-ferritins are between two subunits. In addition, the small helix in the loop between the B and C helix is only found in mini-ferritin. Furthermore, the fifth helix (E-helix) in the maxi-ferritin is not present in mini-ferritin and is responsible for the four-fold symmetry axis which is important for the formation of the 24-mer capsule. The strong interactions between the helices in the helix-bundle motif and between subunits in the assembled ferritins result in a highly stable protein structure that can be heated up to 80 1C or exposed to denaturing conditions such as 6.0 M guanidine at neutral pH without disassembly. 41 The assembly mechanism of maxi-ferritin has been studied most thoroughly and is quite well understood, whereas the assembly of mini-ferritins is relatively poorly understood. The assembly of the apo form of maxi-ferritin from horse spleen has been shown to go through a number of concentrationdependent association steps, which occur after folding of the unstructured monomer (M 1 *). Through chemical crosslinking, intrinsic fluorescence emission spectroscopy experiments, gel permeation chromatography and ultracentrifugation it was shown that assembly of the 24-mer maxi-ferritin (M 24 ) proceeds via several intermediates involving structured monomers (M 1 ), dimers (M 2 ), trimers (M 3 ), hexamers (M 6 ) and dodecamers (M 12 ). The hexamer (M 6 ) is a transient intermediate as it could only detected in small quantities. This assembly scheme was further confirmed via reassociation studies of isolated intermediates that could be obtained upon reversible chemical dissociation with 2,3-dimethylmaleic anhydride. Altogether, these experiments resulted in the following self-assembly scheme: 41 Vaults Vaults are ribonucleoprotein cages of 13 MDa which have been found in various eukaryotic species. Depending on the type of tissue, up to several thousands of copies can be found per cell. Even though vault proteins have been a topic of investigation since their discovery in 1986, their exact function still remains unclear. It has been suggested that vaults are involved in basic cellular activities, such as transport, signal transmission and immune response, because they are highly evolutionary conserved and because they are present in many different tissue types. The vault nanocapsule consists of untranslated RNA and multiple copies of three different proteins; the major vault protein (MVP) and two minor vault proteins (VPARP and TEP1). 43 The assembly of the MVP alone is already sufficient for the formation vault-like particles. Multiple copies of the two other vault proteins and the small vault RNAs are packaged inside the MVP shell. X-ray crystallography of rat liver vault showed that the vault particles exhibit 39-fold dihedral symmetry, which means that each halfvault consists of 39 MVP monomers (Fig. 10D). Each monomer folds into 12 domains: 9 structural repeat domains, a shoulder domain, a cap-helix domain, and a cap-ring domain (Fig. 10E). Most of the interactions that are suggested to drive the assembly of subunits of one half-vault are located in the cap-helix domain. The basis of the interaction between the two half-vaults is a relatively weak anti-parallel b-sheet. Interestingly, the disruption of the assembled vault therefore results in the formation of half-vaults. 43 Recently, a unique assembly mechanism was proposed, in which clusters of cytoplasmic ribosomes (called polyribosomes) template the vault assembly by directing nascent MVP polypeptides to participate in assembly via spatially coordinated synthesis. 44 It was suggested that this assembly mechanism in which the polyribosomes act as a 3D nanoprinter might also be involved in the formation of other homomultimeric protein complexes. Clathrin cages Clathrin-coated vesicles are important vehicles for intracellular trafficking and are found in all eukaryotic cells. Clathrin cages differ from the other protein capsules discussed in this review because in this case the proteins form a lattice coat surrounding a membrane vesicle. The coat is constructed from three-armed assembly units that radiate from a central hub. Such a triskelion consists of three heavy chains (180 kDa) and three light chains (25 kDa), forming legs from a-helical zigzags (Fig. 10F). Triskelions assemble to form a lattice structure with open hexagonal and pentagonal faces. Clathrin can form structures ranging from small coats of 28 and 36 assembly units, named mini-coats and hexagonal barrels (Fig. 10G), up to extended hexagonal arrays. The size of the clathrin-coated vesicles is dictated by the cargo's diameter, and the hexagonal barrel (700 800 ) is probably the smallest polyhedron that can enclose a transport vesicle. 42 Chaperonins Chaperonins are cylindrically-shaped protein assemblies which are members of a set of protein families named molecular chaperones. Like other members of this family, they assist in correct protein folding and proteome maintenance. Chaperonins can be divided into two subtypes; group I is composed of the cylinder and a detachable cap and is found in bacteria (GroEL-GroES) as well as in endosymbiotic organelles such as mitochondria and chloroplasts (HSP60-HSP10). 45 Group II chaperonins have a built-in lid and are found in archaea and in the cytosol of eukaryotic cells. 45 Both groups of chaperonins consist of two back-to-back stacked rings. However, in group I the rings are heptameric, comprised of identical subunits of 60 kDa, whereas the two rings in group II are more complex eight or nine membered heteromultimeric assemblies. The bacterial GroEL-GroES complex (M w E 1 MDa) is the most studied and its very interesting mechanism of action has been unraveled ( Fig. 11A and B). The open internal cavities of the two rings are 5 nm in diameter and are surrounded by a hydrophobic surface. In this state the rings are ready to capture polypeptides with exposed hydrophobic surface. Then upon ATP binding, the GroEL ring recruits a GroES cap, a protein ring comprised of seven identical 10 kDa subunits. This initiates a dramatic structural transformation, in which the open cavity changes into an enclosed chamber with hydrophilic lining. Folding of the protein can occur in this protected environment, after which the chamber is re-opened to release the protein (Fig. 11C). Viruses Viruses need to be robust in order to protect their genomic cargo from changes in the environment (temperature, pH and ionic strength), but yet they need to be fragile enough to dissociate and release the cargo. Virus particles are perhaps the largest class of protein-based nanocages. In approximately half of them, the capsid is spherical, and most commonly their coat proteins are arranged with icosahedral geometry. Even though rod-like viruses, such as TMV, are also examples of protein cages, we have discussed these in the section about tubular protein assemblies. Many different families of spherical virus capsules are known, but structure-wise viruses can be organised in two different types: nucleocapsids with and without a proteoglycan layer. Virus assembly has been studied thoroughly, as understanding of the assembly mechanism could facilitate the design of antiviral agents. The cowpea chlorotic mottle virus (CCMV) was the first icosahedral virus that was assembled in vitro and has served as a model system for investigation of capsid assembly, disassembly and stability. 46 Some viruses require a nucleic acid scaffold or a scaffolding protein for correct assembling process. Interestingly, the 180-mer CCMV capsid can also be formed in the absence of its nucleic acid cargo. Assembly of this virus has been shown to start with the formation of a pentamer of dimers which subsequently associates with free dimers to form the completed capsid (Fig. 12A). In general, assembly of a population of virus capsids is characterised by concurrent reactions for nucleation, elongation and capsid completion. Furthermore, the subunit-subunit association is weak and assembly goes through numerous reversible reactions, which allow for correction of assembly mistakes by dissociation (Fig. 12B). For several viruses it has been shown that increasing of the association energy results in kinetic traps and defects in the assembly. 46 Bacterial compartments Bacterial nano-and microcompartments are reaction chambers that enclose enzymes and other proteins. These supramolecular architectures protect the bacterial cell by sequestering toxic intermediates and by shielding delicate processes from the environment. Encapsulation of enzymatic cascades can also be advantageous because it can result in increased reaction efficiencies. In this section we discuss several examples of proteinaceous bacterial compartments that do not have eukaryotic counterparts. Lumazine synthase is an enzyme that occurs in different topologies. In Bacillaceae, it forms a unique core-shell complex, consisting of an icosahedral shell of 60 lumazine synthase subunits and a core of three riboflavin synthase subunits, with a diameter of 16 nm (Fig. 13A). This supramolecular enzyme complex is involved in the catalysis of the final steps in the synthesis of riboflavin (vitamin B 2 ). This compartmentalisation is suggested to offer protection to intermediates and provide a reaction rate enhancement as a result of substrate-channelling between the active-sites, which are in optimal proximity on the internal surface of the capsid. However, this does not fully explain the structural complexity, because in many other organisms lumazine synthase exists as an empty capsule or as a pentameric or decameric complex, indicating that this remarkable organisation is not required for the function of these enzymes. Furthermore, the recombinantly expressed lumazine synthase from hyperthermophilic bacterium Aquifex aeolicus has been studied for its thermostability, and with an apparent melting temperature of 120 1C it is one of the most thermostable proteins known. 47 Encapsulins form another example of a widespread class of conserved archaeal and bacterial proteins that assembly into a nanocompartment for the encapsulation of enzymes. These homomultimeric reaction chambers arise from encapsulin monomers that were shown to assemble around proteins involved in oxidative stress. Encapsulin nanocompartments consisting of 60 and 180 monomers forming thin icosahedral shells with a diameter of 24 and 32 nm, respectively, have been reported ( Fig. 13B and C). 48 Packaging of enzymes is directed by their C-terminal peptide sequence that anchors the enzymes to the interior of the capsules. 49 Despite the very different functions of encapsulins and viral capsid proteins, there is a striking similarity in the tertiary structure of the HK97 family capsid proteins and the encapsulin monomers, which suggest a common ancestor, although the unique catenation of HK97 bacteriophage has not been observed. In addition to these well-defined bacterial nanocompartments, nature also produces larger very sophisticated proteinaceous microcompartments, which accommodate two or more sequentially acting enzymes. Carboxysomes (Fig. 13E), propanediol utilisation operon (pdu, Fig. 13F) and ethanolamine utilisation (Eut) microcompartments are well-studied examples of these proteinbased organelles that have been reviewed. 50 The Eut microcompartment, for example, sequesters the metabolic pathway for ethanolamine to prevent exposure of the bacterial cytosol to the reactive acetaldehyde intermediate. 51 Bacterial microcompartment (BMC) shells range from 100 to 150 nm and are composed of a few thousand BMC shell proteins. BMC monomers assemble into hexameric discs that form the basic building block of the shell. In carboxysomes, BMC shell proteins forming pentamers have been found that occupy the vertices of an icosahedral shell ( Fig. 13D and E). 50 In contrast to the encapsulin family, no viral counterparts are known for this class of bacterial protein cages. Conclusion and outlook Nature has built with high precision an astonishing number of versatile and functionally well-defined protein architectures. The knowledge on how multimeric protein assemblies spatially arrange and how the three dimensional structure of a protein affects its function may stimulate new research endeavours of the next generation of supramolecular chemists and chemical biologists. Reflecting on the discoveries in the past few decades, it is striking to see how many wonderful architectures nature has produced and one can envision, that with the expected significant advances in chemical and biological research in the coming decade we will discover more intrinsically novel and perhaps unimaginable architectures than currently known in the protein world. The supramolecular protein assemblies presented in this review already exhibit a high level of complexity, but there are even more complex assemblies that are built on proteins only (such as rotary motors, e.g. ATP synthase) or on protein-DNA and protein-RNA complexes (such as the nucleosome and the ribosome). In science the field of supramolecular protein assemblies has successfully passed its first vital phase, in which fundamental biochemical discoveries highlight the great natural diversity of protein structures and a phase in which the examination of the structurefunction relationship has led to a deeper insight in the mechanism of enzyme action and the properties of protein materials. It has now entered the era in which our fundamental understanding of protein assemblies will deepen, such that it can be used for the rational design of new biomolecular systems that exhibit completely novel functions and properties. The research on supramolecular protein assemblies has typically been carried out in simple buffer solutions at low protein concentrations, whereas the 'living environment' of proteins is the crowded habitat of the cell. It remains to be established whether the results and conclusions from in vitro experiments on protein assemblies reflect the formation of these assemblies in the natural cellular environment, in which concentrations are high and the medium is much more complex containing many different species that all interact with each other. 3 Moreover, it is currently not clear whether the formation of multisubunit protein assemblies typically occurs cotranslationally, although there is some evidence that this might be the case for some of them. Nature will remain to serve as a vital inspiration for the research activities in the area of supramolecular protein assemblies. |
Life expectancy in the city of Seattle has positively soared over the last couple of decades! Take a look: in the early 1990s, residents of Seattle (the blue line) could expect a shorter average lifespan than the the state as a whole (the pink line). But by 1997, Emerald City life expectancy shot ahead of the state average; and by the middle of the last decade it had even moved ahead of British Columbia. Today, if Seattle were an independent nation, its life expectancy would rank second in the world, just a month behind Japan’s.
A rapid rise in life expectancy is a big deal, since longer lives usually mean better health. Both internationally and historically, populations that live longer tend to be healthier: they suffer from fewer infectious diseases, face lower risks of injuries and catastrophic ailments, and have access to better medical care—the whole shebang.
So Seattle’s rising life expectancy is clearly good news, but it’s also something of a puzzle. What made Seattle get so healthy, so fast??
The good folks at Seattle / King County Public Health agency looked into things a bit, and came to at least one somewhat unexpected conclusion: during the mid-1990s, the surprising rise in Seattle’s life expectancy could be attributed to the phenomenally steep decline in AIDS deaths.
Finding this article interesting? Donate now to support our independent research!
Take a look at the chart to the right, showing HIV mortality among men and women in King County vs. Washington as a whole. HIV deaths skyrocketed in the years leading up to 1996: among King County males aged 25-44, HIV was the leading cause of death from 1992 to 1996, accounting for 1,286 of the 3,539 deaths to males in this age group. And 77% of those King County HIV deaths were among residents of the city of Seattle proper.
So the numbers suggest two things. First, men in Seattle were particularly hard hit by the AIDS crisis. Second, the rapid surge in Seattle life expectancy in the late 1990s—when Seattle overtook the state as a whole—can be traced first and foremost to the development of an effective treatment for HIV. Clearly, curbing the epidemic of AIDS fatalities was a huge boon to Seattle’s overall health.
So that explains the life expectancy surge in the 1990s. But it doesn’t explain health trends from 2000 to the present. Why did Seattle improve so much faster than the region as a whole? Are the trends real—or are they merely an artifact of a bad data?
Life expectancy figures can be skewed in two ways—first, by mistaken population estimates; and second, by faulty or inconsistent ways of measuring fatalities. We’ll know more about population figures when the 2010 census counts are released; perhaps the state has overestimated the increase in Seattle’s population, and hence underestimated death rates. But the folks at Public Health can’t point to any significant changes in how death statistics are compiled—so there’s no obvious data anomaly that can account for Seattle’s striking rise in life expectancy.
So perhaps, instead, Seattle’s life expectancy gains resulted from rising incomes. All else being equal, wealthier people tend to be healthier people; so maybe a decline in poverty rates, or a sharp rise in median household earnings, was at the root of Seattle’s health advantage. The numbers, however, didn’t quite bear this out. Seattle really has grown wealthier since 1990, and poverty grew a little slower in Seattle proper than in the state as a whole. But still, Seattle didn’t get that much wealthier; plus, poverty levels fluctuated quite a bit, and actually increased within city limits.
So in the end, I’ve got a decent explanation for the 1990s rise, but I still have no solid answer to why Seattle’s life expectancy figures shot up so quickly since 2000, compared with the rest of state.
Any thoughts out there in blog land?
[Data sources for this post include:
1990-2008 Population Estimates: Population Estimates for Public Health Assessment, Washington State;
Department of Health, Vista Partnership, and Krupski Consulting. January 2009.
Washington Department of Health
Special thanks to Eva Wong and Myduc Ta, Public Health-Seattle & King County, for their assistance with data and for their fascinating analysis.] |
1. Technical Field
Embodiments of the present disclosure relate to semiconductor devices and semiconductor systems including the same.
2. Related Art
Fast semiconductor memory devices with improved integration density are increasingly in demand. Synchronous memory devices operating in synchronization with external clock signals have been revealed to improve the operation speed thereof.
At first, single data rate (SDR) synchronous semiconductor memory devices have been proposed to improve the operation speed thereof. The SDR synchronous semiconductor memory devices may receive or output a single data through a single data pin in one cycle of the external clock signal in synchronization with every rising edge of the external clock signal. However, high performance devices operating at a higher speed than the SDR synchronous semiconductor memory devices have been demanded to meet the requirements of high performance semiconductor systems. Accordingly, double data rate (DDR) synchronous semiconductor memory devices have been recently proposed. DDR synchronous semiconductor memory devices may receive and output data in synchronization with every rising edge and every falling edge of an external clock signal. Thus, DDR synchronous semiconductor memory devices may operate at a speed which is at least twice higher than that of SDR synchronous semiconductor memory devices even without an increase in a frequency of the external clock signal. |
Performance and evolution of W/steel joint fabricated by selective laser melting Herein, attempts were made to develop W/steel joints via selective laser melting (SLM) for the first time. The influence of processing parameters on the joint strength was investigated. To further improve the mechanical properties, we performed post-heat treatment on the as-fabricated specimens. The results showed that large amounts of solid solution and dispersed micro-/nanoparticles appear in the interface of W/steel joints fabricated by direct SLM because of the uniqueness of non-equilibrium solidification, which is favourable for the bonding properties. However, the solid solution gradually transforms into the Fe2W(Ni, Cr) phase encapsulated with W particles during heat treatment, while the Fe7W6 phase fills the poles at the W/steel interface, which is extremely detrimental to the bonding properties. |
/**
* Performs supervised training using backpropagation on the network with the
* given input and expected output pair for error calculation.
* @param trainingInput An array containing an input value for each input node.
* @param trainingOutput An array containing an expected output value for each output node.
* @return True if the networks gives the correct output for all output nodes within MARGIN of the target.
*/
public boolean runTraining(double[] trainingInput, double[] trainingOutput){
ArrayList<ArrayList<Node>> hiddenLayers = UpdateLayers();
for(int j = 0; j < inputNodes.size(); j++){
inputNodes.get(j).setInput(trainingInput[j]);
}
boolean match = true;
for(int j = 0; j < outputNodes.size(); j++){
double result = outputNodes.get(j).forwardOperation();
double target = trainingOutput[j];
match = match && (target + MARGIN >= result && target - MARGIN <= result);
}
for(int j = 0; j < outputNodes.size(); j++){
double T = trainingOutput[j];
outputNodes.get(j).computeError(T);
}
for(int j = hiddenLayers.size()-1; j >= 0; j--){
for(Node n : hiddenLayers.get(j)) n.computeError(0);
}
for(Connection c : connections) c.updateWeight();
for(int j = 0; j < hiddenLayers.size(); j++){
for(Node n : hiddenLayers.get(j)) n.updateBias();
}
for(Node n : outputNodes) n.updateBias();
return match;
} |
BAGRAM AIRFIELD, Afghanistan (Reuters) - As the Trump administration debates war strategy in Afghanistan, this Afghan pilot in the U.S.-backed Afghan Air Force has more pressing concerns: He’s worried the Taliban may kidnap or kill his family.
Like other colleagues flying A-29 Super Tucano light attack aircraft, he told Reuters he has received death threats. One came in the form a note left on the door of his home in Kabul.
“It said: ‘If you don’t quit, we’re going to kidnap your kids and kill you’,” the pilot said in an interview, asking not to be identified due to fears for his security.
His children are aged 2, 3, 6 and 7. Three other pilots said such threats were common among Super Tucano pilots, whose skills are quickly becoming among the most sought after assets in the Afghan arsenal. Two said they wanted the Afghan government to do more to help protect their families.
Reuters could not independently verify the accounts.
The fledgling Afghan Air Force (AAF) is a bright spot in a 16-year-old war against Taliban insurgents that American commanders say is at a stalemate, and any future U.S. plans to aid the Afghan military will almost certainly involve strengthening its air power.
U.S. President Donald Trump, Vice President Mike Pence and his national security team are scheduled to discuss Afghan war strategy, and regional policy guiding it, at a meeting at Camp David on Friday.
Even if Trump approves sending thousands more American troops, recruiting, training and retaining Afghan pilots and aircraft maintenance crews is expected to remain a major priority of the U.S. effort.
The 17 Afghan Super Tucano pilots are particularly precious, taking years to train, including in the United States. After dropping their first bomb in April 2016, Afghan Super Tucano pilots are now regularly flying combat missions, something that was hard to imagine just a few years ago.
That appears to have made them targets.
In one case cited by two Super Tucano pilots, a car bomb badly injured an A-29 pilot last year. Wounded in a leg, he no longer flies, they said.
U.S. Air Force Major General James Hecker, commander of the 9th Expeditionary Task Force - Afghanistan, told Reuters the U.S. military was aware of the pilots’ accounts and has discussed the matter with the Afghan government.
“That is something we are very concerned about,” Hecker said.
Afghanistan’s defense ministry said it was also aware of the concerns. Ministry spokesman Dawlat Waziri said measures had been taken to protect the pilots but he declined to offer details, saying that disclosure could compromise their security.
“Overall, there is a plan to protect them, especially those who are directly involved in operations against the terrorists,” Waziri said.
ASYLUM REQUEST
The first female fixed-wing pilot in Afghanistan’s air force made headlines in December for requesting asylum in the United States after completing an 18-month training course there. She was certified to fly a C-208 military cargo aircraft.
In a U.S. State Department award citation, it said she and her family had received direct threats not just from the Taliban but also from some relatives, forcing her family to move house several times.
None of the A-29 pilots suggested they were seeking asylum, however, only protection for their families. They appeared proud of their training and eager to fight the Taliban.
The Super Tucanos are built in the United States under an agreement between their Brazilian manufacturer, Embraer and Sierra Nevada Corp.
The aim of the expanding U.S. assistance is to build an Afghan air force able to support counter-insurgency forces fighting in remote and forbidding terrain with air strikes, supplies and intelligence.
The AAF has about 120 aircraft in service, ranging from small propeller Cessna 208s to old Soviet-era helicopters, as well as the A-29s, MD-530s and veteran C-130 Hercules transporters.
In June, an Afghan air crew parachuted about 400 kg of supplies to an isolated border police outpost, the first time the AAF had conducted an aerial supply drop.
In coming years, the old Russian Mi-17s helicopters, which are increasingly difficult to maintain, will be replaced by American UH-60 Black Hawks.
The planned acquisitions will cost $6.78 billion over the next six years, according to the U.S. military. That does not include maintenance, which is heavily dependent on pricey foreign contractors.
The aim is to increase the effectiveness of the security forces, which advisers hope to get to a point where the Taliban are forced to negotiate a political settlement.
As the AAF has grown, however, it has faced increasing pressure from army units to step up operations. Advisers say one of the main risks it faces is overstretch. |
// reportsHandler is the hadndler for /v1/reports
func reportsHandler(sc *server.ServerContext) func(http.ResponseWriter, *http.Request) {
return http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
si, err := sc.SessionManager.ValidateSession(w, r)
if err != nil {
return
}
switch r.Method {
case "GET":
tx, err := sc.UserDB.Begin()
if util.TXErrorHandler(err, tx, w) != nil {
return
}
reports, err := tx.GetUserReports(si.UserID)
if util.TXErrorHandler(err, tx, w) != nil {
return
}
tx.Commit()
for _, r := range reports {
hideAWSSecretAccessKeys(r)
}
util.SuccessHandler(reports, w)
case "POST":
decoder := json.NewDecoder(r.Body)
report := userdb.Report{}
err = decoder.Decode(&report)
err = errors.Wrap(err, errors.CodeBadRequest, "Invalid report")
if util.ErrorHandler(err, w) != nil {
return
}
tx, err := sc.UserDB.Begin()
if util.TXErrorHandler(err, tx, w) != nil {
return
}
reportID, err := tx.CreateUserReport(si.UserID, &report)
if util.TXErrorHandler(err, tx, w) != nil {
return
}
createdReport, err := tx.GetUserReport(si.UserID, reportID)
if util.TXErrorHandler(err, tx, w) != nil {
return
}
repCtx := sc.CostDB.NewCostReportContext(createdReport.ID)
err = repCtx.CreateRetentionPolicy(createdReport.RetentionDays)
if util.TXErrorHandler(err, tx, w) != nil {
return
}
err = tx.Commit()
if util.TXErrorHandler(err, tx, w) != nil {
return
}
hideAWSSecretAccessKeys(createdReport)
util.SuccessHandler(createdReport, w)
default:
util.ErrorHandler(errors.Errorf(errors.CodeBadRequest, "Unsupported method %s", r.Method), w)
}
})
} |
<reponame>akulamartin/lumberyard
#
# All or portions of this file Copyright (c) Amazon.com, Inc. or its affiliates or
# its licensors.
#
# For complete copyright and license terms please see the LICENSE at the root of this
# distribution (the "License"). All use of this software is governed by the License,
# or, if provided, by the license below or the license accompanying this file. Do not
# remove or modify any license notices. This file is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
#
# Original file Copyright Crytek GMBH or its affiliates, used under license.
#
from waflib.Configure import conf, Logs
from waflib import Options
from lumberyard import deprecated
@deprecated("Logic replaced in load_msvc_common_settings with a combination of setting ENV values and env settings in the common.msvc.json")
def set_ltcg_threads(conf):
"""
LTCG defaults to 4 threads, and can be set as high as 8. Any setting over 8 is overridden to 1 by the linker.
Any setting over the number of available hw threads incurs some thread scheduling overhead in a multiproc system.
If the hw thread count is > 4, scale the setting up to 8. This is independent of jobs or link_jobs running, and
these threads are only active during the LTCG portion of the linker. The overhead is ~50k/thread, and execution
time doesn't scale linearly with threads. Use the undocument linkflag /Time+ for extended information on the amount
of time the linker spends in link time code generation
"""
hw_thread_count = Options.options.jobs
if hw_thread_count > 4:
ltcg_thread_count = min(hw_thread_count, 8)
conf.env['LINKFLAGS'] += [
'/CGTHREADS:{}'.format(ltcg_thread_count)
]
@conf
def load_msvc_common_settings(conf):
"""
Setup all compiler/linker flags with are shared over all targets using the microsoft compiler
!!! But not the actual compiler, since the compiler depends on the target !!!
"""
v = conf.env
# MT Tool
v['MTFLAGS'] = ['/NOLOGO']
# AR Tools
v['ARFLAGS'] += ['/NOLOGO']
v['AR_TGT_F'] = '/OUT:'
# CC/CXX Compiler
v['CC_NAME'] = v['CXX_NAME'] = 'msvc'
v['CC_SRC_F'] = v['CXX_SRC_F'] = []
v['CC_TGT_F'] = v['CXX_TGT_F'] = ['/c', '/Fo']
v['CPPPATH_ST'] = '/I%s'
v['DEFINES_ST'] = '/D%s'
v['PCH_FILE_ST'] = '/Fp%s'
v['PCH_CREATE_ST'] = '/Yc%s'
v['PCH_USE_ST'] = '/Yu%s'
v['ARCH_ST'] = ['/arch:']
# Linker
v['CCLNK_SRC_F'] = v['CXXLNK_SRC_F'] = []
v['CCLNK_TGT_F'] = v['CXXLNK_TGT_F'] = '/OUT:'
v['LIB_ST'] = '%s.lib'
v['LIBPATH_ST'] = '/LIBPATH:%s'
v['STLIB_ST'] = '%s.lib'
v['STLIBPATH_ST'] = '/LIBPATH:%s'
v['cprogram_PATTERN'] = '%s.exe'
v['cxxprogram_PATTERN'] = '%s.exe'
# shared library settings
v['CFLAGS_cshlib'] = v['CFLAGS_cxxshlib'] = []
v['CXXFLAGS_cshlib'] = v['CXXFLAGS_cxxshlib'] = []
v['LINKFLAGS_cshlib'] = ['/DLL']
v['LINKFLAGS_cxxshlib'] = ['/DLL']
v['cshlib_PATTERN'] = '%s.dll'
v['cxxshlib_PATTERN'] = '%s.dll'
# static library settings
v['CFLAGS_cstlib'] = v['CFLAGS_cxxstlib'] = []
v['CXXFLAGS_cstlib'] = v['CXXFLAGS_cxxstlib'] = []
v['LINKFLAGS_cxxstlib'] = []
v['LINKFLAGS_cxxshtib'] = []
v['cstlib_PATTERN'] = '%s.lib'
v['cxxstlib_PATTERN'] = '%s.lib'
# Compile options appended if compiler optimization is disabled
v['COMPILER_FLAGS_DisableOptimization'] = [ '/Od' ]
# Compile options appended if debug symbols are generated
# Create a external Program Data Base (PDB) for debugging symbols
#v['COMPILER_FLAGS_DebugSymbols'] = [ '/Zi' ]
v['COMPILER_FLAGS_DebugSymbols'] = [ '/Z7' ]
# Linker flags when building with debug symbols
v['LINKFLAGS_DebugSymbols'] = [ '/DEBUG', '/PDBALTPATH:%_PDB%' ]
# Store settings for show includes option
v['SHOWINCLUDES_cflags'] = ['/showIncludes']
v['SHOWINCLUDES_cxxflags'] = ['/showIncludes']
# Store settings for preprocess to file option
v['PREPROCESS_cflags'] = ['/P', '/C']
v['PREPROCESS_cxxflags'] = ['/P', '/C']
v['PREPROCESS_cc_tgt_f'] = ['/c', '/Fi']
v['PREPROCESS_cxx_tgt_f'] = ['/c', '/Fi']
# Store settings for preprocess to file option
v['DISASSEMBLY_cflags'] = ['/FAcs']
v['DISASSEMBLY_cxxflags'] = ['/FAcs']
v['DISASSEMBLY_cc_tgt_f'] = [ '/c', '/Fa']
v['DISASSEMBLY_cxx_tgt_f'] = ['/c', '/Fa']
# ASAN and ASLR
v['LINKFLAGS_ASLR'] = ['/DYNAMICBASE']
v['ASAN_cflags'] = ['/GS']
v['ASAN_cxxflags'] = ['/GS']
"""
LTCG defaults to 4 threads, and can be set as high as 8. Any setting over 8 is overridden to 1 by the linker.
Any setting over the number of available hw threads incurs some thread scheduling overhead in a multiproc system.
If the hw thread count is > 4, scale the setting up to 8. This is independent of jobs or link_jobs running, and
these threads are only active during the LTCG portion of the linker. The overhead is ~50k/thread, and execution
time doesn't scale linearly with threads. Use the undocument linkflag /Time+ for extended information on the amount
of time the linker spends in link time code generation
"""
hw_thread_count = Options.options.jobs
if hw_thread_count > 4:
v['SET_LTCG_THREADS_FLAG'] = True
ltcg_thread_count = min(hw_thread_count, 8)
v['LTCG_THREADS_COUNT'] = str(ltcg_thread_count)
else:
v['SET_LTCG_THREADS_FLAG'] = False
|
. The threshold of stereopsis was measured by static stereopter for 111 children with amblyopia aged from 4 to 13 years. They were divided into three groups: 38 cases ranging in age from 4 to 5 years, 43 cases 6 to 7 and 30 cases 8 to 13. Fifty six cases were unilateral and 55 bilateral amblyopia. The average threshold of stereopsis was 149.64 second angle for the 4-5 group, 66.69 for the 6-7 and 48.48 for the 8-13. The average threshold for unilateral amblyopia was 65.3, in which mild type was 38.6 and under moderate type was 88.84, whereas the average threshold for bilateral cases was 115.81, in which mild type was 44.09 and under moderate type was 175.44. The distribution of stereopsis for different age groups, the threshold of stereopsis influenced by different types of amblyopia and the comparison of amblyopia between bilateral and unilateral are also discussed. |
def qemu_get_base_image() -> str:
return QEMU_IMG_PREFIX + "-base" + QEMU_IMG_SUFFIX |
<reponame>JJK96/dance<gh_stars>100-1000
import * as vscode from "vscode";
// ===============================================================================================
// == CHARACTER SETS ===========================================================================
// ===============================================================================================
/**
* A list containing all blank characters.
*/
export const blankCharacters = "\r\n\t " + String.fromCharCode(
0xa0,
0x1680,
0x2000,
0x2001,
0x2002,
0x2003,
0x2004,
0x2005,
0x2006,
0x2007,
0x2008,
0x2009,
0x200a,
0x2028,
0x2029,
0x202f,
0x205f,
0x3000,
);
/**
* A character set.
*/
export const enum CharSet {
/** Whether the set should be inverted when checking for existence. */
Invert = 0b001,
/** Blank characters (whitespace), such as ' \t\n'. */
Blank = 0b010,
/** Punctuation characters, such as '.,;'. */
Punctuation = 0b100,
/** Word character (neither blank nor punctuation). */
Word = Invert | Blank | Punctuation,
/** Non-blank character (either word or punctuation). */
NonBlank = Invert | Blank,
}
/**
* Returns a string containing all the characters belonging to the given
* charset.
*/
export function getCharacters(charSet: CharSet, document: vscode.TextDocument) {
let characters = "";
if (charSet & CharSet.Blank) {
characters += blankCharacters;
}
if (charSet & CharSet.Punctuation) {
const wordSeparators = vscode.workspace
.getConfiguration("editor", { languageId: document.languageId })
.get("wordSeparators");
if (typeof wordSeparators === "string") {
characters += wordSeparators;
}
}
return characters;
}
/**
* Returns an array containing all the characters belonging to the given
* charset.
*/
export function getCharCodes(charSet: CharSet, document: vscode.TextDocument) {
const characters = getCharacters(charSet, document),
charCodes = new Uint32Array(characters.length);
for (let i = 0; i < characters.length; i++) {
charCodes[i] = characters.charCodeAt(i);
}
return charCodes;
}
/**
* Returns a function that tests whether a character belongs to the given
* charset.
*/
export function getCharSetFunction(charSet: CharSet, document: vscode.TextDocument) {
const charCodes = getCharCodes(charSet, document);
if (charSet & CharSet.Invert) {
return (charCode: number) => {
return charCodes.indexOf(charCode) === -1;
};
} else {
return (charCode: number) => {
return charCodes.indexOf(charCode) !== -1;
};
}
}
|
import { MatTableDataSource } from '@angular/material/table';
import { FilterChip } from '../filter-component/filter.types';
import { BehaviorSubject } from 'rxjs';
import { Subscriptions } from '../../utils/subscriptions';
import { tap } from 'rxjs/operators';
import { EntityState } from './table.types';
import { MatSort } from '@angular/material/sort';
export class NaxtTableDataSource<T> extends MatTableDataSource<T> {
private readonly originalList: T[];
private filterSubject: BehaviorSubject<FilterChip[]>;
private subscriptions: Subscriptions = new Subscriptions();
constructor(data: T[]) {
super(data);
this.originalList = data;
this._initFilterObserver();
}
get rows(): T[] {
return this.data;
}
updateData(data: T[]): void {
this.originalList.splice(0, this.originalList.length); // clear list
this.originalList.push(...data); // insert new data
this._refreshFilter(); // refresh filter for new data
}
_compareFn = new Intl.Collator('en', { sensitivity: 'base', numeric: true }).compare;
sortData: (data: T[], sort: MatSort) => T[] = (data: T[], sort: MatSort): T[] => {
const active = sort.active;
const direction = sort.direction;
if (!active || direction == '') {
return data;
}
return data.sort((a, b) => {
const valueA = this.sortingDataAccessor(a, active);
const valueB = this.sortingDataAccessor(b, active);
const comparatorResult = this._compareFn(valueA as string, valueB as string);
return comparatorResult * (direction == 'asc' ? 1 : -1);
});
};
/**
* define custom sorts here
*/
sortingDataAccessor = (data: T, sortHeaderId: string): string | number => {
switch (sortHeaderId) {
// custom sort for state column
case 'state': {
switch (data[sortHeaderId]) {
case EntityState.TODO:
return 1;
case EntityState.DOING:
return 2;
case EntityState.DONE:
return 3;
default:
return -1;
}
}
}
return data[sortHeaderId];
};
updateFilter(filterList: FilterChip[]): void {
this.filterSubject.next(filterList);
}
private _refreshFilter(): void {
this.filterSubject.next(this.filterSubject.value);
}
private _initFilterObserver(): void {
this.filterSubject = new BehaviorSubject<FilterChip[]>([]);
this.subscriptions.plusOne = this.filterSubject.pipe(tap(s => this._addNewFilter(s))).subscribe();
}
private _addNewFilter(filterList: FilterChip[]): void {
let tempList: T[] = this.originalList;
filterList.forEach((ele, index) => {
if (index === 0) {
tempList = tempList.filter(s => this._filterKey(s[ele.key], ele.value, ele.notOperator));
} else {
if (ele.andOperator) tempList = tempList.filter(s => this._filterKey(s[ele.key], ele.value, ele.notOperator));
if (ele.orOperator)
tempList = this._mergeOrList(
tempList,
this.originalList.filter(s => this._filterKey(s[ele.key], ele.value, ele.notOperator))
);
}
});
this.data = tempList;
}
// we could do this fancy with the filterPredicate ?
private _filterKey(key: string | number | boolean, value: string, notOperator: boolean): boolean {
value = value.trim().toLocaleLowerCase();
switch (typeof key) {
case 'string': {
if (!value || value.length <= 0) return true;
return notOperator
? !key.trim().toLocaleLowerCase().includes(value)
: key.trim().toLocaleLowerCase().includes(value);
}
case 'number': {
if (!value || value.length <= 0) return true;
return notOperator
? !key.toString().trim().toLocaleLowerCase().includes(value)
: key.toString().trim().toLocaleLowerCase().includes(value);
}
case 'boolean': {
return key;
}
default:
return true;
}
}
/**
* Merges 2 lists and delets dublicated entries
* @param list1 first list
* @param list2 second list
*/
private _mergeOrList(list1: T[], list2: T[]): T[] {
// sadly the beautifull solution with set has abysmal performance -.-
// return Array.from(new Set([...list1, ...list2]));
return [...list1, ...list2].filter((ele, index, array) => array.indexOf(ele) === index);
}
delete(element: T): void {
this.data = this.data.filter(el => el !== element);
}
disconnect(): void {
this.subscriptions.unsubscribe();
}
}
|
def connectionLost(self, reason=None):
if self.fileobj is not None:
self.fileobj.close()
if reactor.running:
reactor.stop() |
The Reserve Bank of India has said interest rates have peaked and indicated that it would reduce them after taking into account the Budget proposals and global commodity prices.
