content
stringlengths 7
2.61M
|
|---|
def ConstrainToTargetPlane(self):
pass |
ORIGINAL ARTICLE ON THE MORPHOLOGICAL STUDY OF FORAMEN MAGNUM Introduction: Centrally in the deepest part of posterior cranial fossa is the largest foramen, Foramen magnum surrounded by basilar part of occipital bone on either side. Because of relation between the FM and the vital structures passing through it, study on its morphometric features is of great signicance. Aim and objectives: The objectives were to study the various morphological features of the foramen magnum in dry skulls using an analogue Vernier calliper. Materials and methods: 50 dry skulls (8 base skulls, 42 full skulls) of human cadaver of unknown age and sex were obtained to study the morphometric features like shapes, anteroposterior and transverse diameters and FM index in the department of Anatomy, Kanyakumari Government medical college, Asaripallam. Results: The classication of determined shapes were round in 29.7%, hexagonal in 18.2%, egg shaped in 16.9%, oval in 12.7%, tetragonal in 11.4%, pentagonal in 3.7% and irregular in 7.4%. In 12% of the skulls the occipital condyles were found to protrude into the foramen. The mean value of anteroposterior and transverse diameter was found to be 35 ±1.2mm, and 28 ± 1.4 mm respectively and average foramen magnum index was 1.25 ± 0.8. Conclusion: Foramen magnum dimensions are used for sex determination. The structural integrity of foramen magnum is usually preserved in re accidents and explosions due to its resistant nature and secluded anatomical position. The data obtained from protrusion of occipital condyles would help in neurosurgical approach of foramen magnum meningiomas. |
/**
* Input stream that reads data from a {@link KvantumOutputStream}.
* Reading from, and writing to the stream is synchronized.
* {@inheritDoc}
*/
@SuppressWarnings("WeakerAccess") public class KvantumInputStream extends InputStream {
private final Object lock = new Object();
private final KvantumOutputStream kvantumOutputStream;
private final int maxSize;
private final byte[] bufferedData;
@Getter private int totalRead = 0;
private int bufferPointer;
private volatile int availableData = 0;
/**
* Construct a new KvantumInputStream, using the file configured buffered
* input size
*
* @param kvantumOutputStream Output stream to read from. Cannot be null.
* @param maxSize Size of the data that is to be read. The stream will never
* read beyond this point. Has to be positive.
*/
public KvantumInputStream(final KvantumOutputStream kvantumOutputStream, final int maxSize) {
this(kvantumOutputStream, CoreConfig.Buffer.in, maxSize);
}
/**
* Construct a new KvantumInputStream
*
* @param kvantumOutputStream Output stream to read from. Cannot be null.
* @param bufferSize Size of the read buffer. Has to be positive.
* @param maxSize Size of the data that is to be read. The stream will never
* read beyond this point. Has to be positive.
*/
public KvantumInputStream(final KvantumOutputStream kvantumOutputStream, final int bufferSize, final int maxSize) {
this.kvantumOutputStream = Assert.notNull(kvantumOutputStream, "output stream");
this.maxSize = Assert.isPositive(maxSize);
this.bufferedData = new byte[Assert.isPositive(bufferSize)];
}
private int readData() {
synchronized (this.lock) {
if (this.kvantumOutputStream.isFinished() || this.kvantumOutputStream.getOffer() == -1) {
return -1;
}
this.availableData = this.kvantumOutputStream.read(bufferedData);
if (availableData == -1) {
return -1;
}
// Reset read state
this.bufferPointer = 0;
return this.availableData;
}
}
@Override public int available() {
return this.maxSize - this.totalRead;
}
@Override public int read() {
synchronized (this.lock) {
// This is here to ensure that the stream NEVER exceeds the upper limit
if (this.totalRead >= this.maxSize) {
return -1;
}
if (this.availableData == -1) {
return -1;
} else if (this.availableData == 0 || this.bufferPointer == this.availableData) {
if (this.readData() < 0) {
return -1;
}
}
final int data = this.bufferedData[bufferPointer++] & 0xFF;
totalRead++;
return data;
}
}
} |
# -*- coding: utf-8 -*-
"""Test helper para datos genericos"""