“The inflation and interest rate cycles have peaked and have to come down and for that the RBI will look at events like the budget (2012-13), international crude oil prices and other variable factors to calibrate its policies,” RBI Governor D Subbarao said here.
Finance Minister Pranab Mukherjee will present the Budget for 2012-13 fiscal in the Lok Sabha on March 16 during which, he is expected to announce steps to arrest slowdown in economic growth.
The GDP growth rate in 2011-12 is expected to moderate to 6.9 per cent from 8.4 per cent a year ago. Defending RBI’s decision to hike interest rates (Repo) 13 times since March 2010, Subbarao said the measures were required to tame inflation which had significantly crossed the threshold levels.
The RBI Governor said growth had to be sacrificed to bring down inflation. Overall inflation in January stood at 6.55 per cent. On the overall health of the financial system, Subbarao said it is sound and prudently regulated to weather the impact of global economic crisis despite the fact that “our economy is more globalised than we tend to acknowledge”.
“However, it was due to sound and prudent monetary policy of the RBI that our country was by and large saved from the blues of global economic crisis and the savings of our people stood protected in contrast to their counterparts in the US and other Western countries,” he said.
The RBI Governor, further said that the central bank has unlocked liquidity with a view to sustain growth. In its third quarter policy review last month, the RBI had lowered the cash reserve ratio by 0.50 percentage points to 5.5 per cent, releasing Rs 32,000 crore worth primary liquidity into the system.
The apex bank, however, did not reduce interest rates ignoring the advise of some of the members of its Technical Advisory Committee (TAC). |
/**
* Recebe um JSON com uma lista de estados a serem atualizados e os atualiza
*
* @param mensagem
*/
@Transactional("localTransactionManager")
public void atualizaEstado(String mensagem) {
try {
ObjectMapper mapper = new ObjectMapper();
FileDTO file = mapper.readValue(mensagem, FileDTO.class);
Estado estado;
List<Estado> es = repo.find(file.getEstado().getHost(), file.getEstado().getPort());
if(es.size() == 0) {
estado = new Estado();
estado.setHost(file.getEstado().getHost());
estado.setPort(file.getEstado().getPort());
estado.setTime(file.getEstado().getTime());
estado = repo.save(estado);
} else {
estado = es.get(0);
}
file.setEstado(estado);
repoFile.save(file);
log.info("Estados internos atualizados.");
} catch (IOException e) {
log.error("Erro ao ler JSON de estados.", e);
}
} |
Passing and moving: negotiating fieldwork roles in football fandom research Purpose Gaining access to the research field has received much academic attention; however, the literature focusing on the changing and/or multiple roles that researchers adopt during fieldwork has at times been oversimplified. The purpose of this paper is to outline the multiple stages of the fieldwork journey, a more reflexive approach to fieldwork and the research process can be attained. Design/methodology/approach Data were generated from a four-year ethnographic study of the match-day experiences of the fans of Everton Football Club. In total, over 100 hours were spent doing fieldwork observations with an additional 25 semi-structured interviews forming the data set. Findings This paper argues that researchers should be more critical of their position in the field of their research, and should seek to identify this more clearly in their scholarship. This in turn would enable for more discussions of how each stage of the fieldwork journey affected the scope and overall findings of the research. Originality/value This paper provides an alternative framework for ethnographic researchers to better recognise and acknowledge reflexivity throughout the research process. This is done by outlining the various stages of fieldwork engagement more clearly to better understand how researchers change and adapt to the research environment. |
ERK1/2 and p38 MAPKs Are Complementarily Involved in Estradiol 17-d-Glucuronide-Induced Cholestasis: Crosstalk with cPKC and PI3K Objective The endogenous, cholestatic metabolite estradiol 17-d-glucuronide (E217G) induces endocytic internalization of the canalicular transporters relevant to bile formation, Bsep and Mrp2. We evaluated here whether MAPKs are involved in this effect. Design ERK1/2, JNK1/2, and p38 MAPK activation was assessed by the increase in their phosphorylation status. Hepatocanalicular function was evaluated in isolated rat hepatocyte couplets (IRHCs) by quantifying the apical secretion of fluorescent Bsep and Mrp2 substrates, and in isolated, perfused rat livers (IPRLs), using taurocholate and 2,4-dinitrophenyl-S-glutathione, respectively. Protein kinase participation in E217G-induced secretory failure was assessed by co-administering selective inhibitors. Internalization of Bsep/Mrp2 was assessed by confocal microscopy and image analysis. Results E217G activated all kinds of MAPKs. The PI3K inhibitor wortmannin prevented ERK1/2 activation, whereas the cPKC inhibitor G6976 prevented p38 activation, suggesting that ERK1/2 and p38 are downstream of PI3K and cPKC, respectively. The p38 inhibitor SB203580 and the ERK1/2 inhibitor PD98059, but not the JNK1/2 inhibitor SP600125, partially prevented E217G-induced changes in transporter activity and localization in IRHCs. p38 and ERK1/2 co-inhibition resulted in additive protection, suggesting complementary involvement of these MAPKs. In IPRLs, E217G induced endocytosis of canalicular transporters and a rapid and sustained decrease in bile flow and biliary excretion of Bsep/Mrp2 substrates. p38 inhibition prevented this initial decay, and the internalization of Bsep/Mrp2. Contrarily, ERK1/2 inhibition accelerated the recovery of biliary secretion and the canalicular reinsertion of Bsep/Mrp2. Conclusions cPKC/p38 MAPK and PI3K/ERK1/2 signalling pathways participate complementarily in E217G-induced cholestasis, through internalization and sustained intracellular retention of canalicular transporters, respectively. Introduction Bile formation is a highly regulated process that depends upon the coordinated transport activity of basolateral and canalicular carriers on the hepatocyte. Complex signalling pathways regulate the density and activity of these carriers, and its imbalance impairs secretory function, leading to retention of biliary components both in liver and blood. Estradiol 17-D-glucuronide (E 2 17G) is an endogenous metabolite of 17-estradiol that induces an acute, reversible cholestasis in rats. Since its level build up during pregnancy, it has been suggested to be relevant to the pathogenesis of intrahepatic cholestasis in pregnant, susceptible women. The mechanisms involved in E 2 17G induces cholestasis seems to be multifactorial. Trans-inhibition by E 2 17G of Bsep-mediated canalicular transport of bile salts and increase in paracellular permeability leading to dissipation of plasma-to-bile osmotic gradients have been shown to be causal factors. In addition, our group showed that endocytic internalization of both Bsep and Mrp2, two canalicular transporters crucial for bile formation, is also a key cholestatic mechanism. In addition, our group also demonstrated that E 2 17G activates both classical (Ca 2+dependent) protein kinase C (cPKC) isoforms and the phosphoinositide 3-kinase (PI3K)-Akt signalling pathway, and that these signalling events are cooperatively involved in the changes in Bsep/Mrp2 localization status. Whereas cPKC triggers endocytic internalization of these transporters, PI3K-Akt retains them into intracellular, vesicular compartments. Since the biological effects of cPKC and PI3K are often mediated through downstream protein kinases, we speculated that other signalling mediators are involved, provided they can crosstalk with cPKC or PI3K. Likely candidates are the mitogen activated protein kinases (MAPKs). Both cPKC and PI3K/Akt often act as upstream signalling activators of MAPKs in hepatocytes. The main vertebrate MAPKs are the extracellular signalregulated kinases 1 and 2 (ERK1/2), c-Jun amino-terminal kinases 1 and 2 (JNK1/2), and p38 kinase (isoforms a and b in hepatocytes). ERK1/2 is preferentially activated by growth factors, while JNK1/2 and p38 are more responsive to stress stimuli. After recognition of these effectors by surface receptors, MAPKs are activated by three-tiered, sequential phosphorylations mediated by small GTP-binding proteins (e.g., Ras, Rap) and two protein kinases (MAPKKK and MAPKK), which act as dual-specificity enzymes that activate a selective MAPK type. Non-canonical activation of MAPKs involves MAPK autophosphorylation, or direct MAPK phosphorylation by alternative protein kinases, such as Src or ZAP70. These findings have prompted us to study the role for MAPKs in E 2 17G-induced cholestasis. In particular, we ascertained which MAPK types contribute to E 2 17G-induced cholestasis both in isolated rat hepatocyte couplets (IRHCs) and in the isolated, perfused rat liver (IPRL). Our results demonstrate that both p38 and ERK1/2 contribute to the impairment of localization and function of Bsep and Mrp2 induced by E 2 17G, by acting in a complementary manner, and downstream of cPKC and PI3K, respectively. Ethics Statement All animals received humane care according to the criteria outlined in the ''Guide for the Care and Use of Laboratory Animals'' Eight Edition (National Academy of Sciences, 2011). Experimental procedures were carried out according to the local Guideline for the Use of Laboratory Animals (Resolution Nu 6109/012) established by the institutional Bioethical Committee for the Management of Laboratory Animals and approved by the Faculty of Biochemical and Pharmaceutical Sciences of the National University of Rosario. A system of local committeebased regulatory control offers an equivalent level of regulatory control to that exercised by the systems in Canada. The guidelines for the use of laboratory animals has been approved by our Faculty in 2002, based on CCAC (Canadian Council on Animal Care) guidelines documents. The document has been updated since then according to current and emerging issues for the research community and to advances in laboratory animal care. Animals Adult female Wistar rats weighing 250-300 g and bred in our animal house as described, were used in all studies. Treatment were carried out under urethane anesthesia (1 g/kg intraperitoneally), and maintained thus throughout. When necessary, body temperature was measured with a rectal probe and maintained at 37uC. Isolation and Culture of IRHCs To obtain a preparation enriched in IRHCs, livers were perfused according to the two-step collagenase perfusion procedure, and further enriched by centrifugal elutriation. Cell viability, assessed by trypan blue exclusion, was greater than 90%. To allow for restoration of their polarity, IRHCs were plated onto 24-well plastic plates at a density of 0.5610 5 U/mL in L-15 culture medium, and cultured for 5 h. Function and localization of Bsep and Mrp2 in IRHCs Functional changes in Bsep and Mrp2 were evaluated by assessing the canalicular vacuolar accumulation (cVA) of CGamF, a fluorescent Bsep substrate, or glutathione methylfluorescein (GS-MF), a fluorescent Mrp2 substrate derived from CMFDA; this latter compound diffuses passively, and is intracellularly metabolized by esterases and glutathione S-transferases to render GS-MF. After a 20-min equilibration period, the p38 inhibitor SB203580 (250 nM) or the ERK1/2 inhibitor PD98059 (5 mM) or their solvent (DMSO; 370 mL/L) were added to the reservoir. Fifteen min later, a 5-min basal bile sample was collected, followed by administration of E 2 17G (3 mmol/liver; single intraportal injection, over a 1-min period), or its solvent in controls . Bile was then collected at 5-min intervals for another 30-min period. Experiments were considered valid only if the initial bile flow was greater than 30 mL/min/kg of body weight. Liver viability was evaluated by monitoring lactate dehydrogenase release into the perfusate outflow ; experiments exhibiting activities over 20 U/L were discarded. DNP-SG content in bile was measured by high-performance liquid chromatography, as described. Biliary bile salt concentration was determined by a modification of the Talalay's method. At the end of the perfusion period, a liver lobe was excised, frozen in isopentane precooled in liquid nitrogen, and stored at 280uC for further immunofluorescence and confocal microscopy analysis of Mrp2 and Bsep intracellular localization. F-actin staining was carried out to demarcate the limits of the canaliculi, as described. Liver sections were fixed and stained as described, followed by overnight incubation with the specific antibodies against Bsep or Mrp2, and 1 h incubation with the appropriate cyanine 2-conjugated secondary antibodies, and Alexa Fluor 568 phalloidin (1:100, 1 h) for F-actin. To ensure comparable staining and image capture performance for the different groups belonging to the same experimental protocol, liver slices were prepared on the same day, mounted on the same glass slide, and subjected to the staining procedure and confocal microscopy analysis simultaneously. Quantification of the degree of Bsep and Mrp2 endocytic internalization was performed on confocal images using ImageJ 1.34 m (National Institutes of Health), as described. Statistical Analysis Results are expressed as mean 6 standard error of the mean (SEM). Statistical analysis was performed with one-way analysis of variance followed by the Newman-Keuls test. The variances of the densitometric profiles of Bsep and Mrp2 localization were compared with the Mann-Whitney U test. P values,0.05 were considered to be statistically significant. representative Western blottings of phospho (p)-p38, p-ERK1/2, p-JNK1/2 and total forms of all these MAPK types were obtained from whole cellular lysates of primary-cultured rat hepatocytes incubated with E 2 17G (200 mM) for 10 to 60 min, or with E 2 17G (200 mM) for 20 min in cells pretreated with the PI3K inhibitor wortmanin (WM, 100 nM) or with the cPKC inhibitor G 6976 (G, 1 mM) for 15 min. A (right panel), and B and C panels show phosphorylation status of all MAPK types evaluated (calculated as the p-MAPK to total MAPK ratio for each experimental condition). An arbitrary value of 100 was assigned to the band of highest densitometric intensity in every Western blot before the ratio was calculated. The results are shown as mean 6 SEM (n = 5). *P,0.05 vs. control (cells treated only with DMSO), and # P,0.05 vs. E 2 17G (20 min). doi:10.1371/journal.pone.0049255.g001 E 2 17G activates MAPKs Western blots of phospho(p)-p38 ( Fig. 1; A, right panel), p-ERK1/2 ( Fig. 1; B) and p-JNK1/2 ( Fig. 1; C) showed that E 2 17G increased the amount of each p-MAPK in a time-dependent manner, with increments becoming apparent as soon as 10 min after E 2 17G administration. Pretreatment with the cPKC inhibitor G6976 (1 mM) or the PI3K inhibitor WM (100 nM) selectively prevented the increase in p-p38 and p-ERK1/2, respectively ( Fig. 1), indicating that activation of p38 depends upon cPKC whereas activation p-ERK1/2 depends upon PI3K. Instead, pretreatment with either WM or G6976 prevented the increase in JNK2 phosphorylation induced by E 2 17G; although a clear trend towards prevention of JNK1 phosphorylation was also observed, this prevention only achieved statistical significance for G6976. E 2 17G-induced impairment of Bsep and Mrp2 transport function and localization in IRHCs involves p38-and ERK1/2-dependent, additive mechanisms Functional studies in IRHCs revealed that both the p38 inhibitor SB203580 and the ERK1/2 inhibitor PD98059 significantly prevented E 2 17G-induced impairment in cVA of both the Bsep and the Mrp2 substrates (CGamF and GS-MF, respectively; Fig. 2). Contrarily, the JNK1/2 inhibitor SP600125 was without effect, suggesting that JNK1/2 activation does not play a causal role in E 2 17G-induced cholestasis. The effect of E 2 17G on Bsep and Mrp2 transport activity was accompanied by changes in the localization status of these transporters (Fig. 3, top panels). In control IRHCs, the carrierassociated fluorescence was localized mainly in the canalicular vacuoles, whereas in the E 2 17G-treated group, there was extensive relocalization of the fluorescence from the canalicular zone to the cellular body, indicating endocytosis of the canalicular carriers. This phenomenon was markedly prevented by either p38 or ERK1/2 inhibition (Fig. 3, top panels). This was confirmed by densitometric analysis, which showed a flatter Bsep and Mrp2 fluorescence profile in E 2 17G-treated IRHC (Fig. 3, lower panels). ERK1/2 or p38 inhibition prevented partial or totally this relocalization, as densitometric curves were statistically different from that of E 2 17G alone (Fig. 3, lower panels). The preventive effects of PD98059 and SB203580 on CGamF and GS-MF secretory failures were additive in nature (Fig. 2), suggesting that ERK1/2 and p38 act through different but complementary mechanism. However, additivity of effects can only be assumed when recorded at concentrations of the inhibitors producing maximal effects individually. This was actually the case, since the protective effects of each of them remained virtually the same at concentrations 5-time higher than those used here in additivity studies (data not shown). cPKC-p38 and PI3K-ERK1/2signalling pathways are involved in E 2 17G-induced canalicular secretory failure in an additive manner We have previously demonstrated that the cholestatic effect of E 2 17G is partially prevented by the selective inhibition of cPKC and of PI3K, and that these kinases play a partial and complementary role in E 2 17G-induced cholestasis. Since we demonstrated here that cPKC and PI3K are selectively involved in p38 and ERK1/2 activation, respectively (Fig. 1), we tried to reproduce these selective dependencies in the functional field (Fig. 2). As expected, there was a lack of additivity in the protective effects when G6976 (cPKC inhibitor) and SB203580 (p38 inhibitor) were added together, and the same holds true when WM (PI3K inhibitor) was added together with PD98059 (ERK1/2 inhibitor). On the other hand, adittivity was observed when crosscombinations of these inhibitors were used (i.e., G6976 plus PD98059, or WM plus SB203580). Besides, the pretreatment of IRHCs with these same combinations of inhibitors markedly prevented the E 2 17G-induced internalization of canalicular carriers (Fig. 3, upper panels), resulting in control-like densitometric curves of Bsep/Mrp2 localization (Fig. 3, lower panels). This supports the fact that cPKC-p38 and PI3K-ERK1/2 signalling pathways act in parallel and in a complementary manner to account for E 2 17G cholestatic effects. p38-ERK1/2 co-inhibition prevents the colocalization of Bsep/Mrp2 with the endosomal protein Rab11a induced by E 2 17G In E 2 17G cholestasis, canalicular carriers are rapidly endocytosed from the canalicular membrane, and can return back to this membrane, presumably from the apical recycling endosomes (ARE). To visualize this phenomenon, and its prevention by MAPK inhibitors, we evaluated in IRHCs the colocalization status of canalicular carriers with the endosomal protein Rab11a, a marker of ARE that has been shown to colocalize with apically endocytosed proteins in hepatocytes. Using this approach, no colocalization was observed between Rab11a (red) and Bsep or Mrp2 (green) in controls (Fig. 4). Contrarily, E 2 17G induced colocalization of canalicular carriers with Rab11a (yellow/orange merge). IRHCs pretreated with either SB203580 or PD98059 showed lack of colocalization, as in controls; this indicates that either or both the carrier internalization towards ARE was prevented or the carrier reinsertion from ARE has been accelerated. p38 is involved in the initial impairment of bile secretory function induced by E 2 17G, whereas ERK1/2 blocks its recovery in the IPRL model The IRHC model revealed the existence of two complementary mechanisms accounting for E 2 17G-induced cholestasis, which differentially depends upon the cPKC/p38 and PI3K/ERK1/2 pathways (see above). The nature of these two cholestatic mechanisms cannot however be ascertained by using this approach. Using the IPRL model, which unlike IRHCs allow for the dissection of mechanisms occurring separately in time, we had shown a differential role for cPKC and PI3K in E 2 17Ginduced cholestasis: whereas cPKC is involved in the initial reduction in bile flow due to transporter endocytosis, PI3K (via Akt) blocks the otherwise spontaneous reinsertion of the endocytosed transporters. We therefore used this model to assess whether p38 and ERK1/2 reproduce the cholestatic mechanisms induced by their respective upstream activators, cPKC and PI3K. The bolus administration of E 2 17G induced a 61% decrease in bile flow within 10 min, which did not recover throughout the perfusion period (Fig. 5, upper panel). This was accompanied by a decrease in the biliary excretion of the Mrp2 and Bsep substrates DNP-SG and taurocholate, respectively (Fig. 5, middle and lower panels). Whereas the p38 inhibitor SB203580 (250 nM) prevented this initial drop, the ERK1/2 inhibitor PD98059 (5 mM) had little, if any, effect on this event. In contrast, PD98059 accelerated the recovery of both bile flow and biliary excretion of Mrp2 and Bsep substrates from 15 minutes of E 2 17G administration onwards. SB203580 and PD98059, when added alone, did not induce any changes in these parameters (data not shown). Prevention of the biliary secretory failure induced by E 2 17G by SB203580 and PD98059 was accompanied by the recovery of the normal canalicular localization of Bsep and Mrp2 at the end of the perfusion period (Fig. 6). In control livers, transporter-associated fluorescence was confined to the canalicular space (from 21 mm to +1 mm). In E 2 17G-treated livers, relocalization of intracellular fluorescence associated with both carriers from the canalicular space to the pericanalicular area was apparent, as indicated by the decrease in the fluorescence intensity in the canalicular area together with the increased fluorescence at a greater distance from the canaliculus (P,0.001 vs. controls); this indicates endocytic internalization of the carriers. Whereas SB203580 and PD98059 itself did not induce any changes in transporter localization (data not shown), both inhibitors extensively prevented the internalization of Bsep and Mrp2, as illustrated by a control-like pattern of the Bsep and Mrp2 distribution profiles in livers pretreated with the inhibitors, as compared with the E 2 17G-treated group (P,0.005 vs. E 2 17G; n = 20-50 canaliculi per experimental group, from three independent experiments). This supports our contention that p38 contributes to E 2 17G-induced cholestasis by retrieving canalicular carriers from their membrane domain in a cPKC-dependent manner. On the other hand, ERK1/2 would hinder the spontaneous retargeting of the endocytosed transporters by acting downstream PI3K. Discussion This study provides further insights into the intracellular signalling pathways involved in E 2 17G-induced cholestasis. The activation of these two signalling pathways by E 2 17G accounted for two independent and complementary cholestatic mechanisms; whereas the cPKC-p38 signalling pathway was essential for the acute cholestatic effect induced by E 2 17G, a phenomenon largely attributed to the endocytic internalization of Mrp2/Bsep, PI3K-ERK1/2 promoted the intracellular retention of these transporters during the cholestatic phenomenon (see Fig. 7). A similar complementarity had been demonstrated by us for their respective upstream activators, i.e. cPKC and PI3K/Akt. The mechanism by which p38 mediates endocytosis results elusive because of our poor knowledge on the molecular mechanisms that trigger this process, but several reports in other cell models provide some hints. p38 accelerates clathrin-mediated endocytosis of several ligand receptors, such as epidermal growth factor receptor (EGFR), the opioid receptors m and d, and the glutamate receptor, and there are several similarities between the internalization process of these receptors and that of Bsep. Like them, Bsep is internalized in a clathrin-dependent manner through a mechanism requiring the adaptor protein AP-2, and AP-2 phosphorylation by p38 is essential for cargo recruitment to clathrin-coated pits, as shown for EGFR endocytosis. Another putative p38 target is the Rab GDP dissociation inhibitor Rab:GDI. This protein stimulates the membranes/cytosol recycling of Rab5, a protein that coordinates formation of clathrin-coated vesicles and their fusion with early endosomes, thus increasing endocytic rate. p38 increases Rab:GDI activity by direct phosphorylation at serine-121. Whether similar regulations for the endocytic process apply to Bsep, and whether Mrp2 shares with Bsep clathrin-mediated endocytic mechanisms remains to be ascertained. The latter is however probable, since endocytosis of Mrp2 in E 2 17G-induced cholestasis is a microtubule-independent event, and the same applies for clathrin-dependent endocytosis. Our results also show that the PI3K/ERK1/2 pathway is involved in the intracellular retention of the endocytosed transporters, which otherwise would rapidly return to the canalicular membrane in a microtubule-dependent manner. The nature of this vesicular trafficking is speculative at this point, but presumably involves the latest, microtubule-dependent exocytic step of the transcytotic route, which occurs via apical recycling endosomes (ARE). This is supported by the facts that: 1) Bsep and Mrp2 colocalize with the ARE marker Rab11a (see Fig. 4), 2) Mrp2 and Bsep colocalize on the same vesicles with both the microtubule motor dynein and polymeric immunoglobulin receptor (pIgR), a surface protein that is transcytosed via ARE from the basolateral to the apical membrane, 3) insertion of Bsep back to the canalicular membrane from ARE is a microtubule-dependent process, and 4) ARE is interconnected in a microtubule-dependent manner with apical early endosomes (AEE), the first station on the endocytic pathway involved in apical endocytosis. Due to the microtubuledependent nature of the exocytic process, microtubule-based motor proteins are a likely target for ERK1/2 modulation of canalicular transporter exocytosis. In line with this, ERK1/2 impairs binding activity of the microtubule motor kinesin-1 ; this protein is enriched in both AEE and ARE in liver (40% and 45% of total kinesin content, respectively), and may therefore play a key role in the trafficking of canalicular transporters from AEE to ARE and in their exocytosis from ARE, which are both microtubule-dependent events. The dependency of E 2 17G-induced cholestasis on both p38 and ERK1/2 results somewhat paradoxical with respect to other reports showing that both MAPK types are involved in choleretic phenomena, such as those induced by the bile salt tauroursodeoxycholate (TUDC) and cAMP. For example, TUDCA stimulates biliary excretion of the choleretic bile salt taurocholate via an integrin-dependent dual signalling pathway that involves both Ras/Raf/MEK/ERK1/2 and Src/p38 signalling pathways ; of note, activation of ERK1/2 by TUDC depends on PI3K but not on PKC, in agreement with our results (see Fig. 1). The p38-mediated stimulation of taurocholate excretion by TUDCA was due to an enhanced Bsep trafficking from the Golgi compartment to the canalicular membrane, and the further insertion of Bsep into the apical membrane, in a microtubule-dependent manner. The reason for the paradoxical pro-cholestatic and pro-choleretic effects of MAPKs is unclear at present, but several possibilities arise. MAPK activation is a highly compartmentalized process, which requires activation and localization in the organelle of scaffold proteins to allow for the regional recruitment of MAPK cascade components. In quiescent hepatocytes, Raf-1 and MEK (MAPKK of ERK) are restricted to early endocytic compartments. Therefore, rapid effects on endosomal trafficking of a MAPK modulating agent like E 2 17G may be limited to this early endosomal pathway, whereas the influence on Golgi/post-Golgi pathway would require further localization of specific MAPK Figure 7. Schematic representation of the signalling events involved in E 2 17G-induced cholestasis by endocytic internalization and further retention of canalicular transporters relevant to bile formation (Bsep, Mrp2). p38, acting downstream of cPKC, triggers endocytic internalization of the apical carriers presumably towards apical early endosomes (AEE), the first intracellular endosomal compartment receiving internalized proteins from the apical membrane, in a microtubule-independent manner (solid arrow). These transporters traffic to, and accumulate into, apical recycling endosomes (ARE), from where they can be retargeted to the apical membrane during the recovery of the cholestatic process, in a microtubule-dependent manner (dashed arrows). Activation of the PI3K/Akt/ERK1/2 signalling pathway halts this latter process, thus explaining the increased colocalization of Bsep/Mrp2 with Rab11a, an ARE marker. This prolongs the cholestatic effect of E 2 17G by impeding the fast, spontaneous retargeting of intracellular transporters that would lead to a rapid recovery from the cholestatic injury. doi:10.1371/journal.pone.0049255.g007 scaffold proteins at this site. In addition, E 2 17G and TUDC may activate different signalling pathways apart from MAPKs that, by operating in concert with MAPKs, result in opposite final effects. For example, TUDC-induced acceleration of the Golgi/ post-Golgi exocytic pathway depends upon dual activation of p38 and novel PKC isoforms, and we have shown that E 2 17G does not activate the novel PKC isoform e. Alternatively, in a cholestatic context, TUDC-induced MAPK-dependent choleretic mechanisms may not be operative. Actually, MAPKs do not mediate TUDCA anticholestatic effects in TLC-induced cholestasis. Finally, mediation of choleretic and cholestatic effects by p38 may reflect a differential capability of cholestatic and choleretic compounds to evocate p38 isoforms with opposite effects. Only a and p38 isoforms are present in liver, and they have been shown to exert opposite effects in other cell types. Importantly, stimulation of Mrp2 apical translocation by cAMP, which prevents E 2 17G-induced internalization of canalicular carriers, involves exclusively p38a. It would be of interest to find out whether p38a activity is decreased by E 2 17G, and whether cAMP reverses this decrease as part of its anticholestatic action. While further studies will be required to assess the specific molecular mechanisms mediated by ERK1/2 and p38 to impair the localization status of canalicular transporters in estrogeninduced cholestasis, our results strengthen the idea that there is a clear interplay between signalling cascades and intracellular trafficking in this cholestasis, and that both MAPKs are key players in its ethiology. Drugs that inhibit selectively MAPKs have been used in preclinical and clinical studies with reasonable success. As far as our results correlate with clinical cholestatic situations in humans, they should help to envisage new, feasible therapeutic strategies to treat them. |
USA Networks just dropped the first seven minutes to tomorrow’s season 3 premiere of Mr. Robot and we see that the action picks up immediately after where we left off in season 2.
Season 3 of Mr. Robot happens tomorrow night on USA at 10PM. |
In vitro multichannel single-unit recordings of action potentials from the mouse sciatic nerve Electrode arrays interfacing with peripheral nerves are essential for neuromodulation devices targeting peripheral organs to relieve symptoms. To modulate (i.e., single-unit recording and stimulating) individual peripheral nerve axons remains a technical challenge. Here, we report an in vitro setup to allow simultaneous single-unit recordings from multiple mouse sciatic nerve axons. The sciatic nerve (∼30 mm) was harvested and transferred to a tissue chamber, the ∼5 mm distal end pulled into an adjacent recording chamber filled with paraffin oil. A custom-built multi-wire electrode array was used to interface with split fine nerve filaments. Single-unit action potentials were evoked by electrical stimulation and recorded from 186 axons, of which 49.5% were classed A-type with conduction velocities (CV) greater than 1 m s−1 and 50.5% were C-type (CV < 1 m s−1). The single-unit recordings had no apparent bias towards A- or C-type axons, were robust and repeatable for over 60 min, and thus an ideal opportunity to assess different neuromodulation strategies targeting peripheral nerves. For instance, ultrasonic modulation of action potential transmission was assessed using the setup, indicating increased nerve conduction velocity following ultrasound stimulus. This setup can also be used to objectively assess the design of next-generation electrode arrays interfacing with peripheral nerves. Introduction Neuromodulation targeting the peripheral nervous system (PNS) has attracted growing research interest as evidenced by its effectiveness in managing a variety of disease conditions, including epilepsy, obesity, incontinence, urinary tract disorders, and chronic pain. PNS neuromodulation generally requires the stimulation of and recording from peripheral nerves via single electrodes or electrode arrays, the applications of which in clinical settings are limited to either microneurographic needles, a slanted Utah array, or transversal multichannel intrafascicular electrodes. In contrast, electrode arrays that interface with the brain have been routinely used in pre-clinical studies to allow recordings of action potentials from individual neurons (i.e., single-unit recordings) and focal stimulation of a small volume of the brain tissue. Further, successful translation of the electrode-brain interface is evidenced by the wide application of deep-brain stimulators and recent success of electrode arrays that interface somatosensory and motor cortices in patients. The challenge presented for electrodes to successfully interface with, and especially record from peripheral nerves, relates to the extra layers of connective tissues associated with the nerves (which are absent in the central nervous system): the epineurium that encircles the nerve trunk and provides mechanical protection, the perineurium that wraps each nerve fascicle and forms tight-junctions (i.e., blood-nerve barrier), and the endoneurium that protects micronsthick nerve fibers consisting of either one myelinated axon or a Remak bundle of unmyelinated axons. Electrodes outside the fascicles usually record a population of neural activities in the form of compound action potentials, consisting of a large peak from volleys of myelinated A-fibers and a much smaller peak from unmyelinated C-fibers, despite the greater number of C-fibers in peripheral nerves. Utah slanted arrays are one of the few clinically available intrafascicular electrode arrays that penetrate the epineurium. While capable of single-unit recordings, most action potentials recorded by the Utah slanted array are from myelinated A-type axons; unmyelinated C-type axons are underrepresented in recordings using intra-fascicular arrays (unpublished communications). Although efficient at recruiting individual muscles via focal stimulation, longitudinal and transverse intrafascicular multichannel electrode arrays generally do not record single-units from axons. The needle electrodes used in microneurography allow reliable recordings from unmyelinated C-fibers in humans. Through intricate manipulation, the needle tip can penetrate the endoneurium to be in close proximity with C-type axons so as to achieve reliable extracellular recordings. Micro lesions of recorded axons have been documented in some microneurographic recordings, indicative of its potential risk. Single-unit recordings from peripheral nerves have been mainly conducted in animals in vitro or in vivo (e.g., recordings from vagus nerves, aortic depressor nerves, spinal nerves, splanchnic nerves, and pelvic nerves ). To achieve a high signal-to-noise ratio, the technically challenging approach taken was to carefully isolate nerve trunks, gently splitting the trunk into fine nerve bundles, and teasing a bundle into microns-thick fine filaments. Commercially available microelectrode arrays, most of which are designed to interface with the brain, are not suitable for recordings from teased peripheral nerve filaments. Thus, the aforementioned studies were limited to one-channel electrode, singleunit recordings from one to two easily identifiable axons in a single nerve filament. Successful nerve splitting is technically demanding and of low efficiency, and these appear to be factors that limit the broad application of single-unit recording/stimulation of peripheral nerves. In the present study, we aimed to increase the efficiency of single-unit recordings by developing an in vitro setup that permits interfacing of multiple peripheral nerve filaments with a custom-built multi-wire electrode array, resulting in simultaneous single-unit recordings from multiple peripheral nerve axons. Our recordings consist of a significant proportion of single-unit action potentials from unmyelinated C-fibers, which are usually underrepresented in extracellular recordings from peripheral nerves. Portions of the data have been previously reported in abstract form. Methods All experimental procedures were approved by the University of Connecticut Institutional Animal Care and Use Committee. In vitro perfusion chamber A custom-built tissue perfusion chamber, which also served as the platform for multi-channel single-unit recordings, consisted of a tissue compartment perfused at 2-6 ml min −1 with oxygenated (95% O 2, 5% CO 2 ) Krebs solution (in mM: 117.9 NaCl, 4.7 KCl, 25 NaHCO 3, 1.3 NaH 2 PO 4, 1.2 MgSO 4, 2.5 CaCl 2, and 11.1 D-glucose at room temperature) and an adjacent nerve recording compartment filled with paraffin oil (Thermo Fisher Scientific, Waltham, MA). The volume of the tissue compartment was limited to be no more than 6 ml to facilitate rapid replacement of chamber solutions during pharmacological tests (within 1 min for 6 ml min −1 flow). The nerve was pulled into the recording compartment through a mouse hole under a plastic gate that separated the tissue and recording compartments. The bottom of the tissue compartment was covered by a layer of silicone for pinning down the nerve (Sylgard 182 silicone elastomer; Dow Corning, Midland, MI); a hydrophilic glass surface was placed at the bottom of the recording compartment filled with hydrophobic paraffin oil media to attract the hydrophilic nerve fibers via surface tension. Design of a multi-wire electrode array A custom fabricated 5-channel electrode array was utilized to interface with teased nerve filaments as shown in figure 1(C). The electrode array was built with a self-designed PCB board and micro-wires (Nichrome, 65 m thick, A-M Systems Inc., Sequim, WA). One end of the PCB board was configured with connectors compatible with the recording device; the other end was configured with micro-wires through metal pads. Micro-wires were deployed parallel (100-150 m apart) to each other and the insulation (Formvar) for the distal ∼0.2 mm wires was peeled off via a fine-tipped soldering iron under a stereomicroscope to expose the conducting tips (figure 1). The electrode impedance of each channel when submerged in Krebs solution was below 100 k at 1120 Hz measured by PZ5-32 (Tucker Davis Technologies, FL). The shaft of the electrode array was insulated by a thin layer of silicone. Mouse sciatic nerve harvesting Sciatic nerves were harvested from male C57BL/6 mice aged 6-8 weeks, 20-30 g (Taconic, Germantown, NJ). Mice were anesthetized by isoflurane inhalation, euthanized by exsanguination via perforating the right atrium, and transcardially perfused with oxygenated Krebs solution. The whole length of the sciatic nerve (∼30 mm) was harvested bilaterally from their proximal projection to the L4 spinal cord to their distal branches innervating gastrocnemius muscles. Sciatic nerves were then transferred to the custom-built perfusion compartment containing oxygenated Krebs solution at room temperature. One end of the sciatic nerve was pinned down in the tissue compartment. The ∼5 mm distal end of the sciatic nerve was gently pulled over and laid onto a mirror in the recording compartment filled with paraffin oil. Single-unit recording Action potentials were evoked by electrically stimulating the proximal end of the sciatic nerve in the tissue compartment. To minimize stimulus artifacts, a suction electrode fabricated with a quartz glass capillary was used to deliver monopolar cathodal current pulses (0.2-0.8 mA, 0.2 ms duration, 0.5 Hz). To enhance the signal-to-noise ratio (SNR) of recordings, the connective tissue layers (epineurium and perineurium) were carefully dissected away before splitting the individual nerve fascicle into fine filaments (∼25 m thick). The ∼5 mm distal nerve trunk in the recording chamber was exposed to collagenase (2 mg ml −1 ) at room temperature for 10 min (type 4 collagenase, Worthington, NJ) to disrupt the epineurium and mitigate nerve axon damage during splitting. The collagenase solution was subsequently removed and the nerve was rinsed thoroughly with oxygenated Krebs solution. Single-unit action potentials from all five electrodes were recorded simultaneously, digitized at 25 kHz, band-pass filtered (300-3000 Hz) and stored on a PC using an integrated neural recording Figure 1. In vitro setup for simultaneous single-unit recordings of mouse sciatic nerve axons. The schematic diagram and a picture of the setup are displayed in (A) and (B), respectively. The sciatic nerve in the tissue compartment is perfused with oxygenated Krebs solution (2-6 ml min −1 ) and the distal 5 mm of the nerve is pulled into the adjacent recording compartment filled with paraffin oil for single-unit recordings. Action potentials are evoked by a suction stimulus electrode (0.2-0.8 mA, 0.2 ms duration, 0.5 Hz) and recorded using a custom-built multi-wire electrode array (C). The distal nerve segment in the recording chamber was treated with collagenase type 4 and split into fine filaments to interface with the multi-wire electrode array (D). and stimulating system (RZ5D and PZ5-32 and included software, Tucker Davis Technologies, FL). Transmission electron microscopy Peripheral nerve tissues were processed for electron microscopy as described previously. Briefly, tissues were fixed at 4°C for 60 min in a 0.12 M phosphate buffer solution (PB, pH 7.2) containing 2.5% glutaraldehyde, 2% paraformaldehyde, and 3 mM MgCl 2. Following two rinses in PB, the tissues were further fixed with 1% Osmium tetroxide in 0.12 M PB at room temperature for 2 h in the dark. After an additional rinse in PB, the tissues were dehydrated through a series of 30, 50, 70, 95 and 100% ethyl ethanol for 10 min each followed by two exposures to 100% propylene oxide for 10 min each. Tissues were then flat-embedded in an epoxy resin mixture and polymerized at 60°C for 48 h. Blocks were sectioned on an ultramicrotome (Leica, Bannockburn, IL) and collected on grids. The grids were then stained in 2% uranyl acetate and 2.5% Sato's lead citrate, washed with water, and dried at room temperature. The images were taken with an FEI Tecnai T12 transmission electron microscope equipped with an AMT 2 K XR40 CCD (4 megapixel) camera at an accelerating voltage of 80 KV. Ultrasonic stimulation A preliminary study of neuro-modulation by high intensity ultrasound was performed. A focused 1.1 MHz ultrasound transducer (H-101G, Sonic Concepts, WA) was used to generate ultrasonic stimulation to the sciatic nerve in vitro. A power amplifier (E&I A-300, Electronics & Innovation, Rochester, NY) was used to drive the ultrasound transducer, which receives a repetitive pulse wave from a function generator (square pulse, 100 ns duration, BK Precision 4054, Yorba Linda, CA). A pulse repetition frequency (PRF) of 100/150/200/250 kHz was used. An adjustable attenuator (50DR-046, JFW Industries, Indianapolis, IN) was placed between the function generator and power amplifier for fine-tuning ultrasound intensity. Ultrasound was delivered to the nerve segment using a water cone filled with degassed water placed above the perfusion chamber. A needle hydrophone (HNR-1000, Onda Corp., Sunnyvale, CA) was used to establish the ultrasound intensity at the nerve trunk, which is quantified using the spatial peak temporal average intensity method ( i.e., the I SPTA ). The ultrasound intensity delivered to the nerve ranged from 0 up to 100 W cm −2. Statistical analysis Data were processed off-line using customized MATLAB programs (MathWorks R2016b). A rootmean-square (RMS) value of 5 ms pre-stimulation noise was calculated and five times that value was set as the detection threshold for action potential spikes. The time of first exceeding the threshold was deemed as the onset of the action potential. To avoid confusion, only easily discriminable single-unit spikes temporally separated from adjacent spikes in the record were analyzed. Conduction delays were measured as time between the onset of stimulus artifacts and the onset of recorded action potentials. Conduction velocity was computed from the conduction delay and distance between stimulating and recording electrodes. Data are presented as means±SE. One-way ANOVA was performed as appropriate using SigmaPlot v9.0 (Systat Software, San Jose, CA). Differences were considered significant when p<0.05. Results Design of the perfusion chamber for conducting the in vitro single-unit recordings is reported in section 2, and the schematic diagram and a picture of the in vitro recording setup are displayed in figures 1(A) and (B), respectively. The custom-built multi-wire electrode array (figure 1(C)) allows interfacing with split sciatic nerve filaments in the recording compartment for simultaneous single-unit recordings ( figure 1(D)). Collagenase treatment did not cause apparent morphological changes of the sciatic nerve ( figure 2(A)). After collagenase treatment, the epineurium and perineurium covering the distal 5 mm of the sciatic nerve were carefully removed, and the nerve trunk was split into fine filaments as displayed in figures 2(B) and (C). Electron microscopy images show the cross sections of nerve axons in the non-split sciatic nerve trunk (figure 2(D)) in contrast to axons in one split nerve filament (figure 2(E)). As indicated by arrows in figure 2(D), several unmyelinated axons form a Remak bundle, which is wrapped by the endoneurium. The myelinated axons, indicated by arrow heads in figure 2(D), are relatively larger in size and each is wrapped by the endoneurium. Fine nerve fiber splitting resulted in a limited number of axons in each nerve filament and disrupted endoneuriums as indicated by double arrows in figure 2(E). The dry-up of split nerve filaments in figure 2(E) was caused by the processing of tissue with volatile propylene oxide for electron microscopy. Action potentials were evoked by a monopolar cathodal stimulus delivered via a suction electrode on one end of the sciatic nerve (0.2-0.8 mA, 0.2 ms duration, 0.5 Hz). Single-unit action potentials were successfully recorded from 186 axons. Typical single-unit recordings from the custom-built multi-wire electrode array are displayed in figure 3; five electrode wires were in contact with five split nerve filaments. The tight seal of the suction stimulating electrode with the nerve resulted in a stimulus artifact width <4 ms. As displayed in figure 4(A), the robustness of the recording is demonstrated by the almost overlapping records of single units (labeled 1 to 10 in figure 3) evoked from 10 consecutive nerve stimulations (0.5 Hz). The standard deviations of the 10 consecutive conduction delays from each of the 186 axons were quantified and plotted in figure 4(B), which reveals that greater than 95% of the recorded conduction delays have normalized standard deviations less than 1.5%. The stability of the in vitro single-unit recordings was assessed by repeated recordings of 10 consecutive stimulus-evoked spikes at 5-30 min intervals for up to 60 min. Forty-three fibers were tested at 5 min after the baseline recordings and 17 fibers tested at 15, 30, and 60 min after baseline. As displayed in figure 5, the pooled data indicated that the normalized conduction delay did not change over time. Displayed in figure 6(A) is an example of conduction delays from fast-conducting A-fibers and slowconducting C-fibers, which are used to calculate nerve conduction velocities. The histogram of conduction velocities (CV) from 186 axons is plotted in figure 6(B), which were recorded from 91 split nerve filaments in 15 sciatic nerves. Usually, there are singleunit recordings in three to five channels in the electrode array and each channel includes 1-4 easily discriminable single-unit spikes. The CV appears to. Representative single-unit recordings from mouse sciatic nerve filaments. The custom-built multi-wire electrode array was in contact with five different nerve filaments, each ∼25 m thick. The 100-150 m distance between adjacent electrode wires ensure that each of the five channels of recordings are from different nerve axons. Easily discriminable single-units with peak-to-peak magnitude over 10 times the root-mean-square (rms) of the background noise are labeled from 1 to 10. figure 3 (#1-10) overlaid upon one another, illustrating the minimum variation in action potential shapes and latencies. Ten consecutive conduction delays were recorded from each of the 186 axons and the normalized standard deviation of the conduction delays were plotted as a histogram (B); the standard deviation is less than 1.5% in 95% of the axons. in vitro setup along with the reduced stimulus artifact (<4 ms) permitted recordings from nerve fibers with CV up to 7.5 m s −1. Of the 186 axons, 49.5% had CV over 1 m s −1, which are classed as myelinated A-fibers in mice. The other 50.5% axons had CV less than 1 m s −1 and are classed as unmyelinated C-type axons. The in vitro setup achieved robust and repeatable single-unit recordings from multiple sciatic nerve axons and we were able to record from a significant proportion of unmyelinated C-fibers, ideal features for an objective test bench for assessing the effect of neuromodulation of peripheral nerves. As a proof of concept, the effect of focused ultrasonic stimulation (US) was assessed using our in vitro setup. As displayed in figure 7(A), single-units evoked by 10 consecutive electrical stimuli (0.5 Hz) were recorded before, immediately after and 15 min after the delivery of focused US stimulation (up to 100 W cm −2, ∼30 s) to the nerve truck between the stimulating and recording electrodes. A representative recording is displayed in figure 7(B) and summarized data from 15 axons are presented in figure 7(C), showing a signficant reduction of the conduction delay immediately after US and complete recovery 15 min afterwards. Discussion In contrast to the widely available techniques to interface with individual neurons in the brain, studies on electrode arrays to interface with mammalian peripheral nerve axons are comparably scarce. Nerve cuff electrodes outside the peripheral nerve epineurium can only record compound action potentials contributed to by a large number of activated nerve axons. Although penetrating the perineurium, intrafascicular electrode array recordings are biased towards myelinated A-type axons. A recent study employed a floating microelectrode array (Microprobes for Life Sciences) to record from rodent sciatic nerves, but the conduction velocities were not measured and the type of recorded fibers cannot be determined. Needle electrodes used in microneurography are capable of penetrating the endoneurium to record from unmyelinated C-type axons, but are limited to single-electrode recordings. A microchannel electrode array fabricated on PDMS was used to interface with rodent dorsal rootlets which, unlike the relatively easy access to peripheral nerves, are protected by the vertebral column and enclosed by the dura mater. The recent advances in intracellular calcium indicators allow optical methods of recording rapid calcium transients as surrogate of neural action potentials. However, image recordings have not been conducted at peripheral nerve axons but only at neural somata in the DRG, and not all DRG neurons have detectable calcium transients. To the best of our knowledge, this study is the first to achieve simultaneous single-unit recordings from mammalian peripheral nerves with no apparent bias towards myelinated A-fibers. Of the 186 axons studied, more than half were from unmyelinated fibers with slow conduction velocity, a much larger proportion than previous reports of intrafascicular recordings using penetrating electrode arrays. Anatomically, the small diameter of C-fibers relative to A-fibers contributes to the weaker total transmembrane currents associated with action potentials, generating comparably smaller extracellular source signals. Thus, it is common to encounter myelinated A-type fibers more frequently than C-type when conducting extracellular recordings from peripheral nerves. The proportion of our recorded unmyelinated fibers is almost comparable to the lower range of the C-fiber proportions reported by anatomic studies (60-70%), supporting the efficiency of our approach to record from unmyelinated peripheral axons. The maximal conduction velocity (CV) recorded in our in vitro preparation was limited to <7.5 m s −1 by the total length of the sciatic nerve (30 mm) and width of stimulus artifact (4 ms). The histogram of the recorded CV follows a continuous bimodal distribution contributed by A-type and C-type fibers; the distribution tapers as CV is greater than 6.5 m s −1. Previous reports showed that most myelinated axons in mouse spinal saphenous and sural nerves have CVs below 10 m s −1 when recorded at 37°C, comparable to the maximum CV recorded in our setup at room temperature. Regarding the strength of recorded extracellular action potential spikes from peripheral nerve axons, mathematical models and experimental data have indicated that the amplitude of the single-unit action potential increases inversely as a function of the distance between the recording electrodes and nerve axon [45,, and directly as a function of axon CV, which correlates with axon diameter and myelination. Thus, the efficiency of our methods to record from unmyelinated axons is likely due to the short distance between the axons and recording electrodes, which is presumably contributed to by the fine nerve splitting to disrupt the endoneurium layers that separate unmyelinated axons from electrodes. Compared to the concentration and duration used to dissociate DRG neurons for patch-clamp recordings, the strength of collagenase treatment used in the current study is mild and does not cause apparent morphological changes to the nerve trunk. However, collagenase treatment contributes to the increased rate of successful single-unit recordings by reducing the circumferential mechanical strength of the epineurium, and thus facilitates the nerve splitting process by mitigating constrictive damage to the split filaments. In addition to effectively reduced recording noise, the large contact area between the micro-wire recording electrode surface and individual split nerve filament increases the likelihood of forming a focal small distance between the axon and electrode. Besides nerveelectrode distance and axon CV, the impedance between the recording and reference electrodes also positively correlates with the amplitude of recorded single-unit action potentials. Thus in this study, the recording amplitude was further enhanced as paraffin oil in the recording compartment effectively increased the impedance of the recording electrode relative to the reference electrode. Following a similar design concept, novel microelectrode arrays can be developed to interface with peripheral nerve axons and achieve reliable recordings which can be conveniently tested using this in vitro setup. In this study, action potentials generated by suprathreshold electrical stimulation can be reliably recorded as evidenced by negligible variations in conduction delays following consecutive stimuli. Further, the conduction delays recorded from the same axon are robust and repeatable for at least 60 min (the longest time tested in this study), providing an ideal testing bench to assess a variety of neuromodulation and interventional schemes targeting peripheral nerves. As a proof of concept, we tested the effect of focused ultrasound applied briefly between the stimulus and recording electrodes, which caused a significant increase in nerve conduction velocity immediately after ultrasound application that completely recovered within 15 min. The effects of ultrasonic modulation on peripheral neural transmission have been studied previously and the results are contradictory. Tsui et al reported increased nerve conduction velocity following ultrasonic stimulation, consistent with our current findings, whereas others reported that ultrasound reduced nerve conduction velocity and attenuated action potential transmission. It is worth mentioning that all prior studies of ultrasonic modulation were based upon measuring changes in compound action potential shapes and amplitude, which are not solely determined by action potential transmission but also are affected by changes in recording conditions likely occurring following mechanical disturbances from ultrasound (e.g., change of nerve-electrode distance). In comparison, the recordings in the present experiments avoided those recording-related confounding factors. With carefully designed experiments, the current setup has the potential to provide convincing data to resolve the aforementioned controversies on ultrasonic neuromodulation of peripheral nerves. In addition, the ability to record multiple single-unit action potentials simultaneously will allow the application of this setup to efficiently study other neuromodulatory strategies including, but not limited to electrical, infrared light, and chemical/ pharmacological. Conclusions We developed a novel in vitro setup to allow simultaneous single-unit recordings from multiple mouse sciatic nerve axons that are not biased towards recordings of fiber type. An enhanced action potential amplitude from extracellular recordings was achieved by reducing the axon-recording electrode distance via collagenase treatment, fine fiber splitting, and increased contact area between electrodes and nerve axons. This in vitro setup can be used as a test bench to objectively assess the design of next-generation electrode arrays for interfacing with peripheral nerves. This setup can also serve as an objective and convenient platform to study a variety of neuromodulation strategies that target peripheral nerves, including electrical, infrared light, ultrasonic and pharmacological. |
/*|-----------------------------------------------------------------------------
*| This source code is provided under the Apache 2.0 license --
*| and is provided AS IS with no warranty or guarantee of fit for purpose. --
*| See the project's LICENSE.md for details. --
*| Copyright (C) 2019-2022 Refinitiv. All rights reserved. --
*|-----------------------------------------------------------------------------
*/
package com.refinitiv.eta.codec;
class PriorityImpl implements Priority
{
int _class;
int _count;
@Override
public void clear()
{
_class = 0;
_count = 0;
}
@Override
public void priorityClass(int priorityClass)
{
assert (priorityClass >= 0 && priorityClass <= 255) : "priorityClass is out of range (0-255)"; // uint8
_class = priorityClass;
}
@Override
public int priorityClass()
{
return _class;
}
@Override
public void count(int count)
{
assert (count >= 0 && count <= 65535) : "priorityCount is out of range (0-65535)"; // uint16
_count = count;
}
@Override
public int count()
{
return _count;
}
}
|
Q:
Are there any time-sensitive events in Skyrim?
Are there any time-sensitive events in Skyrim, whose occurrence is based on how many in-game days have passed? Or, to phrase it another way, are there any disadvantages to waiting out time indefinitely, as is often suggested for merchant restocking?
A:
The only time-sensitive event is your marriage ceremony.
Once you've proposed to your Future Significant Other, you'll need to arrange the wedding.
After you've arrange the wedding, you have about 24 hours before the ceremony takes place (it always takes place between dawn and dusk, 6:00 AM - 6:00 PM).
However, if you do miss the ceremony, you can go to your Future Significant Other and ask for another chance.
A:
The Blood on the Ice quest may fit some definition of time-based. I believe this happened very early in my game I believe the murder happened, but I basically ignored it because I was on other business.Everything I have read online, seems to indicate that I am screwed, and will never be able start/complete that or own property in Windhelm. |
Radio producer Scott Carrier quit his job at a low moment in his life. His wife left him and took the kids. And he got a job interviewing schizophrenics for some medical researchers. After doing it a while, he began to wonder if he was a schizophrenic himself. And more stories. |
The study shows high rates of turnover and fraying work and emotional relationships.
It is no secret that urban school districts are more likely to see perpetual turnover among leaders than higher performing, more stable districts in the suburbs. In fact, creating that turnover — in the name of reforming bad systems — has been a matter of federal and state education policy for more than a decade.
The impact of that turnover on important working and emotional relationships among school administrators is less well known, and is the focus of a new study led by Kara Finnigan, a professor at the University of Rochester's Warner School of Education.
Her research reveals the extent of "leadership churn" and the way it has disrupted the flow of ideas and the growth of productive relationships.
In the district Finnigan and her colleagues studied from 2010 to 2013 — it appears to be Rochester, though she would not confirm it — more than half the administrators left within four years. The administrators who remained had fewer emotional and research-based ties at the end of the study than at the beginning.
The conclusions are based on a survey administered to 181 high-ranking central office administrators and building principals in the district for four consecutive years.
Each person was given a list of every other top administrator and principal in the district and asked to indicate for each of them: Do you work with this person regularly? Is this person a source of knowledge and new ideas for you? Do you have an emotional connection with this person?
The responses showed a scarcity of reciprocal relationships, particularly emotional ones. Principals were increasingly isolated or divided into new-guard and old-guard groups, and information tended to flow from central office to principals and not in the other direction. Just as people emerged as important information hubs, they often departed and left broken networks behind.
"Constant churn can be viewed as disruptive to fiscal, human and social capital within organizations," she and her co-authors wrote. The study appears in a recently-released book that Finnigan co-edited,Thinking and Acting Systemically: Improving School Districts Under Pressure.
The research points to a paradox: For things to change, there needs to be stability in leadership. A revolving door of principals and directors can lead to stagnation and a lack of progress.
"Our policies assume stability: 'Do things better; work toward these standards,' " Finnigan said. "But just when someone's starting to (do) it, they move out and you've lost that capacity in the system."
Similar research in higher-performing districts, where turnover is less common, shows many more reciprocal ties, with people exchanging research-based ideas.
The lesson, Finnigan said, is that districts — and, in turn, their state and federal overseers — need to focus less on accountability and more on capacity-building.
"In a place that hasn't been functioning well for a long time, it's hard to create (positive) relationships," she said. "There's a lot of blame and people are less likely to go out on a limb and take risks. You have to find a way to change that dynamic." |
Aside from the 11th anniversary of 9/11, there was no bigger story on Sept. 11, 2012 than the attack made upon the U.S. embassy in Cairo, Egypt. But for half the day on Tuesday, MSNBC didn't mention the incident.
A careful review of MSNBC's programming on Tuesday proved that up until 6PM eastern, MSNBC had still not bothered to report the storming of the U.S. embassy.
Late on the morning of Sept. 11, Muslim protesters angry, they claimed, over an American movie that portrays the Prophet Muhammad in an unsavory light, scaled the walls of the embassy in Cairo. The protesters removed the U.S. flag and replaced it with the black al Qaeda flag. Some five hours later, the U.S. embassy in Benghazi, Libya was also attacked. One American was killed and the embassy was burned to the ground.
By 6PM eastern both CNN and Fox News had televised stories about the attacks while MSNBC remained quiet.
The U.S. embassy in Cairo also made waves by issuing a statement criticizing the purported movie makers in the U.S. for having "hurt the religious feelings of Muslims."
GOP Presidential nominee Mitt Romney called the embassy's statement "disgraceful."
I'm outraged by the attacks on American diplomatic missions in Libya and Egypt and by the death of an American consulate worker in Benghazi. It's disgraceful that the Obama Administration's first response was not to condemn attacks on our diplomatic missions, but to sympathize with those who waged the attacks.
Embarrassed, the Obama White House later disavowed the U.S. embassy's statement. |
package dk.in2isoft.onlineobjects.ui.jsf.model;
import dk.in2isoft.onlineobjects.model.Image;
public interface ImageContainer {
public Image getImage();
}
|
<gh_stars>0
package com.dranawhite.study.springboot;
import org.junit.runner.RunWith;
import org.springframework.boot.test.context.SpringBootTest;
import org.springframework.test.context.junit4.SpringRunner;
/**
* @author dranawhite
* @version : BaseTest.java, v 0.1 2019-07-26 14:44 dranawhite Exp $$
*/
@RunWith(SpringRunner.class)
@SpringBootTest(classes = {Application.class})
public abstract class BaseTest {
}
|
Early-Life Social Isolation Impairs the Gonadotropin-Inhibitory Hormone Neuronal Activity and Serotonergic System in Male Rats Social isolation in early life deregulates the serotonergic system of the brain, compromising reproductive function. Gonadotropin-inhibitory hormone (GnIH) neurons in the dorsomedial hypothalamic nucleus are critical to the inhibitory regulation of gonadotropin-releasing hormone neuronal activity in the brain and release of luteinizing hormone by the pituitary gland. Although GnIH responds to stress, the role of GnIH in social isolation-induced deregulation of the serotonin system and reproductive function remains unclear. We investigated the effect of social isolation in early life on the serotonergicGnIH neuronal system using enhanced green fluorescent protein (EGFP)-tagged GnIH transgenic rats. Socially isolated rats were observed for anxious and depressive behaviors. Using immunohistochemistry, we examined c-Fos protein expression in EGFPGnIH neurons in 9-week-old adult male rats after 6weeks post-weaning isolation or group housing. We also inspected serotonergic fiber juxtapositions in EGFPGnIH neurons in control and socially isolated male rats. Socially isolated rats exhibited anxious and depressive behaviors. The total number of EGFPGnIH neurons was the same in control and socially isolated rats, but c-Fos expression in GnIH neurons was significantly reduced in socially isolated rats. Serotonin fiber juxtapositions on EGFPGnIH neurons were also lower in socially isolated rats. In addition, levels of tryptophan hydroxylase mRNA expression in the dorsal raphe nucleus were significantly attenuated in these rats. These results suggest that social isolation in early-life results in lower serotonin levels, which reduce GnIH neuronal activity and may lead to reproductive failure. MaTerials anD MeThODs animals Male transgenic Wistar rats expressing enhanced green fluorescent protein (EGFP) under rat GnIH promoter (GnIH-EGFP transgenic rats), after weaning (3 weeks of age), were randomly assigned to group housing (2-4 male littermates per cage) (n = 63) or individual housing (isolated) condition (n = 65) up to 9 weeks of age. The animals were maintained under a controlled 12 h light/dark cycle (lights on from 12:00 a.m. to 12:00 p.m.) with temperature maintained at 22°C in the SPF animal facility for 6 weeks prior to sampling. Autoclaved water and food were available ad libitum to the rats. Body weight of each rat was measured once a week. All aspects of animal welfare and experiments were in accordance with the guidelines and authorization of Monash University Animals Ethics Committee, AEC (MARP/2012/140, MARP/2013/041). anxiety and Depression-like Behavior Tests The open field test (OFT) was conducted in an open field area made up of a black box (width: 1.2 m, length: 1.2 m, and height: 0.30 m) for use with white rats of 9 weeks old. A handheld camcorder (Sony Corp., Japan) was used to record the movement of the rat in the arena. The OFT experiments were performed in light (white lighting, 10:00 a.m.-12:00 p.m.) and dark conditions (red lighting, 3:00-6:00 p.m.). The video file from the camcorder was analyzed using an automated motion detector software (Lolitrack v2.0, Loligo Systems, Denmark) to track the movements of the rat. Control rats (group housed; n = 11/light phase and n = 9/night phase) and isolated rats (n = 12/light phase and n = 13/night phase) were subjected to OFT for 30 min in order to observe anxiety-like behavior in the dark phase and light phase. The total distance traveled (cm), total time of activity (s), total number of center intrusions, and total time spent in center (s) were measured. Forced swimming test (FST) was carried out using an automated behavior analytical system (MicroAct system, Neuroscience, Inc., Tokyo, Japan). FST was carried out twice; a pre-test as a habituation session and the actual test was performed 24 h later. The FST apparatus consisted of a glass cylinder (height: 45 cm and diameter: 20 cm) which was surrounded by round coil. Prior to the FST, the glass cylinder was filled with water (25 ± 1°C) to a depth of 30 cm, and a magnet (diameter: 1 mm and length: 3 mm) was taped to each front paw of the rats. Individual rat was gently lowered into the water-filled glass cylinder for a 7 min swim test. Electrical currents were generated in the coils corresponding to the movements of the magnet taped to the front paws. The currents were amplified, transformed into voltage, and recorded by the system. The duration of immobility was detected automatically using the MicroAct™ Scratch software (Neuroscience, Inc., Tokyo, Japan). Twenty-six male rats were randomly allocated to group-housing conditions (n = 17) or isolated conditions (n = 9) and used in the FST. The sucrose preference test (SPT) was conducted over a period of 5 days. Prior to the test, the rats were habituated to the presence of two water bottles in their cages for a minimum of 5 days beforehand. The rats undergoing the experiment were provided with two water bottles: one containing sucrose water (sucrose powder diluted in distilled water, 200 mL) and the other containing only distilled water (200 mL). The concentration of the sucrose water was increased every day over a period of 5 days (0, 0.25, 0.5, 1.0, and 2.0%). Water consumption was measured at 12:00 p.m. everyday by weighing the water bottles to determine consumption by weight. The position of the water bottles was swapped daily at the time of measurement to reduce preference bias in the results. The data for percentage of sucrose consumed against total water consumed was calculated from the results and used as an indicator for sucrose preference. A repeated measures test was used for statistical analysis in order to confirm any significant difference between group-housed and isolated rats in their pattern of sucrose preference over the 5-day test period. Thirty-one male rats were randomly allocated to group-housing conditions (n = 14) or isolated conditions (n = 17) at 9 weeks of age. We divided the control and the post-weaning social isolation rats each into two groups for three behavioral tests. Group I was used for OFT and FST and group II was tested for SPT. Polymerase chain reaction Control (n = 6) and socially isolated rats (n = 6) at 9 weeks of age were deeply anaesthetized with an intraperitoneal injection of ketamine xylazine (4.5 mg/kg/BW) followed by rapid removal of the brain and immediately dissected by 1 mm rat brain slicer (Neuroscience, Inc., Japan). The POA (bregma +1.2 to −0.12) and dorsal raphe (bregma −6.96 to −8.16) areas were collected with a sterile blade. Total RNA from these tissues was extracted using TRIzol (Invitrogen, Carlsbad, CA, USA) and transcribed using High Capacity Transcription Kit (Applied Biosystems, Foster City, CA, USA) according to the manufacturer's protocols. Quantitative real-time PCR (ABI 7300, Applied Biosystems Foster City, CA, USA) was performed using primers for gnrh, gnih, sert, tph, and IMPDH2 (0.2M, Table S1 in Supplementary Material) in a final volume of 10 l of 2X Power SYBR Green PCR mix (Applied Biosystems). The house keeping gene, IMPDH2 is listed as a reference gene in real-time PCR to show geometric average expression level. The resulting PCR products were validated using an ABI PRISM 310 Genetic Analyzer and Sequence Analysis Software (Applied Biosystems) and ran on a 2.5% agarose gel with ethidium bromide used for visualization. immunohistochemistry Immunocytochemistry for cFos, 5-HT, and 5-HT2A was performed on the DMN sections obtained through coronal sectioning (30 m). The perfusion fixed (4% PFA) brain tissue sections were washed with 0.1M PBS, in an incubation chamber for 10 min at room temperature and gently shaken at 60 rpm. The sections were then incubated in a blocking solution (40 L normal goat serum (NGS), 10 l 0.5% Triton-X, and 1950 l PBS in 2 mL/well) for 1 h in the same conditions as above. After washing, the sections were incubated with polyclonal rabbit anti-c-Fos antiserum diluted 1:600 (sc52, Santa Cruz Biotechnology, Inc., USA), goat anti-5HT antiserum diluted 1:1000 (20079, Immunostar Inc., WI, USA), rabbit anti-5HT2A antiserum diluted 1:200 (24288, Immunostar Inc., WI, USA) in 2 mL 0.1M PBS containing 2% NGS, 0.5% Triton-X/well for 24 h at 4°C for c-Fos, 5-HT and 5-HT2A respectively at 4°C. Next, the sections were washed in 0.1M PBS incubated in biotinylated anti-rabbit immunoglobulin G (IgG) or biotinylated anti-goat IgG (Vectastain ABC Elite kit, Vector Laboratories, Burlingame, CA, USA) for 45 min. Subsequently, the sections were incubated with avidin-biotinylated horseradish peroxidase complex for 45 min (Vectastain ABC Elite kit, Vector laboratories, Burlingame, CA, USA). Sections were visualized with Alexa Fluor 594 streptavidin conjugates (S32356, Invitrogen Corporation, USA) and pasted on microscope slides (Superfrost PLUS, Fisher Scientific, Pittsburgh, PA, USA). Mounting medium was applied (VectaShield, Vector Laboratories) followed by coverslips. The number of immunoreactive EGFP-GnIH cells within the DMN were determined using the laser scanning confocal microscope (C1si, Nikon, Tokyo, Japan), equipped with NIS-Element 4.0 Advance software. The specificity of both c-Fos and 5-HT antibody was tested using the rat brain from previous study. We divided the control and the post-weaning social isolation rats each into two groups. Group I was used for c-Fos, 5-HT (control: n = 20, isolated condition: n = 17) and group II was tested for 5-HT2A (n = 3). Double-labeled images of c-Fos and GnIH staining, viewed under the red channel were converted to magenta. The brightness and the contrast were adjusted using Adobe Photoshop CS2 (Adobe, San Jose, CA, USA). confocal analysis of c-Fos expression and 5-hT Fiber Juxtapositions to gnih neurons The procedure for confocal analysis of c-Fos expression in GnIH neurons has been described previously. Briefly, immunoreactive c-Fos positive GnIH neurons were visualized using digitized images captured with a Nikon-30 confocal microscope (C1si, Nikon Instruments Inc., Tokyo, Japan). The total number of GnIH neurons and immunoreactive c-Fos positive GnIH neurons were determined using 0.225 m Z-steps in 10-15 sections which included all EGFP-GnIH neurons in the DMN. To confirm the colocalization of c-Fos in GnIH neurons, the Z-steps were carefully inspected with 3D image rotation using NIS Elements AR Version 4.0 (Nikon Instruments Inc.). We then calculated the percentage of c-Fos positive GnIH neurons. Only cells with visible nuclei were counted. The procedure for confocal analysis of fiber projections to GnIH neurons has been described previously. Briefly, 5-HT fiber juxtapositions were captured with a confocal microscope at 0.225 m Z-steps using 60 water immersion objective lens, 4 digital zoom function to cover the entire depth of the neuron (ECLIPSE 90i, Nikon instruments Inc., Japan). Scans of 488 and 543 nm excitation wavelength were also performed sequentially across optical sectioning to avoid bleed-through between the channels. The number of GnIH neurons with intimate 5-HT fiber juxtapositions was determined in 10-15 sections to include all EGFP-GnIH neurons in the DMN. To confirm close juxtapositions between 5-HT fibers and GnIH neurons, the Z-steps were carefully inspected with 3D rotation image using NIS Elements AR Version 4.0 (Nikon Instruments Inc.). GnIH neurons with 5-HT fiber juxtapositions on the cell soma or dendrites were Social-Isolation Inhibit 5-HT and GnIH Frontiers in Endocrinology | www.frontiersin.org counted. A contact was scored only if 5-HT fiber varicosity was in direct contact with the GnIH neuron. The percentage of GnIH neurons with visible nuclei in the DMN and with at least one close juxtapositions with 5-HT fiber was calculated. statistics Data are presented as means ± SEM in all bar graphs. Behavioral data were analyzed by two-way apposition using SPSS 20 (IBM, Chicago, IL, USA). SPT was analyzed by a univariate repeated measures using SPSS 20. Immunohistochemistry and gene expression results were analyzed using the Student's t-test. Significance was set as p < 0.05. resUlTs social isolation, anxiety, Depression, and the serotonin system After 6 weeks of social isolation, we conducted three behavioral tests and took samples of brain tissues for biological study (Figure 1A). Although the total distance traveled (cm), total time Postweaning social isolation decreased the levels of gonadotropin-releasing hormone mRNA in the pre-optic area (n = 6/housing) but did not change the levels of gonadotropin-inhibitory hormone mRNA in the hypothalamus (n = 6/ housing). The relative mRNA expression levels were normalized to that of inosine 5-monophosphate dehydrogenase 2 mRNA. All data are presented as mean ± SEM. Significant differences were determined using the Student's t-test for unpaired values; significance was set at *p < 0.05. Soga et al. Social-Isolation Inhibit 5-HT and GnIH Frontiers in Endocrinology | www.frontiersin.org of activity (s), total number of intrusions into the center, and total time spent in the center (s) were not significantly different between control and isolated rats, both control and isolated rats traveled more and were active for longer durations in the dark phase compared with the light phase ( Figure 1B). No significant difference in activity in the light phase was evident between control and isolated animals ( Figure 1B). However, the total time spent in the center by socially isolated rats was significantly shorter in the dark phase than in controls (control: 145.11 ± 23.28 s, isolated: 65.85 ± 12.73 s, p < 0.05; Figure 1C). Control (n = 17) and isolated (n = 9) rats were subjected to a forced-swim test for 7 min to measure time spent immobile. A pre-test was conducted 24 h earlier for habituation. No significant difference was observed in time spent immobile between control and isolated rats during the pre-test or test sessions (Figure 1D). The difference in sucrose consumption between groups was measured over 5 days. Using a univariate repeated measures test for analysis, a significant difference was observed in sucrose preference between control and isolated rats . social isolation and reproduction There was no difference in body weight post-weaning between controls and socially isolated male rats (Figure 2A). The level of gnrh mRNA in the POA was significantly lower in socially isolated rats compared with controls (control: 1 ± 0.2 and isolated: 0.29 ± 0.2, p < 0.05; Figure 2B). However, there was no difference in the expression of gnih mRNA in the hypothalamus between socially isolated and control rats ( Figure 2B). social isolation and serotonergic regulation of gonadotropin-inhibitory hormone cells 5-HT2A-positive cells were evident in the DMN. Some EGFP-GnIH neurons co-expressed 5-HT2A (Figures 4A-C). In addition, close juxtapositions between 5-HT-immunoreactive fibers and GnIH cell bodies were observed in the DMN (Figures 4D-G). To study the effect of social isolation on the serotonergic regulation of GnIH cells, close juxtapositions between 5-HT-immunoreactive fibers and GnIH cell bodies was determined and analyzed using a laser scanning confocal microscope. The percentage of 5-HT-immunoreactive fibers in close juxtapositions to GnIH cells in the entire DMN and DTM was significantly decreased in isolated males compared with group-housed control males (control: 13.96 ± 3.05% and isolated: 5.6 ± 0.7%, p < 0.05; Figures 5A,B). However, 5-HT fiber density per unit area in the DMN was the same in control and isolated rats ( Figure 5C). Expression of the 5-HT-related genes sert and tph2 in the dorsal raphe nucleus (DR) was measured using quantitative real-time PCR. There were no differences in the levels of sert mRNA expression between control and socially isolated rats. However, tph2 mRNA expression was significantly lower in socially isolated rats (n = 6) compared with controls (n = 6; control: 1 ± 0.22 and isolated: 0.3 ± 0.11, p < 0.05; Figure 5D). DiscUssiOn In this study, we showed that post-weaning social isolation impairs GnIH neuronal activity and the serotonergic system in the DMN, which may contribute to the deregulation of GnRH neuronal activity in the POA and sexual dysfunction. social isolation, Behavior, and the serotonin system It is established that post-weaning isolation for 6 weeks from postnatal day (P) 21 (the time of weaning) has serious consequences for brain development, causing alterations in neurotransmission and behavioral abnormalities (aggression, anxiety, and depression) in rodents. This suggests that social stimuli received after weaning are critical to the development of social behaviors and related neuronal circuits. We observed a daily variation in anxiety-like behavior in both group-housed and socially isolated rats. Importantly, during the dark phase, anxiety-like behaviors were observed in socially isolated rats. However, total locomotor activity was unaffected by group or socially isolated housing. These data suggest that the anxiogenic effect of postweaning social isolation could depend on light conditions and their effect on circadian rhythm. Indeed, disrupted sleep patterns are evident in socially isolated rats. The total time spent immobile (a parameter of depressive-like behavior) is unaffected by post-weaning social isolation in rats. Several studies have shown that social isolation increases despair-like immobility and immobile time in male rats. These conflicting results may be explained by differences between rat strains and the duration of social isolation. Furthermore, in this study, post-weaning isolation resulted in a decreased sucrose intake and reduced preference for sucrose. This anhedonia-like phenotype induced by social isolation can be reversed by treatment with the antidepressant imipramine in male rats, suggesting that it is mediated by the 5-HT pathway. For the first time, we show that post-weaning social isolation specifically decreases 5-HT fiber projections to the DMN; this may, in turn, cause the downregulation of sert gene Frontiers in Endocrinology | www.frontiersin.org expression in the DR, where 5-HT neurons are primarily located. This is supported by the decrease in central 5-HT and 5-HT receptors evident in socially isolated animals during episodes of increased anxiety. Although evidence of serotonergic and 5-HT receptor activity in the DMN remains inconclusive, serotonergic projections to DMN neurons have been reported. Indeed, in this study, we found that 5-HT2A is co-localized in GnIH neurons in rats. Although the magnitude of changes in 5-HT2C in GnIH neurons in socially isolated rats is unknown, antagonists of 5-HT2C receptors reportedly increase sucrose preference. Several 5-HT receptor types, including 5-HT2C, are expressed in GnIH neurons in the DMN of female mice. Therefore, the alteration of serotonergic signaling in the DMN may underlie the reduced preference for sucrose in socially isolated rats. Therefore, GnIH neurons and other neurons in the DMN may be targets of the circuitry for anxiety and anhedonia that mediates serotonergic activity following post-weaning social isolation. social isolation and reproduction Reproductive senescence can be caused by factors related to the social environment. Stress in early life delays pubertal onset, lowers GnRH expression, lowers testosterone synthesis, and impairs sexual behavior, all of which eventually lead to sexual dysfunction in mammals (31,32,. Post-weaning social isolation impairs male sexual behavior, as indicated by an increased latency of ejaculation during adulthood. GnRH expression and release from the POA is a key regulator of gonadotropin release and reproductive behavior. Our results show that post-weaning social isolation decreases the expression of GnRH mRNA in male rats, which could lead to sexual dysfunction. social isolation and gonadotropininhibitory hormone neuronal activity This study is the first to demonstrate GnIH neuronal activity following post-weaning social isolation. Although it is established that GnIH inhibits GnRH neuronal activity and GnRH induced-LH release by the pituitary gland, we found that postweaning social isolation decreases both GnRH mRNA expression and GnIH neuronal activity. Thus, both the GnRH and GnIH systems are down regulated following post-weaning social isolation. In rats, neural activity and the release of GnRH are increased just before and after puberty in the POA, which coincides with pubertal processes, such as changes in -aminobutyric acid (GABA) and glutamate levels. Morphological changes, such as structural remodeling of the dendrites of GnRH neurons, are the key change during puberty. Although the timing of the formation of inhibitory GnIH neuronal inputs to GnRH neurons during the post-natal period remains unknown, a lack of social stimuli pre-and post-puberty may have an impact on the formation of GnIH inputs to GnRH neurons. Post-weaning social isolation may disturb normal GnIH-GnRH neuronal signaling during pubertal development, which may result in reduced expression of GnRH and GnIH in the brain. Accumulated evidence from recent studies shows that stress increases GnIH expression, with an associated suppression of the hypothalamic-pituitary-gonadal axis suggesting that the inhibitory effect of stress on reproductive function may be mediated by the GnIH system. Post-weaning social isolation causes hypofunction of the HPA axis in adult rats, suggesting that the HPA axis becomes desensitized to stressful stimuli. GnIH neurons are sensitive to stress ; thus, in socially isolated rats, they may also become desensitized, as implied by their low neuronal activity. Long-term social isolation lowers GnRH mRNA expression in the POA during adulthood; therefore, inhibitory GnIH signaling may be reduced as a result of short-loop negative feedback from GnRH neurons. Post-weaning social isolation may disrupt the normal development and balance of the GnIH-GnRH neuronal pathway for reproductive activity. We did not observe any erroneous positioning of EGFP-GnIH neurons in the DMN of socially isolated rats. Likewise, social isolation had no effect on the total number of EGFP-GnIH cells. During development, GnIH expression starts at embryonic days (E) 13-14. GnIH neurons migrate to the dorsal and ventral regions of the third ventricle at E16-17 and establish their positions in the medial hypothalamus by E18. GnIH neurons send ascending and descending projections to other regions of the brain by P1 and the GnIH neuronal system is almost completely formed by P28. Our study shows that the development and positioning of GnIH neurons in the DMN during the prenatal period is not altered by post-weaning (after P21) social isolation. social isolation and the serotonergic regulation of gonadotropin-inhibitory hormone neurons 5-HT-immunoreactive fibers form close juxtapositions to GnIH neurons in the DMN of the rat brain. The DMN receives 5-HT fibers and terminals through the medial forebrain bundle from 5-HT nerve cell bodies of the DR. Our findings related to 5-HT2A, and those of our previous study, show that 5-HT receptors are co-expressed in GnIH neurons in female mice. Additionally, administration of citalopram increases the number of GnIH neurons in the DMN and the density of GnIH fibers in the POA, supporting the concept that GnIH is under direct serotonergic control. The significant decrease in c-Fos expression evident in GnIH neurons, combined with the decreased 5-HT innervation of GnIH neurons in socially isolated animals, suggest a reduction in GnIH neuronal activity. Early environmental manipulations impair long-term synaptic potentiation. The decrease in 5-HT fiber juxtapositions to GnIH neurons reflects neuroanatomical adaptation of hypothalamic neuronal circuits in response to a deprived social environment in early life. 5-HT synaptogenesis is important for brain 5-HT concentration during the critical period of brain development. Moreover, 5-HT receptor complements are established between P30-50 in rats. Thus, the reduction in 5-HT fiber juxtapositions to GnIH neurons following post-weaning social isolation may delay post-natal maturation, which may weaken the strength of existing synapses. cOnclUsiOn In this study, we showed that post-weaning social isolation enhances anxiety-like behavior and an anhedonia-like phenotype that is related to altered sert expression in the serotonergic system. Furthermore, we demonstrated that post-weaning social isolation reduces GnIH neuronal activity and decreases 5-HT fiber juxtapositions to GnIH neurons, suggesting that serotonergic regulation may participate in GnIH signaling to accomplish normal GnRH neuronal activity and reproductive function. Although a complex molecular and neuronal mechanism is involved in post-weaning social isolation-induced reproductive dysfunction, altered serotonergic activity may be one factor that mediates GnIH-GnRH signaling in the brain. Our findings characterize the long consensus on the negative effects of post-weaning social isolation and provide insights into the neuronal and molecular mechanisms underlying the serotonergic regulation of GnIH. |
A mere six months after releasing the Four Seasons drama “Jersey Boys,” Clint Eastwood has again lapped his younger directing colleagues with his second film of 2014 and his best movie in years. “American Sniper” is quintessentially Eastwood: a tautly made, confidently constructed examination of the themes that have long dominated his work.
“American Sniper,” based on Navy SEAL marksman Chris Kyle’s best-selling memoir, is both a tribute to the warrior and a lament for war. Shirking politics, the film instead sets its sights squarely on its elite protagonist (Bradley Cooper), a traditional American war hero in an untraditional war.
Here is an archetypal American: a chew-spitting, beer-drinking Texas cowboy who enlists after the 1998 bombings of American embassies with resolute righteousness and noble patriotic duty. The once wayward Kyle finds his true calling in the Navy, and he heads to Iraq with a moral certainty that no amount of time served or kills will shake. He’s there to kill bad guys — “savages” he calls them at one point.
And kill he does. With 160 confirmed kills, Kyle is believed to be the most lethal sniper in U.S. history. The film starts with a remarkable scene of Kyle poised on an Iraq rooftop with a young boy holding a grenade in his scope. Eastwood and screenwriter Jason Hall flashback to Kyle’s upbringing, where his father taught him about “the gift of aggression” and the honor of defending others.
It’s the first of many cuts between far-away battle and the personal life Kyle leaves behind. Shortly before shipping out, he weds Taya, played by Sienna Miller, who gives a refreshingly lively take on a usually one-dimensional character. She’s more cynical than her husband, who returns to their growing family between tours, his head increasingly stuck in Iraq.
He’s much like a terse and weary Western hero torn from home; an early shot through the front door of their home evokes the famous final image of John Ford’s “The Searchers.” Instead of a Stetson, Kyle wears a baseball cap, turned backward when he takes aim. “I’m better when it’s breathing,” he tells an early instructor after shooting a snake.
Cooper is extraordinary as Kyle. He has beefed up, adopted an authentic Texas drawl and endowed Kyle with a commanding swagger. The war steadily takes its toll on his psyche, even if he’d never admit it. When Kyle’s younger brother, passing him on a Tarmac in Iraq, curses the war, Kyle looks him at with genuine befuddlement.
Eastwood has, of course, long been drawn to stories about violence — necessary if regrettable — in meting out justice and the cost to those that carry its heavy burden. The question is if the mythical rending of “American Sniper” fits its more complex basis of reality. Kyle, who died tragically in early 2013, belies easy summary. He, for one, boasted of shooting looters in New Orleans after Hurricane Katrina. His clarity of mission could also be said to mirror the mistaken convictions of politicians that put him in Iraq.
But I believe Eastwood’s purpose here is to depict a straight arrow in the fog of a questionable war. (A pivotal late scene takes place in a gathering sand storm that obliterates the frame in clouds of dust.) The soldier is true; the war — confused, bureaucratic — isn’t.
The film’s narrow perspective, centered on Kyle, is both the best and worst thing about it. “American Sniper” may be a much needed tribute to the sacrifice of American soldiers, but it’s lacking context. Few Iraqis here are seen as anything but the enemy.
“American Sniper,” a Warner Bros. release, is rated PG-13 by the Motion Picture Association of America for “strong and disturbing war violence, and language throughout including some sexual references.” Running time: 124 minutes. |
MR of vasculitis-induced optic neuropathy. PURPOSE To describe the MR characteristics of optic neuropathy caused by vasculitis. METHODS Nine cases of optic neuropathy with diagnosis of vasculitis (six with systemic lupus erythematosis and one each with rheumatoid arthritis, Sjgren disease, and radiation vasculitis) were reviewed retrospectively. Patients were 31 to 62 years old, and all but one were women. All patients had MR imaging through the orbits and anterior visual pathways, five with fat suppression, with and without gadopentetate dimeglumine. Five patients also had MR imaging of the entire brain. The size and enhancement of various segments of the optic nerve and anterior visual pathways were studied. RESULTS MR imaging with contrast material showed enhancement and enlargement of segments of the optic nerves and/or chiasm in six of the nine patients (all but three with systemic lupus erythematosis). Enlargement of a segment of the anterior visual pathway never occurred without enhancement, but enhancement alone did occur in three cases. Of the five patients who had MR imaging of the whole brain, abnormalities were seen in three: periventricular hyperintensity in two and a lacunar infarct in one; none had vessel abnormalities. CONCLUSION Because the MR enhancement seen represents disruption of the blood-brain barrier within the optic nerve, MR imaging with gadopentetate dimeglumine and fat suppression should be performed to detect increased permeability of the blood-brain barrier in acute optic neuropathy. |
<reponame>babtsoualiaksandr/ProjectAngular<filename>app/src/app/core/core.module.ts
import { NgModule } from '@angular/core';
import { CommonModule } from '@angular/common';
import { HomeComponent } from './home/home.component';
import { MaterialModule } from '../share/material/material.module';
import { from } from 'rxjs';
import { HeaderComponent } from './header/header.component';
import { SidebarComponent } from './sidebar/sidebar.component';
import { FooterComponent } from './footer/footer.component';
import { RouterModule, Routes } from '@angular/router';
const routes: Routes = [
{
path: '',
component: HomeComponent,
children: [
{
path: 'auth',
loadChildren: () =>
import('../auth/auth.module').then((m) => m.AuthModule),
},
{
path: 'storage',
loadChildren: () =>
import('../storage/storage.module').then((m) => m.StorageModule ),
},
],
},
{ path: '', redirectTo: '/', pathMatch: 'full' },
];
@NgModule({
declarations: [HomeComponent, HeaderComponent, SidebarComponent, FooterComponent],
imports: [
CommonModule,
MaterialModule,
RouterModule.forChild(routes),
],
exports: [
HomeComponent
]
})
export class CoreModule { }
|
Despite not playing each other since 1978, the 2015 Chick-fil-A Peach Bowl will be the 17th time that Houston and Florida State will play each other, the first time in bowl play. Houston leads the all-time series 12-2-2 with the Cougars victorious in the last matchup played in 1978 at Doak Campbell Stadium, the only game between the two in which Bobby Bowden was head coach for FSU.
The teams played 16 times from 1960-1978 with the two FSU victories coming in 1968 and 1975. Houston's fast pace offense was evident back then, as the Cougars scored at least 20 points in every meeting except for one between 1966 and 1978 (11 games). In 1968, the Houston Cougars put up 100 points the week prior to losing to the 'Noles, falling 40-20.
One of the two Seminoles victories were at Houston with the teams splitting the two neutral location games. Those neutral games were the site of each team's largest margin of victory with Houston winning in Tampa by 32 during the 1970 season, and FSU winning by 20 in Jacksonville in '68 with Bill Cappleman at the helm.
Since the last matchup, Houston has finished with a better regular season record than Florida State seven times, most recently the 2015 season. As the 'Noles prepare for their 36th bowl game since 1978, Houston will play in its 17th with a record of 5-11. The Cougars have never spent a week at the top of the AP Polls with the 'Noles receiving those honors for a total of 72 weeks, the sixth most in the nation.
New Year's Eve will be Houston's first appearance in the Peach Bowl with their last major bowl game coming in the 1984 Cotton Bowl when they lost to BC. Florida State will play in the Peach Bowl for the fourth time, winning two of their three games played.
The 2015 Chick-FIl-A peach bowl between the Florida State Seminoles and Houston Cougars will be played on New Years Eve from Atlanta, Georgia at 12 PM EST. |
Free trade agreements and economic territory as a geoeconomic imaginary in South Korea ABSTRACT Since the early 2000s, the discourse of economic territory has surfaced in conjunction with economic neoliberalization in South Korea. This paper argues that economic territory as a geoeconomic imaginary not only facilitated the expansion of free trade agreements as an accumulation strategy but also served as a hegemonic project which masked the nature of an accumulation strategy as a class project and consolidated political legitimacy by manipulating nationalism. To examine this linkage, it critically draws upon the idea of cultural political economy (CPE) developed by Lancaster-based sociologists Bob Jessop and Ngai-Ling Sum. This paper offers a fresh and more substantial interpretation of South Koreas political economy and opens up new analytical space for CPE. |
package com.fluper.printerdemo;
import android.content.Context;
import android.os.Bundle;
import android.os.CancellationSignal;
import android.os.ParcelFileDescriptor;
import android.print.PageRange;
import android.print.PrintAttributes;
import android.print.PrintDocumentAdapter;
import java.util.concurrent.ExecutorService;
import java.util.concurrent.Executors;
abstract class ThreadedPrintDocumentAdapter extends PrintDocumentAdapter {
abstract LayoutJob buildLayoutJob(PrintAttributes oldAttributes,
PrintAttributes newAttributes,
CancellationSignal cancellationSignal,
LayoutResultCallback callback,
Bundle extras);
abstract WriteJob buildWriteJob(PageRange[] pages,
ParcelFileDescriptor destination,
CancellationSignal cancellationSignal,
WriteResultCallback callback,
Context ctxt);
private Context ctxt=null;
private ExecutorService threadPool= Executors.newFixedThreadPool(1);
ThreadedPrintDocumentAdapter(Context ctxt) {
this.ctxt=ctxt;
}
@Override
public void onLayout(PrintAttributes oldAttributes,
PrintAttributes newAttributes,
CancellationSignal cancellationSignal,
LayoutResultCallback callback, Bundle extras) {
threadPool.submit(buildLayoutJob(oldAttributes, newAttributes,
cancellationSignal, callback,
extras));
}
@Override
public void onWrite(PageRange[] pages,
ParcelFileDescriptor destination,
CancellationSignal cancellationSignal,
WriteResultCallback callback) {
threadPool.submit(buildWriteJob(pages, destination,
cancellationSignal, callback, ctxt));
}
@Override
public void onFinish() {
threadPool.shutdown();
super.onFinish();
}
protected abstract static class LayoutJob implements Runnable {
PrintAttributes oldAttributes;
PrintAttributes newAttributes;
CancellationSignal cancellationSignal;
LayoutResultCallback callback;
Bundle extras;
LayoutJob(PrintAttributes oldAttributes,
PrintAttributes newAttributes,
CancellationSignal cancellationSignal,
LayoutResultCallback callback, Bundle extras) {
this.oldAttributes=oldAttributes;
this.newAttributes=newAttributes;
this.cancellationSignal=cancellationSignal;
this.callback=callback;
this.extras=extras;
}
}
protected abstract static class WriteJob implements Runnable {
PageRange[] pages;
ParcelFileDescriptor destination;
CancellationSignal cancellationSignal;
WriteResultCallback callback;
Context ctxt;
WriteJob(PageRange[] pages, ParcelFileDescriptor destination,
CancellationSignal cancellationSignal,
WriteResultCallback callback, Context ctxt) {
this.pages=pages;
this.destination=destination;
this.cancellationSignal=cancellationSignal;
this.callback=callback;
this.ctxt=ctxt;
}
}
}
|
Feb 12, 2014; Oakland, CA, USA; Golden State Warriors guard Jordan Crawford (55) controls the ball against Miami Heat guard Mario Chalmers (15) in the fourth quarter at Oracle Arena. The Heat defeated the Warriors 111-110. Mandatory Credit: Cary Edmondson-USA TODAY Sports
As the rebuild continues for the Miami Heat, who revamped the roster following a loss in the NBA Finals and of one LeBron James, plenty of veterans remain on the market.
Most notably, guard Jordan Crawford, wing Wayne Ellington and big man Andray Blatche are still looking for work. They come with their baggage, but have gotten significant playing time last season at positions of need for the Heat.
According to the South Florida Sun-Sentinel’s Ira Winderman, the Heat are likely to add one more veteran before the season and it could be one of those aforementioned guys.
Not only does/did he know that, but he apparently always knew they still would be available this late in the offseason, making it more of a buyers’ market for teams to dictate such additions on their own terms. As stated above, I still see the Heat adding the type of veteran who essentially would become a lock to make their roster.
The Heat figure to have 11 locks on the roster: The projected starting lineup of Mario Chalmers, Dwyane Wade, Luol Deng, Josh McRoberts and Chris Bosh as well as Norris Cole, Chris Andersen, Shabazz Napier, Danny Granger, Udonis Haslem and James Ennis.
That leaves four available roster spots for Reggie Williams, Shawne Williams, Shannon Brown, Justin Hamilton, Tyler Johnson and Khem Birch to compete for. Hardly a daunting bench crew. Hamilton figures to have an advantage because of his experience with the Heat, as do Reggie Williams (for his three-point shooting) and Brown (his handles).
Still, it seems Winderman believes Riley is letting the market play out in hopes of landing one of Crawford, Ellington or Blatche at the veteran minimum, since that is all he has left to offer.
Teams can carry up to 20 players in training camp but are limited to a maximum of 15 for the regular season. |
<reponame>SourceVortex/sourcevortex-gatsby-template<gh_stars>0
import styled from 'styled-components'
import { Link } from 'gatsby'
// Icons
import HomeIcon from '@material-ui/icons/Home'
import BookIcon from '@material-ui/icons/Book'
import MenuBookIcon from '@material-ui/icons/MenuBook'
import EditIcon from '@material-ui/icons/Edit'
// Configs
import device from '@Config/devices.config'
export const SidebarContainer = styled.div`
display: none;
align-items: center;
position: fixed;
width: 60px;
height: 100vh;
@media ${device.tablet} {
display: flex;
}
`
export const SidebarSubContainer = styled.div`
display: flex;
flex-direction: column;
justify-content: space-between;
align-items: center;
background-color: #111111;
border-radius: 0.3rem;
width: 40px;
height: 300px;
margin-left: 10px;
`
export const TopContainer = styled.div`
display: flex;
flex-direction: column;
align-items: center;
`
export const BottomContainer = styled.div`
display: flex;
flex-direction: column;
align-items: center;
`
export const StyLink = styled(Link)`
color: white;
text-decoration: none;
margin: 10px 0;
`
export const Anchor = styled.a`
color: white;
text-decoration: none;
`
export const StyHomeIcon = styled(HomeIcon)``
export const StyBookIcon = styled(BookIcon)``
export const StyMenuBookIcon = styled(MenuBookIcon)``
export const StyEditIcon = styled(EditIcon)``
|
Michael Dell, the founder and CEO of Dell, recently spoke to the Telegraph's Sophie Curtis about the PC market, something called the “post-PC era”, as well as the financial value and potential of data.
The “post-PC era” is a term Apple executives, both late Steve Jobs and the current CEO Tim Cook, like to use to describe the trend in which PC sales are dropping as laptops, smartphones, phablets and tablets take over.
Dell seems to believe there is no such thing, as the PC market, at least for his company, is still going strong.
“The post-PC era has been great for the PC. When the post-PC era started there were about 180m PCs being sold a year and now it’s up to over 300m, so I like the post-PC era,” he said.
He also talked about the cloud – a series of giant warehouses of data people access to use their favourite apps, do shopping, banking and stream music and communicate. Powerful computers can use the data cloud generates to crunch big numbers and help businesses create better products and services.
Mr Dell believes that this so-called “data economy” is the next big opportunity for the technology industry, which is why he is investing $67bn (£44bn) in data storage provider EMC — the biggest technology takeover in history.
“It used to be PCs and smartphones; now you have tablets and sensors and cars spinning out all kinds of telemetry data, and all kinds of machines and products and services becoming digital products,” he said. “Within all of those devices, you have a thousand times more applications, and the data is very rich. |
ALBUQUERQUE, N.M. — The long-planned bathroom or kitchen remodel, replacing drafty doors and windows, catching up on deferred maintenance or a needed repair – home-improvement projects that many homeowners have put off for years – are starting to trickle into the market.
“The biggest factor we’re experiencing, and what’s being experienced around the country, is the pent-up demand,” said Larry Chavez, president of Albuquerque-based Dreamstyle Remodeling.
Spending on home remodels nationwide is forecasted to jump by double-digit percentages in the first half of this year by the Remodeling Futures Program at the Joint Center for Housing Studies of Harvard University. There’s a good chance that remodeling activity in the Albuquerque metro area will follow suit.
“I think we’ll see an increase in the 10-12 percent range,” Chavez said.
Attendance jumped by about 30 percent at the Albuquerque Home & Remodeling Show on Nov. 18-19, from about 8,000 in 2013 to 10,500 this year, said David Griffin of sponsor Interact Productions. A fixture in Albuquerque for 22 years, attendance had dropped to the 6,000-7,000 range from 2008-12, he said.
“There’s no question a recovery is underway,” Griffin said.
Chavez, or more formally his company, was recently appointed to the more-than-50-member steering committee for the Remodeling Futures Program, joining a diverse list that includes federal agencies, trade associations, lenders and titans of the construction sector like Masco and Lowe’s Home Improvement.
“We are really honored, pleased and proud to be part of that program,” he said.
Adding comparatively smaller contractors like Dreamstyle Remodeling, which has six locations in three states, as a steering committee member reflects an effort to get more grass-roots, “in-the-trenches” involvement in the program, he said.
The program is best known for its Leading Indicator of Remodeling Activity, an index that estimates homeowner spending on improvements. The latest LIRA points to a 14 percent increase in home improvement spending in the first quarter and a 14.7 percent increase in the second.
The LIRA basically reflects the national economy. Home improvement spending peaked in mid-2007, the twilight of the housing bubble, then slumped steadily – sometimes at an alarming rate – through early 2011. Since then, it has been climbing back.
Based on building permits issued for “single-family alterations and additions,” Albuquerque has generally tracked with the LIRA with one major exception. In 2012, permits for home remodels dropped by 15 percent while the LIRA rose nearly 7 percent, likely a reflection of a lack of consumer confidence at the local level.
Permits in Albuquerque and the LIRA realigned in 2013, posting increases of 6.3 percent and 7.7 percent respectively. The year-end permit count of 1,024 is roughly where it was in the early 2000s before the housing bubble.
A lot of home-improvement projects do not require a building permit, although it often depends on the details, said Land Clark, the city’s chief building official.
A kitchen remodel, for example, doesn’t require a permit if it involves a straightforward replacement of old countertops, cabinets fixtures and appliances with new ones. If the project involves reconfiguring the kitchen space and moving, adding or eliminating electrical, plumbing or gas lines and outlets, then it does.
“If in doubt, the best thing to do at the city level is sketch it out and bring it down for us to look at,” Clark said.
In addition to consumer confidence, the pace of remodeling activity is fueled by increasing home values, said both Chavez and Jamie Baxter of Bain Cochran Construction, chair of the Remodelers Council in the Albuquerque metro area.
Consumer confidence and rising home values would appear to be intertwined, since a home is the biggest financial asset for the majority of U.S. households. Remodeling activity fell during years when homes were losing values pumped up by the buying frenzy of the bubble.
In 2012, the metro’s housing market eked out the first real gain in median price in five years, which likely contributed to the upswing in remodels last year. The market continued to see modest gains in median price during 2013, which could feed the forecasted upswing in remodeling activity in 2014.
2012 also marked the first time since 2005 that the number of home sales rose by a significant amount in the metro. Sales were up in 2013 as well. Harvard’s Remodeling Futures Program has identified homebuyers as top prospects for home-improvement projects.
“I have long found this to be true as people are anxious to personalize a new home that they have carefully chosen,” Chavez said, noting that the old maxim that new buyers are “tapped out” financially holds true some of the time, but not all of the time.
A brisker pace of sales also contributes to more prospective sellers fixing up their homes to put on the market, Baxter noted.
The loosening up of credit for home-improvement projects is widely recognized as a major impetus for increased spending on remodels. Home equity-type loans virtually disappeared when the housing bubble burst and property values went in decline.
The recent credit loosening is tied directly to the gradual recovery of home values which, in turn, lowers the lender’s risk in making home equity-type loans, he said. Some lenders, including First Mortgage, have introduced “rehab loans,” basically extra financing for repairs or a remodel, that are linked to first mortgages on home purchases, he said.
Other factors in the pent-up demand for home improvements include a shift in consumer sentiment away from sitting in limbo, postponing home improvements until uncertainty in the economy and housing clears, Baxter said.
Then there is the positive impact of the baby boom generation, born between 1946 and 1961, on remodeling activity, both Chavez and Baxter said. Chavez said most of his customers are aged 55 and over.
Tied to the desire by many older Americans to “age in place,” these remodels can range from ramps, railings and wheelchair modifications to lifestyle or aesthetics-driven remodels. Improvements to outdoor space and the indoor space next to it are becoming more popular, Baxter said. |
Quinacrine promotes autophagic cell death and chemosensitivity in ovarian cancer and attenuates tumor growth. A promising new strategy for cancer therapy is to target the autophagic pathway. In the current study, we demonstrate that the antimalarial drug Quinacrine (QC) reduces cell viability and promotes chemotherapy-induced cell death in an autophagy-dependent manner more extensively in chemoresistant cells compared to their isogenic chemosensitive control cells as quantified by the Chou-Talalay methodology. Our preliminary data, in vitro and in vivo, indicate that QC induces autophagy by downregulating p62/SQSTM1 to sensitize chemoresistant cells to autophagic- and caspase-mediated cell death in a p53-independent manner. QC promotes autophagosome accumulation and enhances autophagic flux by clearance of p62 in chemoresistant ovarain cancer (OvCa) cell lines to a greater extent compared to their chemosensitive controls. Notably, p62 levels were elevated in chemoresistant OvCa cell lines and knockdown of p62 in these cells resulted in a greater response to QC treatment. Bafilomycin A, an autophagy inhibitor, restored p62 levels and reversed QC-mediated cell death and thus chemosensitization. Importantly, our in vivo data shows that QC alone and in combination with carboplatin suppresses tumor growth and ascites in the highly chemoresistant HeyA8MDR OvCa model compared to carboplatin treatment alone. Collectively, our preclinical data suggest that QC in combination with carboplatin can be an effective treatment for patients with chemoresistant OvCa. INTRODUCTION Ovarian cancer remains a leading cause of death among women with gynecological cancers despite significant advances in the systemic as well as surgical cancer treatment modalities. Ovarian cancer patients with advanced or recurrent disease frequently develop chemoresistance against paclitaxel-and platinum-based therapies which further contributes to disease progression, recurrence and, ultimately, high mortality. In order to overcome the shortcomings of standard chemotherapeutic modalities, several alternative strategies including targeted therapies, multi-drug combination treatments as well as drug repurposing are being investigated. Although the majority of these therapeutic compounds induce apoptosis through type I programmed cell death (PCD), compounds inducing type II PCD have also been shown to be effective as anti-cancer agents. Predominant feature of type II or autophagic cell death is the appearance of doublemembrane vesicles engulfing cytoplasmic organelles which are eventually degraded by lysosomal hydrolytic enzymes. Extensive autophagic activity leads eventually to cell death which, however, differs from the homeostatic autophagy associated with normal cellular organelle turnover. Under basal cellular conditions, autophagy maintains the cellular turnover of proteins and organelles via lysosomal degradation whereas, under nutrient-deprived stress conditions such as oxidative and/or endoplasmic reticulum stress, it promotes cellular adaptation by supplying macromolecules for survival. Various strategies have been investigated to explore the potential of autophagy as a putative anticancer modality including development of www.impactjournals.com/oncotarget chemical inhibitors of autophagy as well as genetic silencing of key autophagy proteins. Several studies have shown that inhibiting autophagy using anti-malarial compounds such as chloroquine (CQ) and hydroxychloroquine while combining these compounds with frontline therapeutic agents such as cisplatin and taxol results in significant inhibition of tumor growth. Furthermore, other studies have indicated that drug-induced autophagy promotes synergy with the frontline therapy. Similarly it has been shown that genetic silencing of key autophagic proteins such as beclin 1 (ATG6) favors survival and decreases resistance to chemotherapy. High beclin 1 and LC3 levels in ovarian tumors have been associated with improved overall survival. The two most common markers and key players associated with autophagy are LC3B and p62. Events leading to the conversion of LC3BI to LC3BII and clearance of p62 are considered hallmarks of the autophagic flux. The p62 protein, also called sequestosome 1 (SQSTM1), is an ubiquitin-binding scaffold protein that co-localizes with ubiquitinated protein aggregates and is required both for the formation as well as the degradation of polyubiquitin-containing bodies by autophagy. p62 binds to LC3B through the LIR (LC3 Interacting Region) domain and is then degraded during the autophagic process. Other studies have shown that elimination of p62 by autophagy suppresses tumorigenesis in vivo and cell viability of several human carcinoma cell lines in vitro. Since p62 accumulates when autophagy is inhibited, and alternatively, p62 levels decrease when autophagy is induced, p62 surfaces as a promising marker to study autophagic flux. Selective degradation of p62 is clinically relevant since high levels of p62 found in various types of tumor have been associated with poor prognosis and survival. Studies show that the cisplatin-resistant SKOV3/DDP OvCa cells express higher levels of p62 and that siRNA downregulation of p62 in these cells resensitized them to cisplatin-mediated cytotoxicity. Previous investigations have provided evidence to suggest a promising role of the antimalarial drug quinacrine (QC) in cancer treatment. The acridine "backbone" of QC allows the drug to intercalate into stacked DNA base pairs. QC is known to impair DNA repair activity in a mechanism similar to other topoisomerase inhibitors,. In addition, QC inhibits the FACT (Facilitates Chromatin Transcription) complex that is required for NF-kB transcriptional activity and modulates the arachidonic acid (AA) pathway. Interestingly, QC has been shown to bind and inhibit proteins involved in multidrug resistance. More importantly, it targets several signaling pathways simultaneously by affecting autophagy, apoptosis, p53, NFkB, AKT and methylation-related pathways [27,28,. While QC has been shown to modulate autophagy in a p53-dependent manner in colon cancer cell lines, in our study QC induced autophagic cell death in a p53-independent manner in OvCa cells. Although QC has been shown to effectively block proliferation of several cancer cell lines both in vitro and in vivo, to our knowledge, there are no in vitro or in vivo studies on the use of QC alone or in combination with standard therapy against OvCa. In this study, we have shown that QC promotes autophagic flux across a variety of OvCa cell lines and induces cell death both in a caspase-dependent as well as independent manner utilizing autophagic-mediated cell death to enhance carboplatin sensitivity. This effect was more pronounced in cisplatin-resistant OvCA cells compared to their sensitive controls both in vitro and in vivo experimental setting. These preclinical data have direct clinical implications for OvCa patients with chemoresistant disease for which only limited therapeutic options exist. In this study, we focused our investigation on the anticancer potential of the antimalarial drug QC against OvCA. Based on prior findings, we hypothesized that QC would exert its anticancer effect against OvCA by inducing an autophagic-mediated cell death and that by doing so it would result in restoring cisplatin-sensitivity. Quinacrine inhibits cell growth and induces cell death in ovarian cancer cells Isogenic pairs of OvCA cell lines were evaluated for the effect of QC on cell growth by colony formation and MTT assays. Colony formation assays ( Figure 1A) were performed after treating the cells with 0, 0.125, 0.250, 0.500, 1.0, and 2.0 M of QC for 24 hours. MTT assays (supplemental data) were performed after treating the cells with 0, 5.0, and 10.0 M of QC for 24, 48 and 72 hours-time intervals. Increasing concentrations of QC effectively inhibited colony forming units with maximal inhibition at a QC concentration of 1.0 and 2.0 M. Similarly, cell growth was also inhibited as early as 24 hours of QC treatment with IC 50 determined from the MTT assays in all the cell lines tested were between 2.5 M and 4 M (Figure supplemental S1). To determine if QC treatment induced apoptotic cell death, we treated cells with 2.5, 5.0 and 7.5 M QC for 24 hours and the apoptotic cell population was determined with the annexin/PI staining method using flow cytometric analysis. The apoptotic cell population upon QC treatment reflecting early and late apoptosis as shown in Figure 1B indicates that QC only treatment induces apoptosis. Similarly, western blot analyses of cell lysates of OV2008/ C13 and Hey A8/HeyA8MDR cells treated with 5.0 and 10 M QC showed the presence of cleaved PARP corroborating the previous finding that QC promotes apoptosis in a caspase-dependent manner ( Figure 1C). Quinacrine induces autophagic clearance of p62/SQSTM, upregulates the autophagic marker LC3B and induces apoptosis Since other antimalarial drugs have previously been shown to modulate autophagy, we tested whether QC is able to induce autophagy in addition to promoting apoptosis. Towards this end, OV2008/C13 as well as HeyA8/HeyA8MDR cells were treated with 5.0 and 10.0 M QC for 24 hours. Western blot analysis of QC-treated cell lysates was performed using two different autophagic marker proteins, LC3B and p62. Figures 2A and 2B show induction of the lipidated form of LC3B-II in all four cell lines upon QC treatment. However, QC treatment reduced p62 levels significantly more in the chemoresistant (C13 and HEYA8MDR) cells compared to their sensitive counterparts. This induction of LC3B-II (lower band) and degradation of p62 are considered hallmarks of autophagic activation. QC treatment resulted in similar effects on p62 and LC3B independent of p53 status in high grade serous cell lines such as OVCAR3 (p53 mutant) and CAOV3 (p53 null) (Figure Supplemental S2). To further evaluate QC-induced autophagy, we utilized transmission electron microscopy (TEM). TEM analysis of QC-treated OV2008, C13, HeyA8 and HeyA8MDR cell revealed early autophagic bodies (autophagosomes) harboring intact organelles ( Figures 2C and 2D). Quantitation of autophagosomes is shown next to the respective figures. These data suggest that QC treatment induces autophagy. While TEM analysis revealed qualitative and morphological features of autophagy upon QC treatment, it was unclear whether this represented increased generation of autophagosome or inhibition of autophagosomal maturation. Therefore, to further clarify whether QC induces autophagic flux, OV2008 and C13 cells were treated with QC in the presence or absence of a potent late-autophagy inhibitor, bafilomycin A1, and then stained with the autophagolysosome-specific dye cyto-ID. Cyto-ID retention in the cytoplasm was detected by flow-cytometric analyses ( Figure 2E). This data indicates that QC's ability to promote autophagy can be completely inhibited by co-treatment with bafilomycin A1. QC induces autophagic flux and QC-induced autophagy precedes apoptosis To further evaluate autophagic flux, we determined whether QC-mediated upregulation of LC3BII and downregulation of p62 was affected by co-treatment with bafilomycin A. Western blot analysis revealed that QC effectively downregulated p62 expression in C13 and HeyA8MDR and co-treatment with bafilomycin A protected autophagic p62 degradation ( Figure 3A). No change in p62 mRNA levels was detected upon QC treatment (data not shown). We next tested whether QC-mediated apoptosis is dependent on autophagy. OV2008, C13, HEYA8 and HEYA8MDR cells were treated with QC with or without bafilomycin A1 followed by annexin/PI staining and flow cytometric detection of apoptotic cells. QC treatment in OV2008, C13, HeyA8 and HeyA8MDR cells resulted in varying but significant increase in annexin V positive cells. Co-treatment with bafilomycin A completely abolished QC-mediated annexin V positivity in these cells as shown in Figure 3B. Consistent with the flow cytometric data, immunoflourescence analysis also confirmed that the downregulated p62 by QC were rescued by bafilomycin treatment in C13 cells (Figure supplemental S3). This data indicates that QC-promoted cell death is completely reversed by autophagic inhibitors in all the four cell lines tested. Taken together, this data suggests that QC induces apoptosis in an autophagic-dependent manner. Our data indicates that QC-induces apoptosis and autophagic inhibition by bafilomycin A. Therefore, we next sought to determine whether autophagy preceded the QC-induced apoptotic response. To this end, we determined the temporal regulation of autophagic proteins LC3B, p62 and apoptotic proteins such as cleaved PARP upon QC treatment at set time points in C13 and HeyA8MDR OvCa cells by western blot analysis. Data in Figure 3C indicate that QC treatment induced p62 downregulation and LC3B upregulation as early as 4 hours post treatment and these levels were sustained up to 24 hours following treatment. Detection of cleaved PARP was minimal at 4 hours of treatment; however its expression increased significantly at 16 hours post QC treatment. This data suggests that QC treatment triggered an autophagic response at the earlier time points while this autophagic response coincided with an apoptosis rate at later time points it. QC synergizes with carboplatin in ovarian cancer cell lines After observing increased degree of apoptosis upon QC treatment in chemoresistant OvCa cells, we then sought to investigate whether QC can synergize with carboplatin treatment. In order to determine whether QC synergizes with cisplatin and carboplatin, constant ratio synergy studies were performed in isogenic cisplatinsensitive OV2008 and cisplatin-resistant C13 cells by treating them with 1 IC 50 of cisplatin in combination with 1 IC 50 of QC. We observed that QC had a more potent synergistic anti-proliferative effect in vitro when combined with either cisplatin in C13 compared to OV2008 cells ( Figures 4A and 4B). The combination indices (CI) for the corresponding fractions affected (FA) are shown in the tables below the figures. CI values between 0.1-0.3 indicate extremely strong synergism, 0.3-0.7 strong synergism, 0.7-0.85 moderate synergism, 0.85-0.9 slight synergism and 0.9-1.0 a nearly additive effect. Of note, synergy was demonstrated across nearly the entire range of the drug concentrations. Similar studies with isogenic taxol-sensitive SKOV3 and taxol-resistant SKOV3TR cells indicate that QC has a more synergistic antiproliferative effect in vitro when combined with either cisplatin or carboplatin in SKOV3TR ( Figures 4C and 4D) in equipotent combinations (IC 50 over IC 50 ratio) was assessed for synergy using the Chou-Talalay method. The cells were exposed to each drug alone and in combination per protocol for 48 hours. The combination indices (CI), fraction affected (Fa) in OV2008 and C13 A and B. in SKOV3 and SKOVTR C and D. and in Hey A8 and HeyA8MDR E and F. were generated by the Calcusyn software and plotted with the use of GraphPad. Combination index (CI) values at 25, 50, 75 and 90% fraction affected (FA) are presented in the tables below the graphs. CI between 0.3-0.7 indicates strong synergism, 0.7-0.85 moderate synergism, 0.85-0.9 slight synergism, 0.9-1.10 nearly additive effect and greater than 1.10 antagonism. www.impactjournals.com/oncotarget QC-induced autophagy is required for sensitization to cisplatin-mediated cell death in OvCa cells In order to test whether QC-induced autophagy is essential in order to sensitize OvCa cells to cisplatininduced cytotoxicity, we assessed the effects of the combination of increasing concentrations of cisplatin with QC (1 IC50) with and without bafilomycin A pretreatment in OV2008 and C13 cells. We hypothesized that by inhibiting autophagy, bafilomycin A will inhibit the ability of QC to synergize with cisplatin in inducing autophagic cell death. The cells were pretreated with 50 nM of bafilomycin for 2 hours followed by treatment with cisplatin and QC. Cell viability was assessed by MTT assays 48 hours later. As shown in figures 5A and 5B, pretreatment with bafilomycin A (green dotted line) inhibited the combined QC-and cisplatin-induced cytotoxicity (blue dotted line). This is also demonstrated by the change in the CI values from values indicating strong synergy in C13 (CI = 0.690) and nearly additive effect in OV2008 (CI = 1.054) without bafilomycin to indicating antagonism in both C13 (CI = 1.242) and OV2008 (CI = 1.396) after bafilomycin pretreatment ( Figure 5C). Collectively, these data suggest that QC-induced autophagy is necessary for resensitization to cisplatininduced cell death in OvCa cells. p62 knockdown enhances sensitivity to carboplatin treatment in HeyA8MDR and C13 ovarian cancer cells Elevated levels of p62 have been previously shown to be critical in imparting chemoresistance in OvCA cells. Our data indicate that QC treatment downregulated p62 levels preferentially in the chemoresistant C13 and HeyA8MDR cells (Figure 2A). Previous studies have shown that p62 downregulation sensitizes cells to cisplatin-mediated cytotoxicity. To determine whether p62 plays role in QC-and carboplatin-mediated apoptosis, we generated two different p62 knockdown shRNA clones in HeyA8MDR and C13 cells as described in Materials and methods section with non-targeted control transduced cells (NTC) as controls. Efficient knockdown of p62 was confirmed in C13 and HeyA8MDR cells by western blot analysis using anti-p62 antibody ( Figures 6A and 6B). To further evaluate the effect of QC in inducing apoptosis in NTC (*P = < 0.001, **P = < 0.0001). E. HEYA8MDR NTC and sh280 cells were treated with different concentrations of carboplatin and cell viability was measured by MTT assay. sh280 cells showed more sensitivity towards carboplatin treatment compared to the NTC group (**P < 0.0001, Figure 6E) indicating the critical role of p62 as a determinant of chemoresistance in HeyA8MDR cells. and p62-depleted cells, we treated C13 NTC and p62 shRNA clones with 0, 5.0, 10.0 M of QC for 24 hours. Evaluation of apoptotic marker proteins by western blot analysis revealed that p62 knockdown cells exhibited higher degree of cleaved PARP and caspase 3 whereas no significant change was observed in LC3B II induction upon QC treatment in C13 NTC as well as p62shRNA cells ( Figure 6A). Consistent with this data, QC treatment in HeyA8MDR p62shRNA cells showed increased degree of caspase 3 and PARP cleavage while no change was detected in LC3B II induction ( Figure 6B). This data suggests that p62 downregulation in C13 and HeyA8MDR sensitizes the cells to QC treatment. Similarly, annexin/PI staining of these cells after treatment with QC revealed that C13 and HeyA8MDR p62 knockdown cells were more sensitive to QC-induced cell death when compared to NTC cells ( Figures. 6C and 6D). More importantly, genetic downregulation of p62 in HeyA8MDR cells enhanced carboplatin sensitivity (please note the reduction in carboplatin IC50 from 176 M in NTC cells to 110 M in p62 knockdown cells) ( Figure 6E). Collectively, these findings indicate that p62 downregulation sensitizes cells to QC-mediated cell death. Quinacrine synergizes with carboplatin in reducing HeyA8MDR derived mouse tumor xenografts QC effectively blocked cell growth, induced autophagy and apoptosis in OvCa cell lines in vitro. We next tested whether QC in combination with carboplatin is effective in attenuating tumor growth in vivo. For this purpose we utilized HeyA8MDR cells that have capacity to form tumors when injected intraperitoneally in nude mice. The effect on tumor growth of QC alone and in combination with carboplatin (CBP in the Figure 6) was evaluated in HeyA8MDR in female nude mouse xenografts. One week after 310 6 HeyA8MDR cells were injected intraperitoneally in the mice, they were randomized into four groups of 10 mice when the tumors were palpable and then treated as shown in Figure 7A. Quinacrine (150 mg/kg body weight) was given as described in the materials and methods section. Quinacrine alone effectively reduced tumor volume and ascites formation ( Figures 6B, 6C and 6D). Although carboplatin treatment similarly reduced tumor volume, it was associated with accumulation of ascites to an extent even greater than untreated controls ( Figure 6D). Assessment of tumor weight and ascites volume as measured at the time of necropsy across treatment groups showed that combination treatment was more effective in reducing cancer progression compared to all other treatment groups ( Figures 7B, 7C and 7D). Staining of the tumors with Ki-67 showed significantly reduced proliferation in QC and QC plus carboplatin treated groups, highlighting the ability of QC to reduce tumor cell proliferation in vivo ( Figure 7E). There was no significant body weight loss in the QC or combination treatment groups compared to the untreated control group ( Figure 7F). TEM micrographs of the xenografts showed more autophagosomes (AV in red, Figure 7G) in the QC group compared to untreated and carboplatin groups. Combination treatment resulted in significantly more autophagosomes compared to QC alone ( Figure 7H). H&E staining of the livers from all four groups ( Figure 7I) showed that there was no difference in histology, suggesting that neither QC monotherapy nor combination treatment was toxic to the mice. These data demonstrate that QC plus carboplatin combination treatment leads to significantly enhanced antitumor activity in an OvCa mouse model. DISCUSSION The lack of effective treatment modalities for patients with chemoresistant disease continues to pose a therapeutic challenge in ovarian cancer. There is a growing need to identify therapeutic compounds which could promote efficacy to frontline therapeutic agents such as cisplatin and taxol. In this study, we evaluated the anticancer properties of the anti-malarial drug QC and tested its effectiveness in combination with carboplatin in OvCa. QC treatment effectively caused apoptosis and promoted an autophagic response across a variety of ovarian cancer cell lines in vitro as well as in vivo in mouse-derived HeyA8MDR xenografts. QC treatment caused robust autophagosome formation, LC3B accumulation and p62 downregulation all of which are hallmarks of autophagy. Our findings suggest that QCmediated induction of autophagy preceded the induction of apoptosis as the effects on the autophagic proteins LC3B and p62 were observed at earlier time points. Autophagy and apoptosis have been previously shown to coincide temporarily in order to cause drug-induced cell death. In agreement with these findings, we showed that by chemically blocking autophagy using bafilomycin A, QC-mediated apoptosis was significantly attenuated. At the molecular level, QC promoted autophagic clearance of p62 in OvCa cell lines. p62 is a well-known substrate of autophagy and high levels of p62 have previously been associated with chemoresistance. Genetic silencing of p62 expression by shRNA in the chemoresistant HeyA8MDR cancer cells promoted apoptosis and enhanced sensitivity to carboplatin ( Figure 6E). Therefore, autophagic clearance of p62 expression by QC might be one of the underlying mechanisms responsible for enhanced synergy when QC is combined with carboplatin. Several studies have shown that p62 expression is elevated in breast, pancreatic, colon as well as ovarian cancer which is consistent with our observation of increased expression of p62 in the chemoresistant cell lines that we studied, namely the C13 and HeyA8MDR cell line compared to their chemosensitive counterparts OV2008 and HeyA8. QC treatment down regulated p62 expression in an autophagic-dependent manner in the C13 and HeyA8MDR OvCa cell lines. Lentiviral shRNA-mediated p62 depletion in the OV2008 and HeyA8MDR OvCa cell lines resulted in enhanced synergy between QC and carboplatin. This data suggests that combination treatment with carboplatin and QC results in an enhanced QC-mediated clearance of p62 via autophagy ultimately favoring apoptosis. This observation is further supported by our data showing that inhibition of autophagy by bafilomycin A not only effectively blocked QC-mediated p62 downregulation but also apoptosis. Being a lysosomotropic agent, it is not surprising that QC appears to be implicated in autophagy; however, our study highlights the fact that p62 downregulation plays a key role in QC-mediated apoptosis in OvCa cells and provides a description of the mechanistic interplay between autophagic and apoptotic cell death upon QC treatment. QC treatment of mouse-derived xenografts was shown to be effective in reducing tumor weight, reducing cell proliferation as measured by Ki-67 staining and blocking ascitic fluid formation. While a similar degree of tumor reduction was achieved by carboplatin treatment alone, in contrast to QC treatment, carboplatin treatment induced ascitic fluid accumulation. TEM analysis of QC-treated xenografts clearly showed increased formation of autophagosomes and autolysosomes. These effects were dramatic when QC was combined with carboplatin leading to complete remission of tumor growth and reduction in the proliferation index. This data is in accordance with the high degree of synergy observed when chemoresistant OvCa cell lines were treated with carboplatin and QC in vitro. Taking into consideration the previously proven ability of carboplatin to upregulate p62 expression along with the knowledge that carboplatin-resistant OvCa cells exhibit increased levels of p62, we hypothesize that co-administration of QC with carboplatin may mitigate p62 expression in vivo thereby enhancing the degree of apoptosis. QC has been shown to stabilize p53, inhibit NF-kB, and cause cell cycle arrest. However, QC-mediated p53 stabilization has been associated with its Nf-kB suppressive activities and not due to genotoxic stress. Our findings revealed that QC induced robust autophagic cell death and synergized with carboplatin treatment. It is likely that multiple cellular pathways including p53 and NF-kB in addition to autophagic p62 downregulation are playing a role in QC-and carboplatin-mediated synergy. Other studies have also demonstrated the synergy between QC and other chemotherapeutic drugs such as Lycopene in breast cancer, cedarinib in glioma cells, vincristine in an MDR sub-clone of K562 cells as well as other therapeutic compounds. More recently, quinacrine has been shown to reverse erlotinib resistance in non-small lung cancer cells by targeting FACT complex and NF-kB activities. It is important to note that several reports have implicated autophagy as a survival mechanism in cancer cells and have demonstrated that by inhibiting autophagy with chloroquine, tumor growth can be arrested. While this is true at a tumor-specific level, we believe that prolonged chemical inhibition of autophagy with different autophagic inhibitors might elicit differential responses than just simply inhibition of autophagy. Notably, chloroquine, in addition to inhibiting autophagy, also stabilizes p53 thereby leading to apoptosis. Similarly, another autophagy inhibitor, 3-MA, has been shown to inhibit Akt activation. These observations in conjunction with our data indicate that drugs altering autophagy may also influence apoptosis. QC unlike Bafilomycin A promoted p62 degradation as a function of autophagy induction and also promoted apoptosis both in vitro and in vivo. In summary, the present study demonstrated that QC treatment resulted in autophagic clearance of p62, promoted apoptosis and effectively synergized with carboplatin in vivo. QC's ability to inhibit multiple pathways simultaneously in addition to promoting autophagic cell death offers compelling evidence to support the hypothesis that it may represent an important adjunct to standard treatment against ovarian cancer especially in patients with chemoresistant disease. Cell lines, chemicals and antibodies The human OvCa cell lines SKOV3, SKOV3 TR, HeyA8 and HeyA8 MDR were obtained on an MTA from MD Anderson Cancer Center, Houston, TX. C13 and OV2008, were obtained on a MTA from Dr. Barabara Vanderhyden (Ottawa Hospital Research Institute, Ottawa, Canada). OVCAR3 and CAOV3 were obtained from the American Type Culture Collection (ATCC) (Manassas, VA). All cell lines were cultured according to the providers' recommendations at5% CO 2 and at 37°C. Quinacrine was obtained from Sigma-Aldrich. MTT dye was obtained from Promega. Carboplatin was purchased from Calbiochem (San Diego, CA). Anti-LC3B, anti-PARP, anti-p62, anti-PDI and anti GAPDH antibodies were purchased from Cell Signaling Corporation. Anti-NBR antibody was purchased from Genetex. Anti-p53 (DO-1) antibody is from Santa Cruz Biotech. Colony formation assay 500 cells were plated in 6-well plates and treated with increasing concentrations of QC (0.12, 0.25, 0.5, 1.0 1nd 2.0, 2.0 M) for 24 hrs. The media was replaced after day1. Colonies were fixed in methanol and stained with 0.5% crystal violet on day 8 and counted using a colony counting software, Quantity One (Bio-Rad). Each treatment was carried out in triplicate and repeated twice. www.impactjournals.com/oncotarget Cytotoxicity assay MTT assay was performed in order to assess the effect of QC on OvCa cell lines. Ten thousand cells from each cell line were treated with various concentrations of QC and carboplatin separately or in combination for 48 hours followed by a 4-hour period of incubation with 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT). The violet formazan crystals were dissolved in dimethyl sulfoxide (DMSO) and the absorbance was measured at 490 nm in a microplate reader. ShRNA transductions The control shRNA, pTRC1-NTC (non-target shRNA vector, Sigma) contains a hairpin insert that will generate siRNAs but contains five base pair mismatches to any known human gene. shRNA for p62/SQSTM denoted as p62-sh280 was purchased from Sigma and with a sequence CGAGGAATTGACAATGGCCAT targeting cDNA region of p62. Lentivirus particles were produced by transient transfection of pTRC1-NTC and pTRC1-p62 shRNA along with packaging vectors (pVSV-G and pGag/pol) in 293T cells. The lentiviral supernatants were collected 48 hours after transduction, filtered and either used for infection or stored at -80°C. Vector titers were determined by transducing OvCa cells with serial dilutions of concentrated lentivirus, in complete growth medium containing 8 g/ml polybrene (Invitrogen). After 7 days, the growth medium was supplemented with puromycin (2 g/ml) for selection. The surviving colonies were counted under the microscope and the titer of lentiviral stocks was calculated using the formula: Transducing units = number of colonies lentiviral dilution. All lentiviral stocks used in the study were selected at a multiplicity of infection of 10. MTT assay, synergy assessment and Chou-Talalay calculations Cell lines were treated with a wide range of concentrations of QC and cisplatin for 48 hours and the half maximal inhibitory concentration (IC 50 ) of each drug alone was derived experimentally by MTT assay as previously described and calculated by Prism (GraphPad Software, La Jolla, CA). Subsequently, drug combination studies (QC with cisplatin and QC with carboplatin) were performed and their synergy was quantified using the Chou-Talalay method as previously described in the literature. Both constant ratio synergy (ratio of drugs 1:1) studies were carried out. Synergy was assessed by creating combination indices (CI): CI values less than 0.9 indicate synergism, CI values between 0.9 to 1.1 indicate nearly additive effect and CI values greater than 1.1 indicate antagonism. Flow cytometric analysis of apoptosis and autophagy C13 and OV2008 cells were treated with the indicated doses of QC, then harvested and resuspended in a binding buffer. Cells were then stained with Annexin V-pacific blue and Propidium iodide (BD bioscience) according to the manufacturer's instructions. Cells were analyzed in a Beckman Coulter EPICS XL/MCL flow cytometer (Beckman-Coulter Fullerton, CA, USA) and the data were analyzed with Flowjo Software (Tree Star, Ashland, OR, USA). Cyto-ID TM (Enzo Life Sciences) is a dye specifically labeling the autophagic vacuoles in a cell by colocalizing with LC3B. After treating the cells with various concentrations of QC, cells were collected and stained using Cyto-ID dye according to the manufacturer's instructions. Cells were analyzed as previously described in a Beckman Coulter EPICS XL/MCL flow cytometer (Beckman-Coulter Fullerton, CA, USA) and the data were analyzed with Flowjo Software (Tree Star, Ashland, OR, USA). Western blot analysis Western blot analysis was performed as previously described. Transmission electron microscopy Cells (2 10 6 ) were treated with QC and then harvested and centrifuged at 1200 r.p.m. for 5 minutes. Cell samples were then pre-fixed with Trumps buffer. The images were taken using a Philips 208S electron microscope (FEI Corporation, Eindhoven, Netherlands). Immunofluorescence Cells were plated on coverslips and treated with QC as indicated in materials and methods. Cells were fixed with 100% methanol followed by blocking with 1% BSA in PBS. Cells were then incubated with the corresponding primary antibodies at room temperature for 1 hour followed by three washes with 1X PBS and then incubated in the dark with Alexa fluor rabbit anti-mouse (593 nm) in 1% BSA in PBS. Coverslips were washed three times before mounting with Prolong Gold Antifade reagent (Invitrogen). Stained samples were visualized using a Zeiss-LSM 510 fluorescence microscope. Animal studies Athymic nude mice were purchased from Harlan. 3 10 6 HEYA8MDR cells were injected intraperitoneally. Treatment was initiated after one week following the intraperitoneal inoculation of the cells. Stock solutions of QC (Sigma) were prepared in sterile water and administered by oral gavage. Carboplatin (Hospira pharma) were injected intraperitoneally. Before initiation of treatment, animals were randomly assigned to one of four groups (10 mice per group). The control group received water by oral gavage the QC only group received 150 mg/kg body weight of QC every other day starting on day 7, the carboplatin group received carboplatin by intraperitoneal injection at a dose of 50 mg/kg body weight) on days 7, 14, 21 and 28 and the combination group received carboplatin by intraperitoneal injection at days 7, 14, 21 and 28 and QC at 150 mg/kg body weight starting on day 7, then every other day until the end of the study. All animals were sacrificed on day 28. Animal care and procedures was conducted according to institutional policies and the experimental protocol was approved by the IACUC committee of our Institution. Statistical analysis All results are expressed as mean ± standard deviation (S.D.). Data were obtained from three independent experiments. All statistical analyses were conducted using Graph pad Prism software (San Diego, CA). Data were analyzed using paired t test, and P values less than 0.05, unless mentioned otherwise, were considered statistically significant. |
Multitype branching and graph product theory of infectious disease outbreaks. The heterogeneity of human populations is a challenge to mathematical descriptions of epidemic outbreaks. Numerical simulations are deployed to account for the many factors influencing the spreading dynamics. Yet, the results from numerical simulations are often as complicated as the reality, leaving us with a sense of confusion about how the different factors account for the simulation results. Here, using a multitype branching together with a graph tensor product approach, I derive a single equation for the effective reproductive number of an infectious disease outbreak. Using this equation I deconvolute the impact of crowd management, targeted testing, contact heterogeneity, stratified vaccination, mask use, and smartphone tracing app use. This equation can be used to gain a basic understanding of infectious disease outbreaks and their simulations. |
def init_wiki_repo(self):
repo_dir = os.path.join(MKDOCS_DIR, self._wiki_name)
repo = self.init_repo(repo_dir)
origin = repo.remotes.origin
origin.pull
os.chdir(repo_dir) |
/**
* A BridgeHandler will be passed in a BridgeCallBack object if the Javascript requester has asked for a response
*/
public class BridgeCallBack {
private BridgeManager bridgeManager;
private String successHandlerName, failureHandlerName;
public BridgeCallBack(BridgeManager bridgeManager, String successHandlerName, String failureHandlerName) {
this.bridgeManager = bridgeManager;
this.successHandlerName = successHandlerName;
this.failureHandlerName = failureHandlerName;
}
public void successHandler(String jsonString) {
bridgeManager.callBack(successHandlerName, jsonString);
}
public void failureHandler(String jsonString) {
bridgeManager.callBack(failureHandlerName, jsonString);
}
} |
#ifndef SCRIPT_CONTENT_HPP
#define SCRIPT_CONTENT_HPP
/*
script_content.hpp
2016/08/03
<EMAIL>
Description: Data container for processing and handling scripts from PMD2.
*/
#include <ppmdu/pmd2/pmd2.hpp>
#include <ppmdu/pmd2/pmd2_configloader.hpp>
#include <cstdint>
#include <string>
#include <map>
#include <unordered_map>
#include <vector>
#include <deque>
namespace pmd2
{
//==========================================================================================================
// Constants
//==========================================================================================================
/**********************************************************************************
Script Type
**********************************************************************************/
enum struct eScriptSetType
{
INVALID,
UNK_unionall, //Scripts not in a LSD, found under the "COMMON" directory. Has no other component.
UNK_enter, //Scripts for accessible levels.
UNK_acting, //Scripts mainly for cutscenes
UNK_station, //Scripts not in a LSD, file extension of data file is SSS, may have matched numbered ssb.
NbTypes,
};
/*
Script Data Type
For identifying the purpose of a script data file.
*/
enum struct eScrDataTy
{
SSE,
SSS,
SSA,
NbTypes,
Invalid,
};
extern const std::array<std::string, static_cast<size_t>(eScrDataTy::NbTypes)> ScriptDataTypeStrings;
const std::string & ScriptDataTypeToFileExtension( eScrDataTy scrdatty );
const std::string & ScriptDataTypeToStr( eScrDataTy scrdatty );
eScrDataTy StrToScriptDataType( const std::string & scrdatstr );
/*
eInstructionType
This is meant to help determine how to process and format some instructions,
and differentiate between instructions we've added solely to serves our needs,
and actual instructions to be placed in the target file.
*/
enum struct eInstructionType : uint8_t
{
Command, //For most instructions
//Meta Labels
MetaLabel, //A jump label
MetaCaseLabel, //A jump label for a conditional case //#REMOVEME
//Meta instructions
MetaSwitch, //The instruction contains case sub-instructions
MetaAccessor, //The instruction contains a sub-instruction to be applied to what the accessor is accessing
MetaReturnCases, //For instructions that can have a set of cases applied to their return value
//Must be last
NbTypes,
Invalid,
};
//==========================================================================================================
// Script Instructions
//==========================================================================================================
//! #TODO: Make this a single class, and get rid of inheritance. We could just make a wrapper around the instruction that contains both and instruction and a list of sub instruction, to remedy the problem.
/***********************************************************************************************
ScriptInstruction
Represents a single instruction from a script.
Is also used to represent meta-command which do not exist in the game's script engine,
but are neccessary here to make processing faster. Such as jump labels.
***********************************************************************************************/
struct ScriptBaseInstruction
{
typedef std::deque<uint16_t> paramcnt_t;
uint16_t value; //Depends on the value of "type". Can be opcode, data, or ID.
paramcnt_t parameters; //The parameters for the instruction
eInstructionType type; //How to handle the instruction
size_t dbg_origoffset; //Original offset of the instruction in the source file if applicable. For debugging purpose
};
struct ScriptInstruction : public ScriptBaseInstruction
{
//typedef std::deque<uint16_t> paramcnt_t;
typedef std::deque<ScriptBaseInstruction> subinst_t;
//uint16_t value; //Depends on the value of "type". Can be opcode, data, or ID.
//paramcnt_t parameters; //The parameters for the instruction
//eInstructionType type; //How to handle the instruction
subinst_t subinst; //Contains sub-instructions linked to this instruction
ScriptInstruction():ScriptBaseInstruction::ScriptBaseInstruction()
{}
ScriptInstruction(ScriptInstruction && mv)
{this->operator=(std::forward<ScriptInstruction>(mv));}
ScriptInstruction(const ScriptInstruction & cp)
{this->operator=(cp);}
ScriptInstruction(ScriptBaseInstruction && mv)
{this->operator=(std::forward<ScriptBaseInstruction>(mv));}
ScriptInstruction(const ScriptBaseInstruction & cp)
{this->operator=(cp);}
ScriptInstruction & operator=(const ScriptInstruction & cp)
{
subinst = cp.subinst;
type = cp.type;
parameters = cp.parameters;
value = cp.value;
dbg_origoffset = cp.dbg_origoffset;
return *this;
}
ScriptInstruction & operator=(ScriptInstruction && mv)
{
subinst = std::move(mv.subinst);
type = mv.type;
parameters = std::move(mv.parameters);
value = mv.value;
dbg_origoffset = mv.dbg_origoffset;
return *this;
}
ScriptInstruction & operator=(const ScriptBaseInstruction & cp)
{
type = cp.type;
parameters = cp.parameters;
value = cp.value;
dbg_origoffset = cp.dbg_origoffset;
return *this;
}
ScriptInstruction & operator=(ScriptBaseInstruction && mv)
{
type = mv.type;
parameters = std::move(mv.parameters);
value = mv.value;
dbg_origoffset = mv.dbg_origoffset;
return *this;
}
operator ScriptBaseInstruction()const
{
ScriptBaseInstruction out;
out.parameters = parameters;
out.value = value;
out.type = type;
out.dbg_origoffset = dbg_origoffset;
return std::move(out);
}
};
//==========================================================================================================
// Script Routine
//==========================================================================================================
/***********************************************************************************************
ScriptRoutine
Contains a groupd of instructions.
***********************************************************************************************/
struct ScriptRoutine
{
typedef std::deque<ScriptInstruction> cnt_t;
typedef cnt_t::iterator iterator;
typedef cnt_t::const_iterator const_iterator;
inline iterator begin() { return instructions.begin(); }
inline const_iterator begin()const { return instructions.begin(); }
inline iterator end() { return instructions.end(); }
inline const_iterator end()const { return instructions.end(); }
inline size_t size()const { return instructions.size(); }
inline bool empty()const { return instructions.empty(); }
inline bool IsAliasOfPrevGroup()const { return isalias; }
bool isalias; //Whether this group refers to the same instructions as the previous one.
cnt_t instructions;
uint16_t type;
uint16_t parameter;
};
//==========================================================================================================
// Script Container
//==========================================================================================================
/***********************************************************************************************
Script
The content of a SSB file. Script instructions, strings, and constant names.
***********************************************************************************************/
class Script
{
public:
static const size_t NbLang = static_cast<size_t>(eGameLanguages::NbLang);
typedef std::vector<std::string> strtbl_t;
typedef std::vector<std::string> consttbl_t;
typedef std::deque<ScriptRoutine> grptbl_t;
typedef std::unordered_map<eGameLanguages, strtbl_t> strtblset_t;
Script(){}
Script(const std::string & name)
:m_name(name)
{}
Script(const Script & tocopy);
Script(Script && tomove);
Script & operator=( const Script & tocopy );
Script & operator=( Script && tomove );
inline const std::string & Name ()const { return m_name; }
inline void SetName ( const std::string & name ) { m_name = name; }
//!#TODO: Encapsulate those later.
inline grptbl_t & Routines() { return m_groups; }
inline const grptbl_t & Routines() const { return m_groups; }
//Returns the set of all strings for all languages
inline strtblset_t & StrTblSet() { return m_strtable; }
inline const strtblset_t & StrTblSet() const { return m_strtable; }
void InsertStrLanguage( eGameLanguages lang, strtbl_t && strings );
//Returns all strings for a specific language
inline strtbl_t * StrTbl( eGameLanguages lang );
inline const strtbl_t * StrTbl( eGameLanguages lang )const;
inline consttbl_t & ConstTbl() { return m_contants; }
inline const consttbl_t & ConstTbl() const { return m_contants; }
private:
std::string m_name;
grptbl_t m_groups;
strtblset_t m_strtable; //Multiple deques for all languages
consttbl_t m_contants;
};
//==========================================================================================================
// Script Data Structures
//==========================================================================================================
struct LivesDataEntry
{
int16_t livesid = 0;
int16_t facing = 0;
int16_t xoff = 0;
int16_t yoff = 0;
int16_t unk4 = 0;
int16_t unk5 = 0;
int16_t scrid = 0;
};
struct ObjectDataEntry
{
int16_t objid = 0;
int16_t facing = 0;
int16_t width = 0;
int16_t height = 0;
int16_t xoff = 0;
int16_t yoff = 0;
int16_t unk6 = 0;
int16_t unk7 = 0;
int16_t scrid = 0;
};
struct PerformerDataEntry
{
int16_t type = 0; //Value from 0 to 5
int16_t facing = 0;
int16_t unk2 = 0;
int16_t unk3 = 0;
int16_t xoff = 0;
int16_t yoff = 0;
int16_t unk6 = 0;
int16_t unk7 = 0;
//int16_t unk8 = 0;
//int16_t unk9 = 0;
};
struct EventDataEntry
{
int16_t width = 0;
int16_t height = 0;
int16_t xoff = 0;
int16_t yoff = 0;
int16_t unk4 = 0;
int16_t unk5 = 0;
int16_t actionidx = 0;
};
struct PosMarkDataEntry
{
int16_t xoff = 0;
int16_t yoff = 0;
int16_t unk2 = 0;
int16_t unk3 = 0;
int16_t unk4 = 0;
int16_t unk5 = 0;
int16_t unk6 = 0;
int16_t unk7 = 0;
};
struct ActionDataEntry
{
int16_t croutineid = 0;
int16_t unk1 = 0;
int16_t unk2 = 0;
int16_t scrid = 0;
};
struct ScriptLayer
{
std::vector<LivesDataEntry> lives;
std::vector<ObjectDataEntry> objects;
std::vector<PerformerDataEntry> performers;
std::vector<EventDataEntry> events;
};
/***********************************************************************************************
ScriptData
Data contained in SSA, SSS, and SSE files.
***********************************************************************************************/
class ScriptData
{
public:
typedef std::vector<ScriptLayer> layers_t;
typedef std::vector<PosMarkDataEntry> posmarks_t;
typedef std::vector<ActionDataEntry> trgentry_t;
ScriptData()
:m_datatype(eScrDataTy::Invalid)
{}
ScriptData(const std::string & name, eScrDataTy datatype )
:m_name(name), m_datatype(datatype)
{}
inline const std::string & Name()const {return m_name;}
inline void Name(const std::string & name) {m_name = name;}
inline eScrDataTy Type()const {return m_datatype;}
inline void Type(eScrDataTy ty) {m_datatype = ty;}
inline layers_t & Layers() {return m_layers;}
inline const layers_t & Layers()const {return m_layers;}
inline posmarks_t & PosMarkers() {return m_posmarkers;}
inline const posmarks_t & PosMarkers()const {return m_posmarkers;}
inline trgentry_t & ActionTable() {return m_trgtbl;}
inline const trgentry_t & ActionTable()const {return m_trgtbl;}
private:
std::string m_name;
layers_t m_layers;
posmarks_t m_posmarkers;
trgentry_t m_trgtbl;
eScrDataTy m_datatype;
};
//==========================================================================================================
// Script Set
//==========================================================================================================
/***********************************************************************************************
ScriptSet
A script set is an ensemble of one or no ScriptData, and none or more
Scripts, that share a common identifier.
***********************************************************************************************/
class ScriptSet
{
public:
typedef std::unique_ptr<ScriptData> dataptr_t;
typedef std::map<std::string,Script> seqtbl_t;
typedef seqtbl_t::const_iterator const_seqtbl_iter_t;
typedef seqtbl_t::iterator seqtbl_iter_t;
ScriptSet(const std::string & id, eScriptSetType ty = eScriptSetType::INVALID)
:m_indentifier(id), m_type(ty)
{}
ScriptSet( const ScriptSet & other)
:m_indentifier(other.m_indentifier), m_type(other.m_type), m_sequences(other.m_sequences)
{
if( other.m_data != nullptr )
m_data.reset( new ScriptData(*(other.m_data.get())) );
}
ScriptSet( ScriptSet && other)
:m_indentifier(std::move(other.m_indentifier)), m_type(other.m_type)
{
m_sequences = std::move( other.m_sequences );
m_data.reset(other.m_data.release());
other.m_data = nullptr;
}
ScriptSet& operator=( const ScriptSet & other)
{
m_indentifier = other.m_indentifier;
m_type = other.m_type;
m_sequences = other.m_sequences;
if( other.m_data != nullptr )
m_data.reset( new ScriptData(*(other.m_data.get())) );
return *this;
}
ScriptSet& operator=( ScriptSet && other)
{
m_indentifier = std::move(other.m_indentifier);
m_type = other.m_type;
m_sequences = std::move( other.m_sequences );
m_data.reset(other.m_data.release());
other.m_data = nullptr;
return *this;
}
inline std::string & Identifier() { return m_indentifier; }
inline const std::string & Identifier()const { return m_indentifier; }
//TEMP
inline ScriptData * Data() {return m_data.get();}
inline const ScriptData * Data()const {return m_data.get();}
inline void SetData( ScriptData && data )
{
m_data.reset( new ScriptData(std::forward<ScriptData>(data)));
}
inline seqtbl_t & Sequences() {return m_sequences;}
inline const seqtbl_t & Sequences()const {return m_sequences;}
const_seqtbl_iter_t SequencesTblBeg()const { return m_sequences.begin(); }
seqtbl_iter_t SequencesTblBeg() { return m_sequences.begin(); }
const_seqtbl_iter_t SequencesTblEnd()const { return m_sequences.end(); }
seqtbl_iter_t SequencesTblEnd() { return m_sequences.end(); }
eScriptSetType Type()const { return m_type; }
void Type(eScriptSetType newty) { m_type = newty; }
//Returns the file extension of the data file for this group if applicable.
// otherwise, returns an empty string.
const std::string & GetDataFext()const;
private:
std::string m_indentifier; //The identifier for this set. AKA the filename prefix all its components share.
dataptr_t m_data; //Contains SSA, SSS, or SSE files. There can be 0 or more.
seqtbl_t m_sequences; //Contains SSB data.
eScriptSetType m_type; //Type of the set. Is usually tied to the data file, but not always.
};
//==========================================================================================================
// Script Level Data
//==========================================================================================================
/***********************************************************************************************
LevelScript
A group of scripts contained within a single named script folder in the
PMD2 games. Each of those are tied to a specific level.
***********************************************************************************************/
class LevelScript
{
public:
typedef std::array<char,8> lsdtblentry_t;
typedef std::deque<lsdtblentry_t> lsdtbl_t;
typedef std::deque<ScriptSet> scriptsets_t;
typedef scriptsets_t::iterator iterator;
typedef scriptsets_t::const_iterator const_iterator;
//Constructors
LevelScript (const std::string & name);
LevelScript (const std::string & name, scriptsets_t && sets, lsdtbl_t && lsdtbl );
LevelScript (const LevelScript & other);
LevelScript & operator=(const LevelScript & other);
LevelScript (LevelScript && other);
LevelScript & operator=(LevelScript && other);
//Accessors
inline void Name(const std::string & name) { m_name = name; }
inline const std::string & Name()const { return m_name; }
inline void Components(scriptsets_t && comp) { m_components = std::move(comp); }
inline const scriptsets_t & Components()const { return m_components; }
inline scriptsets_t & Components() { return m_components; }
inline lsdtbl_t & LSDTable() {return m_lsdentries;}
inline const lsdtbl_t & LSDTable()const {return m_lsdentries;}
//Std methods
inline iterator begin() { return m_components.begin(); }
inline const_iterator begin()const { return m_components.begin(); }
inline iterator end () { return m_components.end(); }
inline const_iterator end ()const { return m_components.end(); }
private:
std::string m_name; //The name of the set. Ex: "D01P11A" or "COMMON"
scriptsets_t m_components; //All scripts + data groups (SSA,SSS,SSE + SSB)
lsdtbl_t m_lsdentries; //Entries in the LSD table Stored here for now
bool m_bmodified; //Whether the content of this set was modified
};
};
#endif
|
An online multimedia experience explaining what NATO is, what it does, how it works, what is on its agenda and key links for more information.
Explore this interactive map, illustrating the who, what, where and when of NATO.
An overview of NATO member countries with links to the national information servers of each country and to the website of its national delegation to NATO.
NATO cooperates with a range of countries and international organisations in different structures (Euro-Atlantic Partnership Council, Mediterranean Dialogue, Istanbul Cooperation Initiative,...). A list of these international organisations and countries with links to their information servers.
Biographies and pictures of the current senior government officials, Permanent Representatives to the North Atlantic Council, all NATO Secretaries General, principal officials to the International Staff, members of the Military Committee, major NATO Commanders and the principal officials of the NATO International Military.
An up to date overview of NATO's organisations and structure with links to more specific websites if available.
Step through NATO’s history with videos, posters, declassified documents and other materials that shed light on the Alliance and its evolution since 1949.
This online service gives you access to NATO Headquarters procurement notices. This service is free and may provide you with some business opportunities.
A list of current international civilian staff vacancies at NATO headquarters and in the various NATO locations world-wide.
NATO Headquarters (HQ) introduced its Internship Programme in 2004. The aim of the programme is to provide a small number of current or recent students with the opportunity to intern with the International Staff at NATO Headquarters in Brussels.
The NATO Public Diplomacy Division runs an extensive public information programme towards the Alliance's Partner Countries. This programme includes publications in a number of Partner Countries' languages, sponsored visits by delegations of opinion-formers to NATO Headquarters in Brussels as well as co-sponsored conferences, seminars and workshops taking place in the Partner Countries themselves.
What is NATO? 18 Feb. 2019 A quick glance at NATO, its role, activities and members. |
<reponame>artisjaap/polyglot<filename>artisjaap-crossword/src/main/java/be/artisjaap/crossword/action/to/CrosswordWordsTO.java<gh_stars>0
package be.artisjaap.crossword.action.to;
import lombok.Builder;
import lombok.Value;
import java.util.ArrayList;
import java.util.List;
@Value
@Builder
public class CrosswordWordsTO {
@Builder.Default
private List<CrosswordWordTO> words = new ArrayList<>();
}
|
Barndale House School at the Teddy Bears' Picnic at The Alnwick Garden.
As 2017 draws to a close and Northumberland Freemasons mark the end of the 300th anniversary of the United Grand Lodge of England, the organisation has given away £600,000 to local and national charities.
Tritlington farmer, Ian Craigs, the Provincial Grand Master of Northumberland, explained that this year, the Provincial Grand Lodge of Northumberland has given away £300,000 to local charities to boost worthwhile and deserving projects throughout the region.
“In this special year, we’ve donated money to charities close to each of our 27 lodge buildings across North Tyneside, Newcastle and Northumberland so that we can really make a positive impact on local projects and causes near to where Masonry takes place,” he said.
In total, 78 charities received cheques for the good work they do to help the people of Newcastle, North Tyneside and Northumberland.
In addition, at the Province’s annual general meeting in November, a cheque for a further £300,000 was given to the Masonic Charitable Foundation.
A range of diverse activities have taken place to mark the 300th anniversary, but one of Mr Craig’s favourite events was a Teddy Bears’ Picnic held at The Alnwick Garden in the summer.
The initiative, hosted by Masonic charity, Teddies for Loving Care (TLC), saw more than 200 children from schools in from across Northumberland and Newcastle, enjoy a teddy-bear trail, a picnic box, entertainment by colourful fairy-tale characters and, of course, the obligatory teddy bear which was given to every child. |
import { Renderable, Collider, Sprite, Velocity, Position, Player, ShootsBullets } from '../components'
export function player () {
return [
new Position(0, 0),
new Velocity(0, 0),
new Sprite('player', -137 / 2, -121),
new Collider(-35, -121, 35, 0),
new Renderable(10),
new Player(500),
]
}
|
import crypto from 'crypto';
import util from 'util';
import { UserEntity } from '../domains/users.entity';
import { BadRequestException, ForbiddenException } from '../lib/exceptions';
import { AuthRepository } from '../repositories/auth.repo';
import { JoinDto, SnsJoinDto } from '../dtos/users.dto';
class AuthService {
private readonly authRepository: AuthRepository;
private EXEC_NUM = parseInt(process.env.EXEC_NUM, 10);
private RESULT_LENGTH = parseInt(process.env.RESULT_LENGTH, 10);
private randomBytes = util.promisify(crypto.randomBytes);
constructor(authRepository: AuthRepository) {
this.authRepository = authRepository;
}
public async findAll(): Promise<UserEntity[]> {
const users: UserEntity[] = await this.authRepository.findAll();
return users;
}
public async findById(userId: string): Promise<UserEntity> {
const findUser: UserEntity = await this.authRepository.findById(userId);
return findUser;
}
public async findByEmail(userEmail: string): Promise<UserEntity> {
const findUser: UserEntity = await this.authRepository.findByEmail(userEmail);
return findUser;
}
public async login(email: string, password: string): Promise<UserEntity> {
const findUser: UserEntity = await this.authRepository.findByEmail(email);
if (!findUser) {
throw new BadRequestException('Cannot find user');
}
const cryptedPassword: Buffer = crypto.pbkdf2Sync(
password,
findUser.salt,
this.EXEC_NUM,
this.RESULT_LENGTH,
'sha512'
);
if (findUser.password === cryptedPassword.toString('base64')) {
if (findUser.deactivatedAt !== null) {
throw new ForbiddenException('Deactivated account');
}
return findUser;
} else {
throw new BadRequestException('Check your id and password');
}
}
public async createUser(userData: JoinDto): Promise<UserEntity> {
const findUser: UserEntity = await this.authRepository.findByEmail(userData.email);
if (findUser) {
throw new BadRequestException(`Duplicated email ${userData.email}`);
}
if (userData.userPw !== userData.userPwRe) {
throw new BadRequestException('Passwords are not matched');
}
const generatedId: string = crypto
.createHash('sha256')
.update(userData.email)
.digest('hex')
.slice(0, 14);
const salt: string = await (await this.randomBytes(64)).toString('base64');
const cryptedPassword: Buffer = crypto.pbkdf2Sync(
userData.userPw,
salt,
this.EXEC_NUM,
this.RESULT_LENGTH,
'sha512'
);
const authToken = cryptedPassword.toString('hex').slice(0, 24);
const createUser: UserEntity = await this.authRepository.createUser({
email: userData.email,
nickname: userData.userNick,
screenId: generatedId,
password: <PASSWORD>.toString('base64'),
salt,
token: authToken,
displayLanguage: userData.userLang,
});
return createUser;
}
public async createSnsUser(snsJoinData: SnsJoinDto): Promise<UserEntity> {
const generateId: string = crypto
.createHash('sha256')
.update(snsJoinData.email)
.digest('hex')
.slice(0, 14);
const salt: string = await (await this.randomBytes(64)).toString('base64');
const cryptedPassword: string = await crypto
.pbkdf2Sync(snsJoinData.email, salt, this.EXEC_NUM, this.RESULT_LENGTH, 'sha512')
.toString('base64');
const createUser = await this.authRepository.createUser({
email: snsJoinData.email,
password: <PASSWORD>,
salt,
nickname: snsJoinData.name,
screenId: generateId,
displayLanguage: snsJoinData.displayLanguage,
profile: snsJoinData.profile,
snsId: snsJoinData.uid,
snsType: snsJoinData.snsType,
isConfirmed: true,
});
return createUser;
}
public async updateUser(userId: string, userData: Partial<UserEntity>): Promise<UserEntity> {
if (!userId || !userData) {
throw new BadRequestException('User id and data required');
}
const updateUserData = await this.authRepository.updateUser(userId, userData);
if (!updateUserData) {
throw new BadRequestException('Not a user');
}
return updateUserData;
}
public async deleteUser(userId: string): Promise<UserEntity> {
const deleteUser: UserEntity = await this.authRepository.deleteUser(userId);
if (!deleteUser) {
throw new BadRequestException('Cannot find user');
}
return deleteUser;
}
public async changePassword(email: string, userPwNew: string) {
const findUser: UserEntity = await this.findByEmail(email);
if (!findUser) {
throw new BadRequestException('Cannot find user');
}
const newSalt: string = await (await this.randomBytes(64)).toString('base64');
const newPassword: string = await (
await crypto.pbkdf2Sync(userPwNew, newSalt, this.EXEC_NUM, this.RESULT_LENGTH, 'sha512')
).toString('base64');
await this.authRepository.updateUser(findUser._id, {
salt: newSalt,
password: <PASSWORD>,
});
}
public async confirmUser(email: string, token: string): Promise<boolean> {
const findUser: UserEntity = await this.authRepository.findByEmail(email);
if (!findUser) {
throw new BadRequestException('Cannot find user');
}
if (findUser.token === null && findUser.isConfirmed === true) {
throw new ForbiddenException('Already confirmed');
}
if (findUser.token !== token) {
throw new BadRequestException('Data not matched');
}
const confirmation = await this.updateUser(findUser._id, {
token: null,
isConfirmed: true,
});
if (!confirmation) {
return false;
}
return true;
}
public async findBySnsId(snsId: string, snsType: string): Promise<UserEntity> {
const findUser: UserEntity = await this.authRepository.findBySnsId(snsId, snsType);
return findUser;
}
public async deactivateUser(userId: string): Promise<boolean> {
const findUser = await this.authRepository.findById(userId);
if (!findUser) {
throw new BadRequestException('Cannot find user. Check id.');
}
await this.authRepository.updateUser(userId, { deactivatedAt: new Date() });
return true;
}
}
export default AuthService;
|
<reponame>influx6/rpkit
package sereal
import "reflect"
type tagsCache struct {
cmap map[reflect.Type]map[string]int
}
func (tc *tagsCache) Get(ptr reflect.Value) map[string]int {
if ptr.Kind() != reflect.Struct {
return nil
}
if tc.cmap == nil {
tc.cmap = make(map[reflect.Type]map[string]int)
}
ptrType := ptr.Type()
if m, ok := tc.cmap[ptrType]; ok {
return m
}
m := make(map[string]int)
l := ptrType.NumField()
for i := 0; i < l; i++ {
field := ptrType.Field(i).Tag.Get("sereal")
if field == "-" {
// sereal tag is "-" -- skip
continue
}
if field == "" {
// no tag? make one from the field name
if pkgpath := ptrType.Field(i).PkgPath; pkgpath != "" {
// field not exported -- skip
continue
}
field = ptrType.Field(i).Name
}
m[field] = i
}
// empty map -- may as well store a nil
if len(m) == 0 {
m = nil
}
tc.cmap[ptrType] = m
return m
}
|
IN PURSUIT OF CAREERS IN THE PROFESSORIATE OR BEYOND THE PROFESSORIATE : WHAT MATTERS TO DOCTORAL STUDENTS WHEN MAKING A CAREER CHOICE ? Aim/Purpose This qualitative study was conducted to illuminate the under-researched aspect of doctoral students career decision-making by examining their internal cognitive processes based on the Cognitive Information Processing (CIP) theory. Specifically, this study compared doctoral students career decision-making from two career groups, those pursuing the professoriate versus those pursuing careers beyond the professoriate. Background Due to PhD workforce supply-demand imbalances in academic job markets and to a growing interest in careers outside academia around the world, an increasing number of doctoral recipients have pursued careers beyond the professoriate, which are considered non-traditional career paths in doctoral education. While a growing number of studies have investigated these changing trends, it remains limited to fully capture more introspective domains of the career choice processes. Given that the career decision-making experience is highly individualized, it is critical to explore doctorate students own narratives about career decision-making. Methodology Individual structured interviews were conducted with 30 doctoral students from a public research-oriented university in the United States. Employing Directed Content Analysis, two researchers developed the initial coding categories based on the guiding theory, CIP theory, and deductively analyzed the data to identify emerging major themes. What Matters to Doctoral Students When Making a Career Choice? 616 Contribution Findings from the study highlight the core factors that influence doctoral students career choices across fields, which allows developing centralized career resources and support systems at the institutional level. Specifically, findings pointed to different approaches for doctoral students to (re-)assess their career choice while providing implications for institutions, academic departments, and individual stakeholders such as faculty advisor and doctoral students, to develop systematic career support in this changing academic job market. Findings Data analysis uncovered three core factors impacting doctoral students career decision making, which are roles of the first-hand experience in career confirmation/shift; dissimilar career readiness status by group; and impact of personal career values. Recommendations for Practitioners Both institutions and academic departments could reassess the culture and value of career development and refine co-curricular activities to offer adequate professional development opportunities in doctoral training to develop career support systems aligned with students diversified career needs and interests. As time and first-hand experiences are identified as critical factors facilitating their career progress, doctoral students may want to proactively seek diverse opportunities to gain first-hand experience in and outside campus. Recommendations for Researchers Researchers could continue similar research in other universities and countries where similar concerns exist. These studies would help fully clarify common influential factors on career choices of doctoral students across fields. Impact on Society Considering the realities of doctoral students diversified career interests and career outcomes, institutes of higher education should make intentional efforts to broaden the definition of successful PhD career outcomes, which ultimately helps break the prevailing myth that doctoral students or recipients who pursue careers beyond the professoriate, called nontraditional or alternative career paths, are considered as failures or incompetent. Future Research Future research should consider examining diverse doctoral student populations such as early-stage doctoral students to discover additional factors influencing their career decision-making. The authors also recommend cross-cultural studies in other countries where similar career concerns exist, such as the U.K. and the Netherlands, to develop a more comprehensive understanding of how doctoral students career decisions are made. Background Due to PhD workforce supply-demand imbalances in academic job markets and to a growing interest in careers outside academia around the world, an increasing number of doctoral recipients have pursued careers beyond the professoriate, which are considered non-traditional career paths in doctoral education. While a growing number of studies have investigated these changing trends, it remains limited to fully capture more introspective domains of the career choice processes. Given that the career decision-making experience is highly individualized, it is critical to explore doctorate students' own narratives about career decision-making. Methodology Individual structured interviews were conducted with 30 doctoral students from a public research-oriented university in the United States. Employing Directed Content Analysis, two researchers developed the initial coding categories based on the guiding theory, CIP theory, and deductively analyzed the data to identify emerging major themes. Contribution Findings from the study highlight the core factors that influence doctoral students' career choices across fields, which allows developing centralized career resources and support systems at the institutional level. Specifically, findings pointed to different approaches for doctoral students to (re-)assess their career choice while providing implications for institutions, academic departments, and individual stakeholders such as faculty advisor and doctoral students, to develop systematic career support in this changing academic job market. Findings Data analysis uncovered three core factors impacting doctoral students' career decision making, which are roles of the first-hand experience in career confirmation/shift; dissimilar career readiness status by group; and impact of personal career values. Recommendations for Practitioners Both institutions and academic departments could reassess the culture and value of career development and refine co-curricular activities to offer adequate professional development opportunities in doctoral training to develop career support systems aligned with students' diversified career needs and interests. As time and first-hand experiences are identified as critical factors facilitating their career progress, doctoral students may want to proactively seek diverse opportunities to gain first-hand experience in and outside campus. Recommendations for Researchers Researchers could continue similar research in other universities and countries where similar concerns exist. These studies would help fully clarify common influential factors on career choices of doctoral students across fields. Impact on Society Considering the realities of doctoral students' diversified career interests and career outcomes, institutes of higher education should make intentional efforts to broaden the definition of "successful" PhD career outcomes, which ultimately helps break the prevailing myth that doctoral students or recipients who pursue careers beyond the professoriate, called nontraditional or alternative career paths, are considered as failures or incompetent. Future Research Future research should consider examining diverse doctoral student populations such as early-stage doctoral students to discover additional factors influencing their career decision-making. The authors also recommend cross-cultural studies in other countries where similar career concerns exist, such as the U.K. and the Netherlands, to develop a more comprehensive understanding of how doc- INTRODUCTION In times of globalization and uncertainty, doctoral education around the world has been considered a critical part of national competitiveness in a global knowledge-driven economy as it develops a new generation of advanced-knowledge producers and innovators who create or transfer knowledge to applicable innovations (Nerad, 2020;Rudd & Nerad, 2015). Traditionally, the structure of doctoral training in the United States was based on an apprenticeship or mentor model that guides trainees to follow in the footsteps of their mentors: tenure track faculty careers (). Over the past decade, however, the employment landscape for PhDs has shifted significantly. While the number of academic tenure-track jobs has stayed stable or decreased, there is no indication that the size of PhD workforce is decreasing (;St. ). Due to PhD workforce supply-demand imbalances in academic job markets in the U.S. and to a growing interest in careers outside academia, an increasing number of doctoral recipients have pursued careers beyond the professoriate (Sauermann & Roach, 2012;St. ). This emerging trend does not limit the U.S. context since an increasing body of doctoral recipients worldwide, such as in the U.K., Estonia, Netherlands, France, and Spain, also pursue careers beyond academic research and teaching careers (Auriol 2007;;Kindsiko & Baruch, 2019;;van der ). A growing number of scholars have investigated these trends (e.g., Agarwal & Ohyama, 2013;;;Sauermann & Roach, 2012;;St. ), providing valuable insights and generating empirical evidence to better understand PhD career pathways and career experience during or after their training. However, the majority of the research on doctoral career development has mostly utilized quantitative methods, which remains limited in its ability to fully capture more introspective domains of the career choice processes by which doctoral students gather, transform, and apply information to arrive at a career choice. Given that the career decision-making experience is highly individualized, we believe it is critical to explore how individuals experience their career choice process in their own words instead of asking doctoral students to fit their experiences into researchers' pre-determined measurements. Moreover, a large number of existing empirical studies on doctoral students' career-related experiences focus heavily on the science and engineering fields (e.g., ;;Roach & Sauermann, 2010;Sauermann & Roach, 2012;St. ). These studies illuminated unique field-specific career challenges and experiences of doctoral students or graduates, but due to limited cross-disciplinary studies, an understanding of the other side's experience simultaneously is lacking. Such a limited view of doctoral students' overall career-related experiences prevents institutional leaders and policy makers not only from recognizing the core, or common, factors that influence career choices of doctoral students across fields but also from developing centralized career resources and support systems at the institutional level (). To fill the gap, the purpose of this study was to illuminate the under-researched aspect of doctoral students' career decision-making by exploring their internal cognitive processes through in-depth individual interviews with doctoral candidates. With the Cognitive Information Processing (CIP) theory () as our theoretical foundation, we specifically focused on comparing the career decision-making process of doctoral students from two different career groups-those pursuing a career in the professoriate (CP) versus those pursuing a career beyond the professoriate (CBP)-to uncover factors that influenced their career decision making. We employed a structured qualitative approach in an effort to keep both rigor and exploratory viewpoints. DOCTORAL STUDENTS' CAREER DEVELOPMENT AND OUTCOMES For decades, the majority of new PhD recipients pursued the traditional linear academic career trajectory: moving directly into tenure-track faculty positions (;K. D. Gibbs & Griffin, 2013;Kindsiko & Baruch, 2019). In today's academic job market, however, less than 25% of PhD graduates are able to land tenure-track faculty positions, and some academic fields, such as life sciences, even require 5-6 years of additional postdoc training to attain these faculty positions in the U.S. (Stephan 2012). Similarly, a recent study conducted in Netherlands (van der ) found that only fewer than 3% of PhD recipients actually secured a tenure-track job. Despite a shortage in available tenure-track faculty positions, becoming a tenure-track professor is still considered the common, most desired career in many PhD training universities (Sauermann & Roach, 2012;St. ). Due to the noticeable PhD workforce supply-demand imbalance in both national and global academic job markets (Kindsiko & Baruch, 2019;St. ), doctoral education communities have raised a concern about whether the U.S. higher education institutions overproduce a PhD workforce (). On the other hand, a growing number of scholars claim that the widening concerns about the PhD employment crisis results mainly from a mismatch between how doctoral students are trained and career opportunities actually available to them after completing their doctoral training rather than simple supply-demand imbalances or the oversupply of PhDs (e.g., Sauermann & Roach, 2012;St. ;Stenard & Sauermann, 2016). They viewed the current doctoral workforce challenge as being more about underutilization, rather than overproduction, of PhDs. This may be largely because of the observable PhD career outcomes trend, regardless of field, that many doctoral students and graduates have pursued careers outside academia such as industry, nonprofit organizations, and government, which are traditionally considered less acceptable and desirable career pathways for PhDs (Sauermann & Roach, 2012). According to the findings of the Survey of Earned Doctorates (National Science Foundation, 2019), only 46% of all PhD graduates from U.S. universities expressed definite commitments to traditional academic employment. Sinche and colleagues also found 34% of 8,099 PhDs who graduated from U.S. universities between 2004 and 2014 were employed as faculty members in research-oriented and teaching-oriented institutions. These findings clearly demonstrate that doctoral graduates are involved in a wide range of careers beyond the professoriate, which should be no longer considered as "nontraditional" or "alternative" careers for the doctoral workforce. Further, empirical research has repeatedly found that doctoral students often change their primary career interest or preference over the course of their doctoral training. For example, based on the data collected from 469 biomedical science students across years of doctoral training, Fuhrmann and colleagues found noticeable shifts in doctoral students' career preferences during the third year of their graduate training from tenure-track faculty to careers outside academia (e.g., industry, government). Similarly, Roach and Sauermann also identified declining interests in the traditional academic career path even though the majority of the participants expressed strong interests in becoming faculty members during the early stages of their doctoral training. Such changes in career choices of doctoral students are not driven mainly by fierce competition in academic job markets, or by research interests to pursue academic research careers. Rather, the shifting career choices of doctoral students during their training is more related to personal values (e.g., balancing work and life), the research environment in academia (e.g., slow pace of academic research), and preferences for work activities (e.g., applied research to address practical problems) (;K. D. Gibbs & Griffin, 2013;Roach & Sauermann, 2017;Sauermann & Roach, 2012). Although the existing literature reveals unique patterns of career preferences and challenges of doctoral students and graduates during or after their training, knowledge about the doctoral career experience as a process is still partial. With heavy use of quantitative data, the current literature has not sufficiently explored the dynamic mechanisms through which doctoral students make their career choices and implement them, especially from their own perspective. CAREER DECISION-MAKING PROCESS GROUNDED ON THE CIP THEORY The Cognitive Information Processing (CIP) theory focuses on the role of cognitive factors in an individual's career decision-making process (;). In particular, the CIP theory illustrates an entire process of an individual's career decision making to clarify different types of mental activities regarding career choice. The process is composed of a five-phase cycle: communication, analysis, synthesis, valuing, and execution (the CASVE cycle). It describes how an individual gathers and processes information to make a career choice and develop action steps to achieve that career goal (;). During the communication phase, an individual identifies a career gap between "where they are and where they want to be" (, p. 26) as a result of self-awareness based on interac-tions with themselves (internal cues) or their environment (external cues). When their recognized career gap makes them feel more uncomfortable than the fear of change, people tend to initiate career exploratory behaviors to reduce their career gap (). In the analysis phase, individuals expand their career knowledge domains, including self-understanding (e.g., interest, values, skills, employment preferences, personal/family situations) and occupational knowledge, as result of their career exploratory behaviors. Through these exploratory activities, individuals can recognize key factors, such as a gender stereotype, that may have positively or negatively influenced their career decision-making process. In this phase, people tend to elucidate career problems that may have caused their career gap (;). Based on the self-understanding and occupational knowledge built from the analysis phase, individuals in the synthesis phase develop possible career directions that may be able to reduce or close the career gap. Once they collect all the possible options, they review them again to identify options that are congruent or incongruent with who they are. This process allows individuals to narrow down their career options and keep the list manageable (). Throughout the valuing phase, individuals prioritize the remaining career options by evaluating potential benefits and costs that would result from pursuing each option and the impacts on both themselves and their significant others or communities. This cost-benefit analysis strategy helps them determine a tentative primary career choice (). Individuals develop action plans and implement them during the execution phase to achieve their primary career choice (). Finally, they return to the communication phase to assess whether outcomes of the action items implemented during the execution phase successfully resolved the initially identified career gap. They restart the cycle again if the original career gap has not yet been solved (). The CIP theory also highlights the impact of individual affective factors on the career decision-making process. In particular, negative beliefs or thoughts during the CASVE cycle prevent people from making steady progress in their career decision making (;;;). For example, Bullock-Yowell and colleagues reported that negative career thoughts significantly reduced students' confidence in their abilities to perform career decision-making activities, which in turn prevented them from engaging in career exploratory activities. The current empirical studies demonstrate that affective factors play a critical role in an individual's career decision-making process (;). However, it is difficult to fully gain comprehensive insights of the entire career choice process because the existing studies, driven mostly by quantitative research approaches, focused on a specific phase of the CASVE decision-making process (e.g., ;). We thus employed a qualitative research method to investigate the entire career-decision making process of individual doctoral students. In particular, we used the CASVE cycle to gain a process-oriented understanding, as individuals often use the cycle as a guide for their career decision making (). The process-oriented data ultimately helped us uncover influential factors on these career decisions. METHOD PARTICIPANTS A total of 30 doctoral candidates across disciplines from a public U.S. research-oriented university were recruited. A purposive sampling strategy was used to recruit only participants who suited the purpose of the study. Specifically, we invited only doctoral candidates who had already decided their primary career choice and were about to enter or already in the job market after passing their preliminary exam, as we considered them "job seekers" who had recently experienced the entire career decision-making process. For comparison purposes, we managed the ratio of numbers of participants in each of the two career groups, the CP group and CBP group. In this study, career options in the CBP group include higher education administrators and researchers/employees in research institutions, business, or government. The majority of them were in either the 4th or 5th year of their doctoral training. Over half were under the age of 30. Most of the participants were either White or Asian. Table A1 in the Appendix provides the profiles of interview participants based on self-reported demographic characteristics. We used pseudonyms for all participants reported in this article. DATA COLLECTION A qualitative research approach allows a phenomenon to reveal itself (Kindsiko & Baruch, 2019) to highlight a more introspective aspect of the doctoral career decision-making experience. We conducted individual open-ended, structured interviews. To capture the process-oriented career decisionmaking experience of doctoral students in a theoretically valid manner, core interview questions were initially developed by the first author based on the CASVE cycle framework (). To reinforce the feasibility and quality of our interview instruments, the interview questions were reviewed by two doctoral-level experts who specialize in qualitative research and CIP theory and modified after one-on-one pilot interviews with eight doctoral candidates. Before conducting interviews, doctoral candidates across disciplines were invited to complete a brief online survey to collect their demographic information, including their primary career choice. Among the survey participants, only individuals who expressed their interest to participate in the interview and met the criteria of the study were invited to the one-on-one interviews. With permission, all interviews were audio-recorded. While asking the core interview questions, various probing questions were also used (e.g., "would you mind sharing a specific example?") to illustrate participants' responses if needed. In this way, we were able to obtain "rich data filled with words that reveal the respondents' perspectives" (Bogdan & Biklen, 2007, p. 104). The 30 in-depth qualitative interviews resulted in about 21 hours of audio data and 576 pages of transcript. The first author of the current study conducted all of the interviews and kept a reflective journal during the data collection. Reflective journaling enabled her to not only reflect on her interactions with participants but also recognize her own biases to minimize their impact during data collection and data analysis (Lincoln & Guba, 1985). DATA ANALYSIS Directed Content Analysis (DCA; Hsieh & Shannon, 2005) uses a guiding theory to develop initial coding categories prior to data analysis. Based on the pre-determined coding categories, researchers deductively analyze the data. DCA is a more structured approach than a traditional content analysis that employs an open-coding strategy (Hsieh & Shannon, 2005). Employing DCA, we used the CIP's CASVE cycle as our guiding theory and developed the initial coding categories, which enabled us to identify emerging major themes as they evolved throughout the interview data. All of the interviews were transcribed and uploaded into the NVIVO 11.0 software package for data analysis. Prior to data analysis, participants were invited to review their own transcript to assess whether the transcript accurately described what they said during the interview (Forbat & Henderson, 2005; R. G. Gibbs, 2007). This member-checking activity (Lincoln & Guba, 1985;Savin-Baden & Major, 2013) increased the credibility of the data, as confirmation from the participants ensured "adequate representations of their own (and multiple) realities" (Lincoln & Guba, 1985, p. 314). We analyzed the interview data with four coding rounds. In the first round of coding, the first and second authors coded two randomly selected transcripts to decide how to segment texts to capture meaningful units of analysis in a consistent manner (). After independent coding and discussion, the two researchers reached consensus about a "meaning unit" in this study, which ranged from one sentence to a paragraph. In the second round, both researchers independently coded two randomly selected transcripts by using the pre-established coding guide to reduce variation in understanding of code sub-categories and code definitions between the researchers, which may affect the inter-rater reliability (). Based on repeated comparison and discussion, modifications were made to the coding guide. In the third round of the coding, the two raters independently coded a randomly selected transcript based on the updated coding guide to examine the inter-reliability between the two coders. By using a Kappa statistic, the inter-rater reliability between the two researchers was examined. If the inter-rater reliability of any of the CASVE cycle phases was lower than.80, the two researchers discussed to identify and make sense of the discrepancy between them. If necessary, the coding guide was modified. Then the researchers reanalyzed the data. This coding process was repeated until the inter-rater reliability reached.81 or higher, as the Kappa of.81 is considered the "almost perfect" agreement range between two raters (Viera & Garrett, 2005). Finally, the raters independently coded the remaining data and reviewed all the coded data prior to merging them. After the four coding rounds were complete, emerging themes were initially identified. After careful examination, the initial themes were merged and/or reduced into three primary themes indicating influencing factors on doctoral students' career decision making. The proposed primary themes were reviewed by two external experts. A doctoral-level expert on career development and the CIP theory conducted a peer review session. With the other external doctoral-level expert on qualitative research, the external audit was conducted to evaluate the trustworthiness of the study (Savin-Baden & Major, 2013). RESULTS The following three separate primary themes emerged from the results of the data analysis, peer review, and external audit. Each had a number of subthemes discussed by participants from the CP and CBP career groups. WHAT'S FOR ME AND WHAT'S NOT: FIRST-HAND EXPERIENCES AS CAREER DECISION-MAKING CUES Participants in both career groups indicated the first-hand experience that doctoral students gained during their training helped them not only to gain more self-knowledge and/or career knowledge but also to reassess whether their original career choice was well aligned with who they are (e.g., interests, skills, values). However, the major activities reported from each group were quite different. In the CP group, most participants reported that the experiential learning they gained within their departments, such as research or teaching experiences, helped them recognize that a faculty career was what they wanted to pursue after graduation. For instance, Soojung started to consider a faculty job at a teaching-oriented institution after a positive first solo-teaching experience: I taught classes last semester. I really enjoyed it and I got quite a good student evaluation, so I think that helped me to consider a teaching institution and not just researcher in a research institution. Similarly, Cristina's teaching assistant experience helped her realize that being a professor at a teaching-oriented institution would be a great fit for her. She was convinced that being a faculty member would allow her to focus more on mentoring and teaching while continuing to "keep your research going." On the other hand, Nill described how his research experience during his doctoral program reshaped his interest in an academic research career, which was not part of his original career plan: I was strongly considering being a private practice therapist. When I first got here, I didn't really understand research. I didn't understand what it was. But as I started working through the program, I started understanding what you can do with research and I became obsessed with it. I wanted to be a researcher pretty early on here and it's really sort of been confirmed over the last three or four years that research is more of my strength. He especially emphasized that "publishing my first paper fueled research interest," resulting in shifting his career pursuit from a private psychologist to a faculty member in a research-oriented institution. Unlike the CP group participants, who reported their research or teaching experience as major career cues, most of the CBP group indicated employment experience outside the department and vicarious learning as their major career decision-making cues. For example, Jackson shared that working as a higher education administrator during his gap year "changed my mind because I realized I didn't just have to be a faculty member." Emma also indicated that her first campus job helped her see that working as a professor was not her only career option: I've worked as a graduate mentor, which is kind of an academic advisor. It started to make me think about other ways that you can work with students. The experiences that I've had made me think that that could be a good career path for me. Still in academia, but not necessarily teaching . In addition to these on-or off-campus employment experiences, participants in the CBP group also reported that knowledge learned vicariously from faculty members or colleagues who became faculty members served as career cues to reconsider their career choice. For example, Bella recalled the time when she learned about faculty's daily responsibilities from junior faculty members: Their stories were so discouragingfaculty members were like "I teach four classes each semester. I had put in a lot of service as an assistant professor and I have to write grant proposals." I don't want to work 100 hours a week. Further, the majority of the participants from both groups emphasized that making their current career choice was a gradual process through their experiential learning over the years rather than a specific turning point. Participants described their decision as "a progression over time" or "gradual release of responsibilities" as a result of understanding "what's for me and what's not." For example, as a result of multiple teaching experiences (e.g., teaching assistant, instructor) during her program, Cristina was "slowly introduced to teaching," which helped her realize that teaching "is doable. I could enjoy doing this." She described her decision resulted from a "gradual takeover of responsibilities." CLEARLY AWARE OF OR CONFUSING: DISSIMILAR LEVELS OF PERCEIVED CAREER READINESS While all the participants in the current study were at the final stage of their doctoral program, they perceived their current career statuses differently. Most participants in the CP group considered that they were either about to enter or already on the job markets, whereas the majority of CBP group participants reported that they were still in an exploration stage, collecting occupational information relevant to their sought career path. The majority of participants in the CP group actively participated in action-oriented job search activities to achieve their chosen career goals. Over half reported that they had already applied for faculty jobs. For example, Soojung was "starting to send my applications to schools. This weekend, I have one Skype interview." Similarly, Pei said that "I'm applying to places like the East Coast and wherever there's opening." Most of the doctoral candidates in this group were clearly aware of what types of competencies and experiences were critical for their chosen career, which enabled them to strategically seek relevant opportunities to make their applications more competitive. The most frequently reported experiences were publishing academic articles, writing grant proposals, teaching classes, and presenting papers at academic conferences. For example, prior to entering the academic job market, Pei intentionally sought teaching opportunities even beyond her department as she noticed that her teaching experience might be not enough to become a competitive faculty candidate: I try to teach as much as I canwhenever I can, so I have taught a lot in the community at libraries and stuffto boost up my teaching experience on my CV, so that is one of the things I did. Nill shared how he strategically built his CV by not only "set a goal for how many papers I wanted to have published by the time I leave," but also seeking collaborative research projects "to show interdisciplinary work." In contrast, such action-oriented activities were rarely discussed during the interviews with participants in the CBP group. In fact, over half the group participants (n=9) focused on career exploration activities to better understand the fields or occupations they chose to pursue. For example, Miyoung gained knowledge on possible career options in her field by talking "to friends, friends of friendslike quite a bit of people." In addition to direct interaction with professionals working in their fields of interest, several students, such as Jackson and Cooper, regularly checked online job descriptions for open positions to identify available career options and examine whether there was a possible career fit, as Jackson described: Every week or every other week, I am on the university job board if I see a job that I might want to apply for. I look at the qualifications. I think about how I can talk about my own experiences as a way to fit into these qualifications. Unlike participants pursuing paths to the professoriate, participants in the CBP group were largely unaware of how their doctoral education prepared them for their sought career path or how to translate their doctoral training experience to their potential employment beyond academia. Simply put, it was "a very confusing process" to them. Several students (e.g., Bella, Cooper, Diana, Emma, Joseph) expressed their anxiety or negative thoughts, such as "I have no idea what I am going to get"; "I do not know how to articulate my skills about how to be useful"; and "I am prepared to be a very good research scientist I'm unprepared for a job in the industry." For example, in the absence of sufficient career guidance and education from her department, Emma felt that she was not sure how her skills and knowledge could be transferred to various career options beyond the professoriate: Similarly, Bella, who planned to pursue a government career, was not sure if she could secure a job due to her limited networks and knowledge in the government sector: I just do not know many people in government, so this is very difficult to obtain this kind of insider information. That is so unknown. I would like to have more direction. HAVING FREEDOM OR SECURING BOUNDARIES: PERSONAL CAREER VALUES AS A MAJOR DRIVING FORCE Throughout the career decision-making process, emerging themes showed a clear difference between groups in terms of career values, which impacted their perceived congruence with their chosen career pursuit. The CP group participants indicated freedom as the most crucial element that they desire in their careers, whereas doctoral candidates pursuing CBP reported having a more balanced life with a fixed working schedule was their most critical career value. A majority of doctoral candidates in the CP group (n=11) desired freedom with their time and work, which is a primary benefit associated with faculty careers. The participants used various terms when talking about this freedom, such as "intellectual autonomy," "flexibility," "independence," "intellectual freedom," and "liberty," since they focused slightly different aspects. Several participants (e.g., Abigail, Felicia, Keri, Pei) defined the freedom as "the ability to structure my own time" rather than "being confined to certain hours of the day." For example, Felicia emphasized her desire to develop her own working schedule based on her productivity: I tend to do a lot of my creative work in the mornings and like to meet with people in the afternoon or evening. In my future job, I would want to establish right away. I will be here in the morning working on something. Then if you want to meet with me, set up a time to meet in the afternoon because I don't like to be bothered in the morning that's when I like to be in the lab or writing. In addition to freedom in working schedule, others described freedom as a capacity to decide independently what and how to do their work, as they would "be the boss" of themselves. For example, Abigail believed that being a professor would give her "the ability to have more independence and freedom in my research and in my teaching." Similarly, Brian wanted "to do the science" and being a faculty member at a research institution would give him "intellectual freedom" or "intellectual autonomy" to "approach a problem to solve it" rather than "being a button pusher": I don't want to do something where someone is telling me, "Push this button," and I push a button for however many hours I need to be there. I mean, it is obviously not one button... That is not a job that I would want. The primary idea of freedom described by the CP group was the ability to decide what to pursue and how to accomplish those pursuits based on their own schedule. Over half of the participants in the CBP group (n=9) indicated that a work-life balance was the most important element in their career. Participants in this group shared their desire to have a more structured work schedule so that they could set a clear work-life boundary. For example, Miyoung was not concerned about working long hours as it would be similar to "the graduate school lifestyle." However, what she sought in her next career was to separate her working time from personal time: One thing I do seek once I do graduate is when I finish work and come home, I feel like I am actually done with work and I can do home like this is my hour and I can do what I want. In particular, participants who recently got married or have children in the CBP group (e.g., Emma, Jackson, Helen, Vivienne) emphasized the importance of a standardized working schedule so that their family members can "develop expectations around that structure." After her marriage, Vivienne changed her career pursuit to one that would let her to "work within a defined period of time," as she now valued more on being with a family: Interestingly, each group's career value was reported as one group's benefit, but another group's cost. Approximately half the participants in the CBP group indicated losing the freedom to "research necessarily what you want at all times" as a major cost. For example, although Edwin was excited to return to industry after graduation, he mentioned what he might need to give up: Maybe it is research that I enjoy. Sometimes, when you work in industry you have no time to write papers because it is a business. Maybe that is one of the things I will miss. I will stop doing it, but I still would love to do that. One thing I might give up is that research. Similarly, Bella expressed her concerns about losing her freedom to "research my most passionate research topics" if she left academia even though she was actively looking for a researcher position in the government sector. On the other hand, a work-life balance was the most frequently indicated cost by the participants pursuing the CP. Some participants shared the major cost that they might pay to pursue a faculty career was an unstable life, which may "hurt the family life." For example, Cristina was concerned about how her career choice might affect her work-life balance, especially when she has her own family: If you have a family, make those decisions together. I ultimately want to have this type of career, but I don't know when that will happen. I think that can be just very stressful and sort of be a negative impact on your everyday life. Another group of CP group participants, including Abigail, Nill, Natalie, and Pei, shared that they might face tremendous pressure to get tenure even after becoming an assistant professor, which could have negative impacts on their work-life balance. For example, Abigail described this cost: The process of getting tenure time consuming and sometimes I worry that it is going to feel like being in a doctoral program all over again. There is so much pressure to meet certain benchmarks and to be able to compile a certain kind of CV. Abigail emphasized that going through the tenure process would be "like a repeat of graduate school because graduate school for me has been very, very stressful." She was concerned that such "unhealthy kind of stress" might make it hard to keep a psychologically healthy life even if she could land a faculty job. Despite such a major cost, a challenge to maintain a life-work balance, participants in this group strongly believed that becoming a faculty member outweigh that cost "because as a faculty member you have the freedom," as Nill expressed. DISCUSSION AND IMPLICATIONS FOR RESEARCH AND PRACTICE The main goal of the present study was to uncover major factors that play a major role in doctoral candidates' career decision making based on an examination of their entire career decision-making process. Three core themes affecting their career decision making are roles of first-hand experience in career confirmation/shift; dissimilar career readiness status by group; and impact of personal career values. These findings offer important theoretical and practical implications/recommendations, as discussed below. RESEARCH The findings provide empirical evidence for CIP theory that career knowledge plays a critical role in the career choice process () and for the utility of CIP theory in PhD career development. In this study, a majority of the participants, regardless of their career choice, were clearly aware of their personal core values and able to articulate how their career values impacted their career choice. We also found that our participants perceived their career readiness differently even though they were all at the final stage of their PhD training, because of different levels of occupational knowledge of their chosen career path. When interview questions related to the execution phase, most of the participants in the CBP group still focused on career exploration, one of the major activities in the analysis phase, while the CP participants focused on implementing action-oriented job search activities to achieve their career goals. This finding shows that an incomplete task in a certain phase (e.g., a limited occupational knowledge during the analysis phase) can prevent individuals from successfully completing career activities in other phases, empirically supporting the theoretically proposed cyclical nature of the CASVE cycle (;). Previous research grounded in the CIP theory (e.g., ;) shows strong utilization of quantitative measures such as the Career Thoughts Inventory (), which shares a limited perspective on the entire process through which an individual determines and implements a career choice, as described by the CASVE cycle. The present study provides valuable insights to bridge this methodological gap while shedding light on the potential use of the CIP theory, especially the CASVE cycle for qualitative research. PRACTICE This study contributes to identifying different approaches for doctoral students to gain career knowledge and (re-)assess their career choice while recognizing their feelings of career readiness for a different pursuit of career. The findings are informative to reassess the culture and value of career development and refine co-curricular activities and provide adequate professional development opportunities in doctoral education. In this regard, this study has practical implications that can be established or supported at the institution/college, department, and individual levels. Institution/college level As the traditional primary objective of doctoral education is training future researchers and scholars, the perceived belief about doctoral students aspiring to faculty career is widespread (Nerad, 2004;). Due to such one-dimensional culture of career development in doctoral education around the world, growing career diversity for PhD postgraduate pursuits appears to be overlooked (;St. ). As such, many doctoral students perceive institutional indifference toward career development concerns, especially careers outside academia (K. D. Gibbs & Griffin, 2013;;;Woolston, 2017), and they do not feel properly prepared in terms of career expectations and planning (Kindsiko & Baruch, 2019;). Similarly, this study found that participants in the CBP group experienced social pressures and lower levels of support during their training to pursue careers outside academia, as they had difficulties navigating career options and resources and gaining mentors and networking outside of academia. Although a growing number of institutions have started developing and tailoring career development resources for doctoral students, many of the existing institutional career resources are often focused on tenure-track academic careers (). It is inequitable for those who pursue CBP to have fewer resources and opportunities due to the traditional perspective of career development in doctoral education. An initial action at the institutional level would be acknowledging changes in the employment landscape for PhDs, such as shifting career patterns due to the supply-demand imbalances in academic job markets around the globe (Kezar & Maxey, 2015;Kindsiko & Baruch, 2019;van der ) and growing interests in careers beyond the professoriate (Sauermann & Roach, 2012). Institutions of higher education may consider reviewing and rebuilding the career culture and values in doctoral education by evaluating career needs and preferences as well as the career development experiences of doctoral students. In addition, colleges could track and collect doctoral alumni career outcomes to inform faculty members and doctoral students of career trends and a variety of career opportunities for PhDs. They might also consider hosting annual meetings where PhD-trained professionals from a wide range of career sectors are encouraged to present their work and to network with current doctoral students. Such small but pertinent changes would allow doctoral education stakeholders and employers to become more aware that there are multiple ways that PhDs can make a difference in their discipline in addition to traditional scholarly contributions. In the 21st-century knowledge-based global economy, multiple paradigms coincide in higher education. Thus, it is time to call for each institution of higher education to attempt a cultural paradigm shift in doctoral education. Institutions might make intentional efforts to broaden the definition of "successful" PhD career outcomes, which are now mostly limited to tenure track faculty careers, as well as to reduce doctoral students' perceived anxiety toward careers beyond academia. It ultimately helps break the prevailing myth that doctoral students or recipients who pursue careers beyond the professoriate, called nontraditional or alternative career paths, are considered as failures or incompetent (Hayter & Parker, 2019;). Department level The disparity of perceived career readiness regarding different career paths in our findings suggests a sense of urgency to develop more structured career and professional development supports, in particular, tailored approaches for careers beyond academia, which is consistent with findings of the existing studies conducted inside and outside of the U.S. (Kindsiko & Baruch, 2019;;Woolston, 2017). Academic departments should consider developing consistent and long-term interventions for career advising and mentoring for doctoral students. Without it, as the current study identified, the quality of capacity building for career development is widely variable from individual to individual, which prevents some doctoral students, especially those pursuing CBP, from building knowledge and competencies necessary to pursue their career goals. Departments may want to consider assessing their doctoral curriculum to examine whether their curriculum provides proper learning opportunities that are aligned with the career development needs of doctoral students to achieve their sought career paths, both inside and outside academia. Based on the assessment, departments can identify areas of improvement and redesign their doctoral curriculum. For example, departments can integrate career and professional development components as a formal part of the curriculum, especially from the early stage of graduate training as professional competencies, which are also called generic skills in the U.K., are not typical by-products of doctoral training. Such curriculum reform should not be limited to offering general career workshops or seminars. Based on data-informed decisions, departments can build their own capacity for the integrated career and professional development curriculum and up-to-date career resources. In addition, departments can build partnerships with campus career services offices to develop an experiential learning course where doctoral students can participate in micro-internships or local and campus community projects. In this way, students can explore various career possibilities remotely or locally and examine what career path best suits who they are from a relatively well-informed position. Considering that graduate students rely primarily on resources from their own academic department and fields for their career development instead of using career services on campus (Shen & Herr, 2004), academic departments and graduate schools should take more proactive roles in making existing career resources highly visible to both faculty members and students. Further, academic departments can encourage faculty advisors to develop a structured schedule for career discussion with their doctoral students throughout training, especially from the early stage of graduate training. If necessary, departments can seek services from the campus academic advising center or career service centers to provide advising or training tailored to faculty members so that they can also receive help in supporting their doctoral students. Individual level-Doctoral students and faculty advisors Currently, conversations about careers between doctoral students and faculty advisors occur either near the end of doctoral training or not at all (;Woolston, 2017). For example, from a recent worldwide survey of over 5,700 doctoral students from diverse scientific fields, Woolston found that students barely had conversations with their advisor regarding careers beyond academia. Because of their perceived low levels of career support from their program, doctoral students, especially those pursuing careers outside academia, were less likely to seek advice from faculty members, including their own advisor (St. ). To build an environment where doctoral students can discuss their career aspirations regardless their career pursuits, both student and faculty advisors need to take ownership. Given the importance of personal career values in career decision making (;Gaule & Piacentini, 2018;K. D. Gibbs & Griffin, 2013), as the current study indicates, doctoral students are responsible for identifying what matters to them personally and professionally and continually communicating these personal career values with their faculty advisors to build mutually clear expectations. On the other hand, it is important for faculty advisors to go beyond their advisor role, which usually focuses on programmatic academic support, by considering themselves as a mentor who provides "psychosocial support, career support, and role modeling" (, p. 328). At the same time, both students and faculty advisors should recognize the shifting PhD employment landscape and acknowledge a diverse array of career options available. Such evidence-informed understanding allows doctoral students not only to avoid feeling overwhelmed or incompetent when pursuing careers beyond the professoriate (Hayter & Parker, 2019;) but also to develop a realistic training plan with their advisor for academic and career success. Along with these intentional individual efforts, ongoing and structured career conversations between students and faculty advisors throughout doctoral training would hold students more accountable for their career development while allowing students to continuously reflect on their career interest formation. From active interactions with students, faculty advisors are also able to gain a more holistic understanding of their students' career progress. Similar to previous studies (;K. D. Gibbs & Griffin, 2013;Roach & Sauermann, 2017), we found that career development is a gradual process through experiential learning over years rather than a single turning point, highlighting that time and first-hand experiences are important factors facilitating doctoral students' career progress. Doctoral students may need to proactively seek resources and opportunities to gain first-hand experience in and outside campus, rather than relying on resources from their own academic department and field (Shen & Herr, 2004). Such experientially acquired knowledge enables individuals to discover or reinforce their likes, dislikes, skills, and values (), which eventually serves as indirect career support helping them identify available career option that might be a good fit. LIMITATION AND FUTURE RESEARCH DIRECTIONS As an exploratory study to investigate doctoral students' entire cognitive career decision-making process step by step, this present study research contributes to the literature on graduate career development and theoretical understandings of career choice process. Despite its contributions, there are several limitations of the study on which further research could focus. The first limitation is the data collection source, a single public research-oriented university in the United States. Although we intentionally chose a single institution as a research site to capture the unique institutional environment and its impacts on students' career experiences in a consistent manner, caution is required when applying the findings of the study to other higher education institutions. Considering this context limitation, similar empirical research is needed at other institutions and other countries where similar career concerns exist, such as France, Netherlands, Spain, and the U.K. (;Kindsiko & Baruch, 2019;van der ). Such studies would promote cross-cultural comparisons, which provide a more comprehensive understanding of doctoral students' career decision-making experience. In addition, the current study examined only doctoral candidates who had already decided which career to pursue and the career decision-making experience of early-stage doctoral students or doctoral candidates who have yet to decide their career choice remain unknown. To fully illuminate how doctoral students' career decisions are made and changed throughout training, longitudinal research is recommended. A final limitation of this study is the use of self-developed interview questions. As discussed in the method section, these interview questions were developed based on the original theoretical framework of the CIP theory's CASVE cycle. Although the interview instruments of the study were reviewed by two doctoral-level researchers and tested through a pilot qualitative study, it is possible that the interview data collected in this study might not fully capture the career decision-making process proposed by the CIP theory. Thus, more empirical research is needed to address questions regarding the applicability of the CASVE cycle-based interview instruments of the current study to other populations. CONCLUSION This study empirically illuminates doctoral students' career decision making via participants' own narratives. It is identified as a long-term developmental process through which various first-hand experiences intertwine with a personal career value system. As personal values were identified as the primary lens of doctoral students' career decision, goals of doctoral education, which is still focused mainly on developing future generations of tenure-track faculty members, should be realigned with the realities of doctoral students' diversified career interests and outcomes as well as a growing concern that universities are overproducing PhDs. The academic community should support and celebrate a wide range of career possibilities for the doctoral workforce as it facilitates continuous knowledge creation and transfer both inside and outside academia (Nerad, 2010(Nerad,, 2020). Any students, regardless of their degree levels, should benefit from a wider range of career possibilities. |
package awsadvisor
import (
"fmt"
"time"
"bytes"
"io"
"encoding/csv"
"github.com/aws/aws-sdk-go/aws"
"github.com/aws/aws-sdk-go/service/ec2"
"github.com/aws/aws-sdk-go/service/iam"
"github.com/aws/aws-sdk-go/service/cloudtrail"
)
type AwsAuth struct {
Users *iam.ListUsersOutput
UsersDetails *iam.GetAccountAuthorizationDetailsOutput
CredReport []*UserReport
last_err error
}
type UserReport struct {
UserName string
PassLastRotation time.Time
Key1LastRotation time.Time
Key2LastRotation time.Time
}
func (c Config) Sg_is_excluded(sg_id, proto string, port int64) bool {
for _, sg := range c.Conf.Security_groups {
if( sg.Id == sg_id && sg.Proto == proto && sg.Port == port){
return true
}
}
return false
}
func (ec2 *EC2) Is_attached_to_instance(sg_id string) string {
instances := ""
for idx, _ := range ec2.Instances.Reservations {
for _, inst := range ec2.Instances.Reservations[idx].Instances {
for _, sg := range inst.SecurityGroups {
if *sg.GroupId == sg_id {
instances += fmt.Sprintf("%s,", *inst.InstanceId)
}
}
}
}
return instances
}
func (ec2 *EC2) Is_attached_to_elb(sg_id string) string {
elbs := ""
for _, lb := range ec2.Elbs.LoadBalancerDescriptions {
for _, sg := range lb.SecurityGroups {
if *sg == sg_id {
elbs += fmt.Sprintf("%s,", *lb.LoadBalancerName)
}
}
}
return elbs
}
func (c Config) Instance_is_excluded(role string) bool {
for _, inst := range c.Conf.Public_instances {
if( inst.Role == role){
return true
}
}
return false
}
func (c Config) User_is_allowed(user string) bool {
for _, u := range c.Conf.Iam.Users {
if( u.Username == user){
return true
}
}
return false
}
func (c Config) Ami_is_approved(ami string) bool {
for _, a := range c.Conf.Approved_amis {
if( a.Ami_id == ami){
return true
}
}
return false
}
func (i *AwsAuth) CheckPasswordPolicy(meta interface{}) string {
issues := ""
resp, err := meta.(*AWSClient).iamconn.GetAccountPasswordPolicy(nil)
if err!=nil {
panic(err)
}
if *resp.PasswordPolicy.MinimumPasswordLength < 14 {
issues += "Minimum Password length less than 14 chars,"
}
if resp.PasswordPolicy.PasswordReusePrevention == nil || *resp.PasswordPolicy.PasswordReusePrevention < 3 {
issues += "Last 3 Passwords can be reused,"
}
if *resp.PasswordPolicy.RequireUppercaseCharacters || *resp.PasswordPolicy.RequireNumbers || *resp.PasswordPolicy.RequireSymbols {
issues += "Password Policy doesn't require Uppercase Letters, Numbers and Symbols,"
}
if resp.PasswordPolicy.MaxPasswordAge == nil || *resp.PasswordPolicy.MaxPasswordAge < 90 {
issues += "Passwords don't expire after at least 90 days,"
}
return issues
}
func (i *AwsAuth) GetUsers(meta interface{}) {
params := &iam.ListUsersInput{}
i.Users, i.last_err = meta.(*AWSClient).iamconn.ListUsers(params)
if i.last_err != nil {
panic(i.last_err)
}
}
func (i *AwsAuth) GetUsersDetails(meta interface{}) {
i.UsersDetails, i.last_err = meta.(*AWSClient).iamconn.GetAccountAuthorizationDetails(nil)
if i.last_err != nil {
panic(i.last_err)
}
}
func (i *AwsAuth) GetCredentialsReport(meta interface{}) {
_, err := meta.(*AWSClient).iamconn.GenerateCredentialReport(nil)
if err == nil {
resp2, err2 := meta.(*AWSClient).iamconn.GetCredentialReport(nil)
if err2 == nil {
reader := csv.NewReader(bytes.NewReader(resp2.Content))
reader.Comma = ','
for {
record, err := reader.Read()
// end-of-file is fitted into err
if err == io.EOF {
break
} else if err != nil {
fmt.Println("Error:", err)
return
}
var t1,t2,t3 time.Time
if record[5] == "N/A" {
t1 , _ = time.Parse(time.RFC3339, record[2])
} else {
t1, _ = time.Parse(time.RFC3339, record[5])
}
if record[9] == "N/A" {
t2, _ = time.Parse(time.RFC3339, record[2])
} else {
t2, _ = time.Parse(time.RFC3339, record[9])
}
if record[14] == "N/A" {
t3, _ = time.Parse(time.RFC3339, record[2])
} else {
t3, _ = time.Parse(time.RFC3339, record[14])
}
i.CredReport = append(i.CredReport, &UserReport{UserName:record[0], PassLastRotation: t1, Key1LastRotation: t2, Key2LastRotation: t3})
}
}
}
}
func (a *EC2) GetSgs(meta interface{}) {
params := &ec2.DescribeSecurityGroupsInput{}
a.Security_groups, a.last_err= meta.(*AWSClient).ec2conn.DescribeSecurityGroups(params)
if a.last_err != nil {
panic(a.last_err)
return
}
}
func (i *AwsAuth) User_has_mfa(meta interface{}, username string) bool {
params := &iam.ListMFADevicesInput{
UserName: aws.String(username),
}
resp, err := meta.(*AWSClient).iamconn.ListMFADevices(params)
if err != nil {
return false
}
if len(resp.MFADevices)<1 {
return false
}
return true
}
func (i *AwsAuth) User_has_expired_key(meta interface{}, username string, exp_days int16) string {
keys_expired := ""
params := &iam.ListAccessKeysInput{
UserName: aws.String(username),
}
resp, err := meta.(*AWSClient).iamconn.ListAccessKeys(params)
if err != nil {
return keys_expired
}
if len(resp.AccessKeyMetadata)>0 {
for _, ak := range resp.AccessKeyMetadata {
if *ak.Status == "Active" {
params2 := &iam.GetAccessKeyLastUsedInput{
AccessKeyId: aws.String(*ak.AccessKeyId), // Required
}
resp2, _ := meta.(*AWSClient).iamconn.GetAccessKeyLastUsed(params2)
if resp2.AccessKeyLastUsed.LastUsedDate != nil {
if time.Now().Unix()-int64(exp_days*24*3600) > resp2.AccessKeyLastUsed.LastUsedDate.Unix() {
keys_expired += fmt.Sprintf("%s,", *ak.AccessKeyId)
}
} else {
if time.Now().Unix()-int64(exp_days*24*3600) > ak.CreateDate.Unix() {
keys_expired += fmt.Sprintf("%s,", *ak.AccessKeyId)
}
}
}
}
}
return keys_expired
}
func (e *EC2) Cloudtrail_is_enabled(meta interface{}) bool {
resp, err := meta.(*AWSClient).ctrailconn.DescribeTrails(nil)
if err != nil {
return false
}
if len(resp.TrailList)<1 {
return false
}
for _, t := range resp.TrailList {
params := &cloudtrail.GetTrailStatusInput{
Name: aws.String(*t.Name), // Required
}
resp2, err2 := meta.(*AWSClient).ctrailconn.GetTrailStatus(params)
if err2 != nil {
return false
}
if *resp2.IsLogging {
return true
}
}
return false
}
func (i *AwsAuth) User_has_password(meta interface{}, username string) bool {
params := &iam.GetLoginProfileInput{
UserName: aws.String(username),
}
resp, err := meta.(*AWSClient).iamconn.GetLoginProfile(params)
if err != nil {
return false
}
if (resp.LoginProfile.CreateDate.Unix() > 0) {
return true
}
return false
} |
<reponame>nariakiiwatani/VoectChat
import crypto from "crypto"
export function getHashString<T extends ArrayBuffer>(data: string, algorithm: string = "sha1"): string {
return crypto.createHash(algorithm).update(data).digest("hex");
}
export function makeQueryString(params: { [key: string]: string }): string {
return Object.entries(params).map(([k, v]) => `${k}=${v}`).join("&")
} |
Ever since I read an article about a group of guys named the Waldos claiming to be the originators of 420, I had doubts from the start as something didn’t make sense. Why did they choose 4:20 as the time to meet at the statue and who was the one that first coined the term? This never made any sense to me and I knew something wasn’t right. Suspecting there was a true originator, I began sending out energy for them to contact me for this article. Today was that day.
I received an email from a guy named Brad Bann aka “The Bebe”, claiming to be the Father of 420, saying that it all started in 1970 with a group of guys called “The Bebes”. They lived on a golf course, in a neighborhood called “Peacock Gap” in San Rafael, California.
Bebe says, “The Bebes and the Waldos are still good friends to this day, however it’s time the truth be told.” He goes on to say, “The Waldos were a group of guys I ordained”, referring to Steven Capper as the “Original Waldo”. He went on to explain how the Waldos were a small group of guys he dubbed “Waldos”, because they were goofy. “During the summer of 1970 at San Rafael High School, there were two groups of people involved in bringing forth the term 420, the Bebes and the Waldos. The Bebes beat the Waldos to the punch on nearly every phrase. The Waldos put a story on the Web in 1998, but not the real story. They never mentioned the Bebes because they would have some explaining to do.”
The Bebe and one of the other Bebes named, “Bone Boy”, sent their claim to High Times in 2003, after someone sent them the article they did on 420 and the Waldos. They waited for months, yet never received a reply.
420 Letter From Bone Boy The Bebe Is The Thomas Edison Of 420 With over 420 million Google results, the morphing of the number 420 into an international phenomenon is fairly baffling to me, an early ’70s graduate of San Rafael High School in Marin County, California. I’ve been sent numerous stories over the years regarding tracing the beginning of 420. I’ve listened to syndicated radio talk show hosts devote nearly entire shows to 420 on April 20th, heard radio shows celebrate each weekday with a Reggae tune at 4:20, and have seen 420 features on TV. How about the number being a police code, or penal code, or The Who’s 1965 album cover of “My Generation” in front of Big Ben at 4:20, clocks in the film, “Pulp Fiction” set to 4:20, etc… Anyway, I can tell you, as one who has firsthand knowledge of its true origin, that nearly everybody has been had… It is amazing to me that even sociologists have weighed in with their “expert” 420 viewpoint (and they get paid for this!?) of what it means. “420 creates an intense sense of group belonging among friends, strangers, and crowds” or “a kind of mystical, spiritual, or extraordinary sense of belonging, where the group exists as a reality greater than itself”… What? Way too stoned in Madagascar I’m afraid. True, the term initiated its international acronymic ascent in the early 1970’s- actually 1970 at San Rafael High School (SRHS). And the notions SRHS alumni left Marin County, taking 420 to the collegiate level, mostly along the West Coast, the I-5 and 101 corridors, and in fact, all the way past Seattle and up to Prudhoe Bay are true as well. I know, I took it to the Arctic Sea with an Arctic 4-twone standing on frozen water. The number also surfaced and spread throughout the Grateful Dead concert community, thanks in large part to the Waldos I’m certain, taking the term east, promoting the newly fabled number on a national scale. Deadheads are a great promotional vehicle, however this is about where the myths end and the truth takes off… I’m sorry, but the real story is rather short, unimpressive and unimaginative. It is spontaneous though, just like the character who first coined the term, completely by accident, like most things from his youth. Brad Bann aka The Bebe (Beeb – a nickname) about seventeen at the time, is the Father of 420 and many other terms that caught on around the campus of SRHS in the Fall of 1970. Take for instance “Waldos”, used in the current “420 Smokelore”, denoting a group of guys. It was a word originally made up to describe an odd, awkward, out of place person. Its predecessor was “Gome” or “Gomer”, after the TV character, Gomer Pyle and our neighborhood Gomer, Gary “Gomo” Schweitzer. When Bebe didn’t know you, he would call out, “Hey Waldo” or “Hey Gome”. Bebe also had specific nicknames for everyone and everything in our neighborhood: The Blue Boys, Puff, Du, Hello Andy, Turkey, Bone Boy, Thorgy, The Mead, Pig Newton, the Incredible Walking Man, and Bonfiglio (Or Bonfig), a tag he would use to address bearded hippies of the era. There was Dune (pot, taken from the title of a sci-fi novel) and “Alfred”, a term meaning, “To borrow, but never return”. He spawned Jimmy Bardoni, a fictitious name he would use whenever he got into hot water. Bebe not only concocted 420, he labeled the guys who claim to have created it. “The Waldos” (Though they were not the Waldo Father’s of 420) were perhaps the greatest promoters of the number back then. I mention all of this in an attempt to lay a foundation, I suppose, of a history of “Mindless Term Invention” by the Bebe. Quite simply, the birth of 420 occurred at precisely 4:20 in the afternoon to begin a bedroom bong session at the house of Du and Puff on a Saturday in October of 1970. The Bebe along with the brothers began preparing to “bong out”, when Bebe glanced at the clock on the nightstand and said, “It’s 4:20, time for bong loads”. After getting high, they proceeded to do some audio recording with Bebe, as we did frequently, using his assortment of voices, including his impression of Abraham Lincoln, and said as tape was rolling: “4 score and 20 years ago…” As it turned out, 420 became an instant code in our neighborhood. We gravitated to any and all Bebe terminology he conjured up. 420 seemed to just roll off the tongue better than any other number, 4:19, 2:37, 3:58 etc, and gosh knows we needed a code to use in front of non-stoners, especially for the parental establishment. I remember once Bebe saying, “It’s 420” in front of Hello Andy’s mother, and she responded by saying in a minor panic, “My goodness, it can’t be that late yet, I have to go pick up your sister!” As knucklehead teenagers, I guess we never realized parents were on such prompt time schedules. 420 developed its own lexicon, “Do you have any 4-twone?”, “Who’s got the 4-twone?”, “This is excellent 4-twone” or “420”, and “I’m too 4-twentyed”. Sign language and lip reading also evolved, as well as a hesitation code of sorts, where a person would say 4, then moments later, 2, followed shortly thereafter by a 0. There was the countdown as well 4-2- Zero. Way too stoned: “4-twoned”. It was also used to define certain kinds of weed, “420 Colombo” and “Thai 420” for Thai sticks. I submit the story just shared is the truth and nothing but the whole 420 truth. 420 was only designated as an actual time at the moment of its inception. It was never intended as a time of day to get high, but evolved into that as well as the coronation of April 20th into the stoner holiday all over the world. It was and will always be, first and foremost, just a simple code, period. 420 is an accidental anomaly. There is no deep meaning. A guy looked at a clock as he was about to “smoke out” with some buddies, blurted out the time, and it became local stoner lore. If Bebe would have said, “It’s twenty minutes after four”, the term probably would have never gotten legs to get out of the bedroom that day. 420 just sounded “Stoner Poetic”. The Waldos were a real group of guys, ordained by Bebe. One of them, Bebe referred to as the “Original Waldo” I believe. But now it’s time to examine the story. Of the Waldos in 1970, I believe one was a junior, one or two were sophomores and the others freshmen. Two of them I believe lived in the same neighborhood, and the others in different parts of town. One dude’s dad was a DEA agent as I recall. School finished at about 3pm, for some earlier. Some may have had sports after school, some didn’t. Now, let me get this straight: Guys are going to return an hour and a half after school was dismissed to meet at a statue, get high and go look for pot plants a lengthy drive away? If this is believable, you must be in possession of some excellent 420. Have you ever driven from San Rafael to Pt. Reyes? It’s about an hour each way (without commute traffic). So at 4:20, guys get together at the Louis Pasteur statue in the middle of campus, away from sports fields or gymnasiums, pile into a car and cruise out to Pt. Reyes looking for pot plants based on some map, then return? 4:20 seems kind of late to be driving an hour or so to look for pot plants. Might the sunset have interfered with their ability to find anything? What time did they get home, especially if they indulged in herb and did a bit of wandering out at Point Reyes? Which, by the way is highly likely. Of course they didn’t have any homework. If they played sports, how could one be sure practice was going to end close to 4:20? It is a reasonable assumption that practice for any sport lasts close to two hours, but wouldn’t the coach be the only one with that information? A meaningful question that should be asked of “The Waldos” is, “Why did you choose 4:20 as the time to meet at the statue as opposed to 4:15 or 4:30?” Seems so implausibly random. And, which of the Waldo’s was the one who first coined the term? One might suggest they chose the time because they gravitated to the Bebe’s new code and turned 420 into their time of day to partake in herb. But even that would be flawed, because they chose 4:20 as a time of day to meet and drive somewhere, not get stoned. Postmarks on letters is an interesting tool in tracing the beginnings of 420, but I’m sure that if Bebe saved any of his reel to reel tapes of prank phone calls, 420 would no doubt be heard. I was in Las Vegas with a friend during NCAA March Madness this year. As we were going down the elevator from our hotel room to the lobby, we stopped at several floors acquiring passengers along the way. Most were sporting shirts from their favorite teams. When I asked one young man what time his team played, he replied, “Game starts at 4:20”. Most of the eight other people in the elevator began to chuckle. People from various parts of the country knew exactly what the number represented. 420 is now a brightly colored number in the fabric of Pop Culture. When something enters, then becomes part of Pop Culture, the truth of where that “something” originated demands to be uncovered. That moment for 420 has been “now” for quite some time. How did it come to be? How was it intended to be used? Who first conceived and uttered the term? Who is responsible for creating this iconic three digit number? It’s a simple truth, really. Brad Bann aka The Bebe is the Thomas Edison of 420.”
The moment of truth. I finally had the answer I’d been seeking and it was time to set the record straight, once and for all.
420 Interviews
To verify credibility, I began contacting all of The Bebes, gathering information on the story. I confirmed all of their full names and identities, some of which asked to keep secret for personal and/or professional reasons. After interviewing 10 Bebes and hearing all of their stories, it was easy to conclude that there was definitely a hidden truth, that needed to be revealed. Everyone told the exact same story!
Meet Dave Dixon aka Wild Du, one of the Dixon brothers who was there when Bebe coined the term 420. Bebe describes Wild Du as, “a “Core Bebe”, who I sold knife sets to businesses up and down the California Coast with, just out of high school. Him and I started “The 420 Band” in 1972, and still play to this day.” Wild Du says, “I first met Bebe at the neighborhood gathering when we were 15. We went to the brick yard to play and Bebe started throwing bricks and poking holes in the mortar with his fingers, causing a ruckus, which ended up in getting us both arrested.” He went on to say that the Waldos have admitted that the Bebes coined the term 420 to him, just not to the public.
Dave is now 58 years old, still lives in San Rafael, California grinding knives and playing guitar with Bebe.
Then there’s Wild Du’s brother Dan Dixon aka Puff, the other Dixon brother who was there when Bebe coined the term 420. Bebe describes Puff as another “Core Bebe”, saying, “Puff was popular with the Bebes and the Waldos, this is how all the words the Bebes made up, became language that the Waldos ended up using, like 420. Puff and I were in the Army in Germany together and were always seen together during high school.” Puff says, “The Waldos admit that the Bebes coined the term 420, there is no question. They even tried to recruit me, to make their story more credible.”
Dan is now 57 years old and lives in Oklahoma, where he is a retired caregiver for his mother in law. He spent his career as a Basketball Coach, later to become a Pharmacy Tech, providing health care to those in need. He loves football, basketball, golf and 420.
When brothers Wild Du & Puff were asked to recall the exact Saturday in October, 1970, they both remembered puffing in the house with Bebe on those particular weekends and confirm being there when Bebe coined the term, however were unable to pin down the exact day. Wild Du thinks it happened on the first (10/03/70) or second (10/10/70) Saturday of October, because it was the beginning of school.
Tom Thorgersen aka Thorgy was the neighborhood Norwegian weed dealer, who handled all of the Bebe’s 420 needs. Bebe recalls, “Thorgy was the big 420 weed dealer in the 70s & 80s and the Bebes spent a lot of time at Thorgy’s.” Thorgy started smoking at 12 years of age, to calm his hyperactivity. His mom even offered to help him grow it. He shared stories about calling in to radio shows with the Bebe and doing 420 pranks on the air, and listening to the reel to reel audio recordings of Bebe’s version of Abraham Lincoln’s address, “4 Score and 20 Years”. Thorgy says, “Steve Capper is an opportunist who wasn’t even close to making up 420. We made fun of the Waldos, aka “Wallies”, they were the weaker link, the ones who didn’t fit in. It will be nice to finally have the truth be told.”
Tom is now 55 years old and still lives in San Rafael, California. He is a Carpenter with a passion for rebuilding old cars.
Dave Anderson aka Hello Andy was, “The main spokesman for the Bebes, who tried desperately to organize and explain Bebe behavior”, says Bebe. Hello Andy recalls, “Everything Bone Boy said is true. Bebe was always making up nicknames for everyone and spent a lot of time in his bedroom making prank recordings. He was always making weird sounds and was great with voices. One time we made this recording of police calls where Bebe would say stuff like, “One Adam Twelve. We have a 420 on 4th Street. Send 2 units. Over”. He would do things like aiming rocks at a target, looking long and hard at it, then saying something like, “Estimating angle 420”, then throw it. Hello Andy goes on to say, “I lived in between Bebe and Du & Puff, and took part in plenty of bong outs at their house. It’s highly likely that I could have been there at the time he first said it, however there is no question that Bebe certainly coined the term 420, which later became our special code. To be honest with you, I never even considered Steve Capper a Waldo.
Dave is now 57 years old, lives in Sacramento, California and is an Engineer. He likes Golf, sports, music, movies and a little 420 from time to time at a concert or special occasion like the Bebe’s 420 reunion.
Bone Boy was the designated driver, chauffeuring The Bebes around in his Blue 66′ Barracuda, blasting 8-Tracks of Hendrix, Zeppelin, The Who, Santana, Doobie Brothers & more. Bone Boy says, “I never allowed toking out in my car, so we would drive around Marin County looking for scenic places to take the car and get high. Some of our favorite spots were, “The Wall”, “Windless”, Sequoia” and “360”.” Bebe says, “Bone Boy always had transportation and planned events. He was very good in sports, loved music and always had new albums. All the Bebes were good in sports and had very good communication skills. We all used Sonics, a loud piercing noise that Bebes could identify and locate each other, saved our asses more than once.” Bone Boy says, “We would go to Baskin Robbins and Bebe would make this high pitch sonic noise. People would just look around and wonder where it came from. He was always screwing with people, in an odd, fun way.” He goes on to say, “We lived on a golf course surrounded by houses and Bebe was always doing something crazy. One day, he showed up with a golf cart. When we asked where he got it, he said, “Don’t worry, nobody pays attention.” Bone Boy says, “I asked a teacher from San Rafael High School if they remembered The Bebe and he said they used to have staff meetings about Bebe and his pranks, all of the time”.
Bone Boy is now 57 years old, lives in Huntington Beach, CA and is a music industry veteran. He loves film, music, writing, teaching and the great outdoors.
Turkey was from Georgia and spoke differently with a southern twang. Bebe recalls, “He always had to go home early and would say, “my ass is grass”, then run home. He had a mini bike that would go 42.0 mph, some of our first transportation.”
Then there was The Worm, who Bebe says had a prosthetic arm and used to play tackle football with them. “I love that guy, so full of life. He was a real game person, many stories about and with him. He would love to be part of this.”
Blue and The Mead were two anti-social brothers called Blue Boys, who were a few years younger. Bebe says, “Blue Boy” was a term we gave to the younger guys who hung out with us.” Hello Andy recalls, “One time I watched Blue Boys, Blue and Scraun play a prank on the coach. They watched their watches and when the time came, they asked, “Hey coach, what time is it?” He replied, “4:20″ and they all started giggling.”
And finally we come to Brad Bann aka The Bebe. He was known as a prankster back in high school. He loves music, sports, and scientific facts. He still lives in San Rafael, California where he plays guitar and is the lead singer in a band, doing Frank Sinatra covers. When Bebe isn’t playing live gigs, you can still find him in his studio making funny songs and recordings. Today is actually his 58th Birthday. What a perfect time to reveal his hidden truth to the world. Hempy Birthday, Brother Bebe!
When asked if anyone possibly had any of the reel to reel prank calls and/or random audio recordings of The Bebe with 420; Bebe lost all of his, Wild Du’s were stolen and Bone Boy’s mysteriously disappeared. Unfortunately, nobody else had any recollection of having any of these recordings, however with the release of this new truth, hopefully some of them will manifest in the days to come.
420 Conclusion
The Bebes were a group of athletes from San Rafael, CA, who went to San Rafael High School in 1970. They lived in the same neighborhood called Peacock Gap, which was a golf course, surrounded by houses. They were well known for their prank phone calls and recordings. Brad Bann aka The Bebe was the leader of the group and was joined by all of his friends, whom he ordained and named as well. There was Dave Dixon aka Wild Du, his brother Dan Dixon aka Puff, Dave Anderson aka Hello Andy, Tom Thorgersen aka Thorgy, Bone Boy, Blue, The Mead, Turkey & The Worm.
The Waldos were a group of guys from San Rafael, CA, who went to San Rafael High School in 1970. They were known for being goofy, which is why The Bebe nicknamed them all Waldos. There was Steve Capper and David Reddix who have gone public with their names, Patrick, Larry, Jeff, John and Mark, who have not gone public as of yet. While these guys may have been responsible for promoting 420 across country, there is no question that they did not coin the term and have been dishonest with the world from day one. True credit goes to The Bebe and his brotherhood of Bebes.
420 Closing
One thing is certain to me, Brad Bann aka the Bebe coined the term 420 and the Waldos carried the term across the U.S. on tour with the Grateful Dead. I took the torch in 1993 and promoted 420 to the world via my website/s, reaching over 20 million people a year, totaling over 420 million people worldwide. Now there are billions of us.
People have asked me the same question, 420 million times, “What is 420?” The most common reply was usually an hour long explanation of 420 different things that it is and can be. After 20 years of promoting this magical number, I’ve come to summarize it down to, “It’s anything you want it to be.”
It’s been 42.0 years since Bebe first coined the term 420 and I am very honored and truly grateful for being chosen to send his message to the world, setting the record straight once and for all.
It all makes sense now.
Rob Griffin
Editor-in-Chief
420 Magazine
PO Box 3420
Hollywood, CA 90078
1-888-420-MAGG
http://www.420magazine.com
http://www.twitter.com/420magazine
http://www.facebook.com/420magazine
Contact: editor@420magazine.com
Photo Gallery of Bebes
All content copyright © 1993-2013 420 Magazine ® All Rights Reserved |
Information management and technology in England's large acute NHS hospitals. National strategy versus local reality. Discusses the NHS Executive's information management and technology (IM&T) strategy and its relationship to the 1991 reforms. Examines the recommendation for large acute hospitals to adopt integrated hospital information support systems (HISS). Reports that a recent census of these hospitals, undertaken by the authors, suggests that the implementation of the strategy's recommendations has been slow at the local level. Attempts to diagnose the factors that are impeding implementation, using the evidence provided by the census. Identifies four main problem areas: the lack of success of past IM&T initiatives undermines confidence in the current strategy; the strategy is poorly aligned with other policy initiatives; the legacy of discrete, proprietary information systems within hospitals makes the creation of an integrative information environment difficult to accomplish without massive investment in new systems; and there are implicit contradictions between the following: the absence of a comprehensive post-implementation evaluation of the economic, technological and cultural feasibility of HISS at any of the three HISS pilot sites; the strategy's advocacy of HISS as the way forward for large acute hospitals; the requirement for a comprehensive business case to support any substantial investment in IM&T. Concludes that a massive rethink of policy is required, with a much greater emphasis on research, development and independent evaluation. |
Towing vehicles or trailers are designed to secure and haul cargo. Trailers may be arranged to haul various types of cargo, such as boats, automobiles, consumer products, and the like. Many such cargo items may be large, heavy and difficult to move or maneuver onto the bed or frame of a towing trailer. To assist in moving or maneuvering the cargo onto the towing trailer, such trailers may often be equipped with a winch or winch assembly.
In boat trailers, a stand or base member typically extends upwardly from a front portion of the trailer and a head member is attached to the base member and extends rearwardly therefrom. A bow stop is attached to an end portion of the head member opposite the base member and engages with the bow of a boat to limit the positioning of the boat on the trailer. The winch is commonly attached to the head member between the bow stop and the base member. The winch may be connected to the boat by, for example, a strap, cable, rope, chain or the like that may aid in pulling the boat onto the trailer. The winch may be used to assist in the final positioning of the boat onto the trailer. This type of situation may commonly occur while using a boat trailer to remove a boat from a body of water.
These traditional boat trailers require the assembly of the base member to the trailer, the head member to the base member, the bow stop to the head member and the winch to the head member. These steps are time consuming and may provide for a poor aesthetic quality. Additionally, these various components must be individually selected and aligned in a proper orientation to allow for a preferred positioning of a particular boat on the trailer. |
Random growth lattice filling model of percolation: a crossover from continuous to discontinuous transition A random growth lattice filling model of percolation with touch and stop growth rule is developed and studied numerically on a two dimensional square lattice. Nucleation centers are continuously added one at a time to the empty sites and the clusters are grown from these nucleation centers with a tunable growth probability g. As the growth probability g is varied from 0 to 1 two distinct regimes are found to occur. For g\le 0.5, the model exhibits continuous percolation transitions as ordinary percolation whereas for g\ge 0.8 the model exhibits discontinuous percolation transitions. The discontinuous transition is characterized by discontinuous jump in the order parameter, compact spanning cluster and absence of power law scaling of cluster size distribution. Instead of a sharp tricritical point, a tricritical region is found to occur for 0.5<g<0.8 within which the values of the critical exponents change continuously till the crossover from continuous to discontinuous transition is completed. I. INTRODUCTION A new era in the study of percolation has started in the recent past developing a series of new models such as percolation on growing networks, percolation in the models of contagion, k-core percolation, explosive percolation, percolation on interdependent networks, agglomerative percolation, percolation on hierarchical structures, drilling percolation and two-parameter percolation with preferential growth. In these models, instead of robust second order (continuous) transition with formal finite size scaling (FSS) as observed in original percolation, a variety of new features are noted. Sometimes the transitions are characterized as a first-order transition, sometimes a crossover from second order to first order with a tricritical point (or region) are observed [6,13,, sometimes features of both first and second order transitions are simultaneously exhibited in a single model, sometimes second order transitions with unusual FSS are found to appear. Such knowledge not only enriches the understanding of a variety of physical problems but also leads to creation of newer models beside extension of the existing models for deeper understating of the existence of such non-universal behavior. In this article, we propose another novel model of percolation namely "random growth lattice filling" (RGLF) model adding nucleation centers continuously to the lattice sites as long as a site is available and growing clusters from these randomly implanted nucleation centers with a constant but tunable growth probability g. RGLF can be considered as a discrete version of the continuum space filling model (SFM) with the touch and stop rule in the growth process as that of the growing cluster model * santra@iitg.ernet.in (GCM). However, RGLF displays a crossover from continuous to discontinuous transitions as the value of g is tuned continuously from 0 to 1 in contrast to both SFM and GCM which display a second order continuous transition. Below we present the model and analyze data that are obtained from extensive numerical computations. II. THE MODEL A Monte Carlo (MC) algorithm is developed to study percolation transition (PT) in RGLF defined on a 2dimensional (2D) square lattice. Initially the lattice was empty except one nucleation center placed randomly to an empty site. In the next time step, besides adding a new nucleation center randomly to another empty site, one layer of perimeter sites of all the existing active clusters including the nucleation center implanted in the previous time step are occupied with a constant growth probability g following the Leath algorithm. The process is then repeated. A cluster is called an active cluster as long as it remains isolated from any other cluster or nucleation center at least by a layer of empty nearest neighbors. Each cluster (active or dead) are marked with a unique label. At the end of a MC step, if an active cluster is found separated by a nearest neighbor bond from another cluster (active or dead), they are merged to a single cluster and they are marked as a single dead cluster. The growth of a dead cluster is seized for ever as in GCM. If a peripheral site is rejected during the growth of an active cluster, it will be not available for occupation by any other growing cluster as in ordinary percolation (OP). However, such a site can be occupied by a new nucleation center added externally. The growth of a cluster stops due to the fact that either it becomes a dead cluster by merging with another cluster or all its peripheral sites become forbidden sites for occupation. The process of lattice filling stops when no lattice site is available to add a nucleation center. Time in this model is equal to the number of nucleation centers added. Therefore, for any value of g, there will always be a PT in the long time limit. The model with g = 0 naturally corresponds to the Hoshen-Kopelman percolation as the instantaneous area fraction p(t) reaches the OP threshold p c (OP) ≈ 0.5927 and exhibits a continuous second order PT. For g = 0, the area fraction p(t) at time t is nothing but the number of nucleation centers added per lattice site up to time t whereas for g = 0, it is the number of occupied sites per lattice site at time t. Such a continuous transition is expected to occur as long as the growth probability g remains below p c (OP). It can be noted here that in SFM, PT occurs only at unit area fraction in the limit of growth probability tending to 0. As g is increased beyond p c (OP), large clusters appear due to the merging of compact finite clusters originated from continuously added nucleation centers. As a result, the system will lack small clusters as well as power law distribution of cluster size at the time of PT. Such a transition will occur with a discrete jump in the size of the spanning cluster due to the merging of compact large but finite clusters. Hence, it is expected to be a discontinuous first-order transition. A smooth crossover from continuous transitions to discontinuous transitions is then expected to occur as the growth parameter g will be tuned from below p c (OP) to above p c (OP). III. RESULTS AND DISCUSSION Extensive computer simulation of the above model is performed on 2D square lattice of size L L. The size L of the lattice is varied from L = 64 to 1024 in multiple of 2. Clusters are grown applying periodic boundary condition (PBC) in both the horizontal and the vertical directions. All dynamical quantities are stored as function time t, the MC step or the number of nucleation centers added. Time evolution of the cluster properties are finally studied as a function of the corresponding area fraction p(t) instead of time t directly. Averages are made over 2 10 5 to 5 10 3 ensembles as the system size is varied from L = 64 to 1024. The snapshots of cluster configurations are taken just prior to the appearance of the spanning cluster in the system and are shown in Fig.1 for g = 0.4 (a), and g = 0.8 (b) on a 2D square lattice of size L = 64. The largest cluster is shown in red and the other smaller clusters of different sizes are depicted in other different colors. At the lower growth probability g = 0.4, clusters of many different sizes exist along with a large finite clus- ter. Smaller clusters are found to be enclaved inside the larger clusters. PT occurs in the next step and no significant change in the largest cluster size is expected as the largest cluster in the previous step was about to span the lattice. Such continuous change in the largest cluster size along with enclaved smaller clusters within it are indications of continuous transition. On the other hand, as the growth probability is taken to be high g = 0.8, clusters of smaller sizes are merged with the fast growing other finite clusters. As a result, only a few large compact clusters are found to exist beside the newly planted nucleation centers. Clusters of intermediate sizes are found to be absent. Enclave of smaller clusters by the larger clusters has almost disappeared. As the clusters in cyan and red are merged in the next step and generates a spanning cluster, the PT occurs with a significant change in the size of the largest cluster corresponding to a jump in the largest cluster size at the time of transition. Appearance of compact large cluster with discontinuous jump in the largest cluster size are indications of a discontinuous transition. The time evolution of the size of the largest cluster S large (t) is monitored against the instantaneous area frac-tion p(t) for three different samples for a given g. Their variations are shown in Fig.2(a) for g = 0.4 and for g = 0.8 in Fig.2(b) for a system of size L = 512. The average area fractions corresponding to the thresholds at which PT occur in these systems with given g are marked by the crosses on the respective p(t)-axis. Around the respective thresholds, the evolution of the size of the largest clusters for g = 0.4 and g = 0.8 are drastically different. For g = 0.4, the size of the largest cluster in different samples are found to increase almost continuously with the instantaneous area fraction p(t) around the threshold. This indicates a continuous PT to occur. However, for g = 0.8, S large grows with discontinuous jumps at the threshold with the largest jump of the order of 10 5 ≫ L. The effect would be more prominent with higher values of g. This is another indication of a discontinuous PT. It is then intriguing to study the model with varying the growth probability g and characterize the nature of transitions at different regimes of g. B. Fluctuation in order parameter The dynamical order parameter P ∞ (t), the probability to find a lattice site in the spanning cluster, is defined as where S max (t) is the size of the spanning cluster at time t. The finite size scaling (FSS) form of P ∞ (t) is given by where P ∞ is a scaling function, is the order parameter exponent, is the correlation length exponent and p c (g) is the critical area fraction for a given growth probability g at which a spanning cluster connecting the opposite sides of the lattice appears for the first time in the system. Following the formalism of thermal critical phenomena as well as several recent models of percolation, the distribution of P ∞ is taken as where P is a scaling function. With such a scaling form of P , one could easily show that The fluctuation in P ∞ (t) at an area fraction p(t) is defined as The FSS form of ∞ (t) is given by where / = d − 2/ is the ratio of the average cluster size exponent to the correlation length exponent, d is space dimension and is a scaling function. In Fig. 3, ∞ (t)/L 2 is plotted against p(t) for different lattice sizes at two extreme values of g: g = 0.1 in Fig.3(a) and g = 0.8 in Fig.3(b). There are two important features to note. First, each plot has a maximum at a certain value of p(t) for a given g and L. The locations of the peaks correspond to the critical thresholds p c (L) (marked by crosses on the p(t) axis) at which a spanning cluster appears for the first time in the system. The critical area fraction p c (L) is expected to scale with the system size L as where is the correlation length exponent, as it happens in OP. In the limit L → ∞, the value of p c (L) becomes p c (g), the percolation threshold of the model for a given g. In the insets of respective figures, p c (L) is plotted against L −1/ taking 1/ = 0.75, that of the OP, for different values of g. The scaling form given in Eq.7 is found to be well satisfied for g ≤ 0.5 with 1/ = 0.75, inset of Fig.3(a). For g ≤ 0.5, the linear extrapolation of the plots of different g intersect the p c (L) axis at different p c (g) values. Whereas for g ≥ 0.8, the data do not obey Eq.7 and no definite p c (g) is found to exist in the L → ∞ limit. Such deviation from the scaling form given in Eq.7 is found to occur for the systems those are grown with g > 0.5 too. Second, the peak values of ∞ (t)/L 2 are decreasing with increasing L for g = 0.1 as in continuous transitions whereas they remain constant with L for g = 0.8 as in discontinuous transitions. In order to extract the exponent / for a system with a given value of g, the peak values of the fluctuation ∞ (max) are plotted against L for g ≤ 0.5 in Fig.4(a) and for g ≥ 0.8 in Fig.4(b) in double logarithmic scale. The magnitudes of ∞ (max) are found to be independent of g at a given L for g ≤ 0.5 whereas they increase with g at a given L for g ≥ 0.8. As per the scaling relation Eq.6, a power law scaling ∞ (max) ∼ L / is expected to follow at the threshold. The exponent / is determined by linear least square fit through the data points. For g ≤ 0.5, it is found to be / = 1.79 ± 0.01 whereas for g ≥ 0.8, it is found to be / = 2.0 ± 0.01. The solid straight lines with desire slopes in Fig.4(a) and (b) are guide to eye. It is important to note that the value of / for g ≤ 0.5 is that of the OP (43/24) which indicates continuous transitions whereas for g ≥ 0.8 it is that of the space dimension which indicates discontinuous transitions. For 0.5 < g < 0.8, the exponent / is found to change continuously from 1.79 to 2.0 indicating a region of crossover. The values of / for different values of g are also verified by estimating the average cluster size of all the finite clusters (excluding the spanning cluster) at their respective percolation thresholds. However, there are evidences in some other percolation models such as k-core percolation that the scaling behavior of order parameter fluctuation and that of the average cluster size are not identical. The FSS form of ∞ (t) is verified plotting the scaled fluctuation ∞ (t)/L / against the scaled variable L 1/ for g = 0.1 in Fig.5(a). A good collapse of data is obtained for / = 1.79 and 1/ = 0.75 as those of OP. Whereas, for g = 0.8, a partial collapse is obtained for the plots of ∞ (t)/L / against the scaled variable L 1/, as no p c (g) is available, taking / = 2.0 and tuning the value of 1/ to 0.5 as shown in Fig.5(b). The collapse of the peak values confirms the values of the scaling exponent / as 1.79 for g ≤ 0.5 and 2.0 for g ≥ 0.8. Following the scaling relation / = d − 2/, the exponent / should be 0.105 as that of OP for g ≤ 0.5 and zero as that of a discontinuous PT for g ≥ 0.8. C. Phase diagram A phase diagram separating the percolating and nonpercolating regions is obtained by plotting the variation of p c (L) against g for a system of size L = 1024 in Fig.6(a). It is interesting to note that the critical area fraction has a minimum at a growth probability little above the threshold of OP, p c (OP)≈ 0.5927 and it is as low as ≈ 0.45. It is obvious that area fraction would be ≈ 0.6 at the criticality when g = 0. If g is finite but small, growth of small clusters will stop mostly because of less growth probability beside rarely merging with another small cluster or a newly added nucleation center. A large number of smaller clusters will be there in the system before transition and merging of such small cluster will lead to a spanning cluster which will have many voids in it. As a result, the area fraction will be less. Such an effect will be more predominant when g is around the percolation threshold of OP as at this growth probability large fractal clusters will be grown. PT occurs due to merging of such large fractal clusters which will contain maximum void space in it. Hence, the area fraction is expected to be the lowest. Beyond, such growth probability, compact clusters start appearing which will occupy most of the space at the time of transition. Area fraction will increase almost linearly with g in this regime. Such variation of p c is also observed in a percolation model with repulsive or attractive rule in site occupation. The phase diagram is then complemented by the variations of P ∞ (t) against p(t) for various values of g which are shown in Fig.6(b). Not only the the critical threshold decreases with increasing g and takes a turn at g ≈ 0.65 but also the transitions become more and more sharper as g increases beyond g ≈ 0.65. It is also interesting to note that values P ∞ (t) at p c also increases with increasing g even when the critical area fraction (p c ) is decreasing. Therefore the spanning cluster mass is always increases with the growth probability g. D. Dimension of spanning cluster For system size L ≪, the size of the spanning cluster S max at the criticality varies with the system size L as where d f is the fractal dimension of the spanning cluster. Since the clusters are grown here applying PBC, the horizontal and vertical extensions of the largest cluster are stored. If either the horizontal or the vertical extension of the largest cluster is found to be greater than or equal to L, it is identified as a spanning cluster. The value of S max are noted at the respective thresholds for several lattice sizes L for a given g. For a continuous PT, the spanning cluster is a random object with all possible sizes of holes in it and is expected to be fractal whereas in the case of a discontinuous transition it becomes a compact cluster. The values of S max are plotted against L in double logarithmic scale for the different values of g ≤ 0.5 in Fig. 7(a) and for g ≥ 0.8 in Fig. 7(b). For g ≤ 0.5, S max scales with L as a power law with d f = 1.895 ± 0.002 almost that of OP (91/48). On the other hand, for g ≥ 0.8, S max scales with L as a power law with d f = 2.0 ± 0.01 as that of space dimension d. The solid lines with desire slopes 1.896 and 2.0 in Fig.7(a) and (b) respectively are guide to eye. Thus for g ≤ 0.5, the spanning cluster is found to be fractal as in OP whereas for g ≥ 0.8 they appear to be compact as expected in a discontinuous transition. As a result, there would be enclaves in spanning clusters for g ≤ 0.5 whereas such enclaves would be absent in the spanning clusters for g ≥ 0.8 as it is evident in the cluster morphology shown in Fig. 1. Such presence or absence of enclaves in the spanning cluster determines whether it would be fractal or compact which essentially determines the nature of the transition as continuous or discontinuous. In the regime 0.5 < g < 0.8 the dimension of spanning cluster d f changes continuously from d f = 1.895 to d f = d = 2.0. The FSS form of P ∞ (t) given in Eq.2 as L −/ P ∞ should scales with the system size L as P ∞ (t) ∼ L −/ at the criticality where / = d − d f. As the value of d f is found to be 1.895 for g ≤ 0.5 and 2.0 for g ≥ 0.8, it is expected that the order parameter exponent / should be 0.105 as that OP (5/48) for g ≤ 0.5 and zero for g ≥ 0.8 leading to discontinuous jump. A continuous variation in / is expected in the regime 0.5 < g < 0.8. Variation of P ∞ (t) is plotted against p(t) for different lattice sizes for g = 0.1 in Fig. 8(a) and for g = 0.8 in Fig. 8(b). As the system size L increases, P ∞ (t) becomes sharper and sharper for both g = 0.1 and g = 0.8. However, the plots of P ∞ (t)L / cross at a particular value of p(t) corresponding to the critical threshold p c (g) taking / = 0.105 for g = 0.1 as shown in inset-I of Fig. 8(a). As / = 0 for g = 0.8, by no means they could make cross at a definite p(t). However, for g = 0.1, after re-scaling the P ∞ (t) axis if the p(t) axis is re-scaled as L 1/ taking 1/ = 0.75 a complete collapse of data occurs as shown in inset-II of Fig. 8(a). Whereas, for g = 0.8, collapse of P ∞ (t) plots are obtained by re-scaling only the p(t) axis as L 1/ taking 1/ = 0.5 as shown in the inset of Fig. 8(b). Such collapse of data not only confirms the validity of the scaling forms assumed but also confirms the values of the scaling exponents obtained. The observations at g = 0.1 are found to be the same for all g ≤ 0.5 and those are at g = 0.8 are same for g ≥ 0.8. Though discontinuous jump in the order parameter is also observed in SFM, the PT is characterized as continuous. On the other hand, in GCM, discontinuous transition is found to occur only in the vanishingly small fixed initial seed concentration but for intermediate seed concentrations the transitions are found to continuous that belong to OP universality class. For 0.5 < g < 0.8, collapse of data is observed for continuously varied exponents that depend on g as also seen in Ref.. The variations of the critical exponents /, / and fractal dimension d f with the growth probability g are presented in Fig. 9. The values of the critical exponents clearly distinguishes the discontinuous transitions for g ≥ 0.8 from the continuous transitions for g ≤ 0.5. F. Binder cumulant The evidences presented above indicate a continuous transition for g ≤ 0.5 and a discontinuous transition for g ≥ 0.8. In order to confirm the order of transition in different regimes of the growth probability g, a dynamical Binder cumulant B L (t), the fourth moment of S max (t), is studied as function of area fraction p(t). The dynamical Binder cumulant B L (t) is defined as The cumulants when plotted against the area fraction p(t) for different system sizes L are expected to cross each other at a definite p(t) corresponding to the critical threshold of the system for a continuous transition whereas no such crossing is expected to occur in the case of a discontinuous transition. Though the cumulant has some unusual behavior, it is rarely used in the study of recent models of percolation. The values of B L (t) are plotted against p(t) for different system sizes L in Fig. 10(a) for g = 0.1 and in Fig. 10 The FSS form of B L (t) is given by where B is a universal scaling function. The FSS form is verified by obtaining a collapse of the plots of B L (t) against the scaled variable L 1/ taking 1/ = 0.75 for g = 0.1. For g = 0.8, however, such a collapse is obtained when the cumulants are plotted against L 1/ taking 1/ = 0.5. The data collapse is shown in the insets of the respective plots. Such scaling behavior of B L (t) for g = 0.1 is found to occur for the whole range of g ≤ 0.5 and that of g = 0.8 is found to occur for g ≥ 0.8. Once again, Binder cumulant provides a strong evidence that the dynamical transition is continuous for g ≤ 0.5 whereas it is discontinuous for g ≥ 0.8. For 0.5 < g < 0.8, a region of crossover, the cumulants do not cross at a particular value of p(t) rather they cross each other over a range of p(t) values but do collapse when plotted against the scaled variable L 1/ for the respective value of 1/ for a given value of g. G. Cluster size and order parameter distributions Power law distribution of cluster sizes at the critical threshold is an essential criteria in a second-order continuous phase transition. Following OP, a dynamical cluster size distribution n s (t), the number of clusters of size s per lattice site at time t, is assumed to be where and are two exponents and f is a universal scaling function. For OP, an equilibrium percolation model, the exponents are = 187/91 and = 36/91. The distribution at the percolation threshold n s (p c ) is expected to scale as ≈ s −. The cluster size distributions n s (p c ) are determined taking p c (g) as threshold for g ≤ 0.5 and taking p c (L) as threshold for g ≥ 0.8 for a system of size L = 1024. The data obtained are binned of varying widths and finally normalized by the respective bin widths. In Fig. 11(a), the distributions n s (p c ) are plotted against s in double logarithmic scale for g ≤ 0.5 (0.4 (green) and 0.5 (magenta)) and for g ≥ 0.8 (0.9 (orange) and 1.0 (blue)) for L = 1024. It is clearly evident that the distributions for g ≤ 0.5 describes a power law behavior whereas for g ≥ 0.8 the distributions develop curvature and deviate from power law scaling. In the inset, the measured exponent s = ∂ log 10 n s (p c )/∂ log 10 s is plotted against log 10 s. The value of s remains constant to ≈ 2.055 as that of OP over a wide range of s for g ≤ 0.5 whereas s varies with s for g ≥ 0.8 indicating no definite value of. The existence of a crossover from continuous transition of OP type to a discontinuous percolation transition is further confirmed by the value of in the different regimes of the growth probability g. This is in contrary to the observations in SFM or cluster merging model where a power law distribution of clusters size is found to occur beside discontinuous transition. Beside the cluster size distribution, distribution of order parameter is also studied for different values of g as usually it is studied in thermal phase transitions where a bimodal distribution of order parameter is expected in a discontinuous transition whereas single peaked distribution is obtained in a continuous transition. An ensemble of largest clusters on different configurations are collected at the percolation threshold of a given g and the values of the order parameter P large = S large /L 2 are estimated. A probability distribu-tion P (P large ) is then defined as P (P large ) ∼ L / P where P is a scaling function. Bimodal nature of P is found to be a powerful tool to distinguish discontinuous transitions from continuous transitions in some of the recent percolation models. The distributions of P (P large )s are plotted in Fig. 11(b) for different values of g. For g ≥ 0.8, instead of sharp bimodal distributions, broad distributions with two weak peaks are obtained. No FSS of the distributions is found as given Eq.12 but the width of the distribution ∆ = 2 1/2 for a given g is found to increase with the system size L, shown in the inset of Fig. 11(b), as a signature of discontinuous transition. For a given L, the width of the distributions ∆ is also found to increase with g. However, the distributions P (P large ) for g ≤ 0.5 are found to be single humped and follow the scaling form given in Eq.12 as shown in the other inset. The width of the distributions for a given g ≤ 0.5 is found to decrease with L. The model, thus, exhibits characteristic properties of discontinuous transition for g ≥ 0.8 and those of continuous transition for g ≤ 0.5. IV. CONCLUSION In a dynamical model of percolation with random growth of clusters from continuously implanted nucleation centers through out the growth process with touch and stop rule, a crossover from continuous to discontinuous PT is observed as the growth probability g tuned from 0 to 1. For g ≤ 0.5, the order parameter continuously goes to zero and the geometrical quantities follow the usual FSS at the critical threshold with the critical exponents that of OP. The cluster size distribution is found to be scale free and a single humped distribution of order parameter occurred in this regime of g. On the other hand, for g ≥ 0.8, the PT occurs with a discontinuous jump at the threshold, the order parameter fluctuation per lattice site becomes independent of system size, the spanning cluster becomes compact with fractal dimension d f = 2 as that of discontinuous transitions. No scale free distribution is found for the cluster sizes and the order parameter distribution is weakly double humped broad distribution of increasing width with the system size. The order of transitions in different regimes of g are further confirmed by the estimates of Binder cumulant. The intermediate regime of growth probability 0.5 < g < 0.8 remains a region of crossover without a definite tricritical point. |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.