from und_microservice.helper.string import concat
def test_concat_helper():
"""Prueba de concatenacion de elementos de un array"""
test_data = [
'salu',
'do',
'hola'
]
espected_string = 'salu_do_hola'
result = concat(test_data, '_')
assert espected_string == result
|
Internal Calibration System Using Learning Algorithm With Gradient Descent We present a novel approach to internal calibration of a radar system. A Ku-band radar system with internal calibration paths is designed. Thermal drift of a system is mainly caused by active components, which are a high-power amplifier (HPA) and a low-noise amplifier (LNA). We aimed to reduce the drift using a learning algorithm with a gradient-descent method. Hardware offset factors and calibration factors are introduced for the process. In the learning algorithm, a penalty term is formed based on the analysis of local minimum points. The result verifies the proposed internal calibration method. Maximum deviations of gain are 0.0477 dB for the HPA and 0.0132 dB for the LNA. In addition, the maximum deviations of phase are 0.2481° for HPA and 0.0722° for LNA, respectively. |
<filename>source/cosa/include/ansc_dkuo_interface.h
/*
* If not stated otherwise in this file or this component's Licenses.txt file the
* following copyright and licenses apply:
*
* Copyright 2015 RDK Management
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/**********************************************************************
Copyright [2014] [Cisco Systems, Inc.]
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
**********************************************************************/
/**********************************************************************
module: ansc_dkuo_interface.h
For Advanced Networking Service Container (ANSC),
BroadWay Service Delivery System
---------------------------------------------------------------
description:
This wrapper file defines all the platform-independent
functions and macros for Daemon Socket Udp Object.
---------------------------------------------------------------
environment:
platform independent
---------------------------------------------------------------
author:
<NAME>
---------------------------------------------------------------
revision:
12/10/01 initial revision.
**********************************************************************/
#ifndef _ANSC_DKUO_INTERFACE_
#define _ANSC_DKUO_INTERFACE_
/*
* This object is derived a virtual base object defined by the underlying framework. We include the
* interface header files of the base object here to shield other objects from knowing the derived
* relationship between this object and its base class.
*/
#include "ansc_co_interface.h"
#include "ansc_co_external_api.h"
/***********************************************************
PLATFORM INDEPENDENT DAEMON SOCKET UDP OBJECT DEFINITION
***********************************************************/
/*
* Define some const values that will be used in the object mapper object definition.
*/
#define ANSC_DKUO_MAX_BUFFER_SIZE 2048
/*
* Since we write all kernel modules in C (due to better performance and lack of compiler support),
* we have to simulate the C++ object by encapsulating a set of functions inside a data structure.
*/
typedef ANSC_HANDLE
(*PFN_DKUO_GET_CONTEXT)
(
ANSC_HANDLE hThisObject
);
typedef ANSC_STATUS
(*PFN_DKUO_SET_CONTEXT)
(
ANSC_HANDLE hThisObject,
ANSC_HANDLE hContext
);
typedef ULONG
(*PFN_DKUO_GET_SIZE)
(
ANSC_HANDLE hThisObject
);
typedef ANSC_STATUS
(*PFN_DKUO_SET_SIZE)
(
ANSC_HANDLE hThisObject,
ULONG ulSize
);
typedef PUCHAR
(*PFN_DKUO_GET_ADDRESS)
(
ANSC_HANDLE hThisObject
);
typedef ANSC_STATUS
(*PFN_DKUO_SET_ADDRESS)
(
ANSC_HANDLE hThisObject,
PUCHAR address
);
typedef USHORT
(*PFN_DKUO_GET_PORT)
(
ANSC_HANDLE hThisObject
);
typedef ANSC_STATUS
(*PFN_DKUO_SET_PORT)
(
ANSC_HANDLE hThisObject,
USHORT usPort
);
typedef ANSC_STATUS
(*PFN_DKUO_RETURN)
(
ANSC_HANDLE hThisObject
);
typedef ANSC_STATUS
(*PFN_DKUO_RESET)
(
ANSC_HANDLE hThisObject
);
typedef ANSC_STATUS
(*PFN_DKUO_FINISH)
(
ANSC_HANDLE hThisObject
);
typedef ANSC_STATUS
(*PFN_DKUO_OPEN)
(
ANSC_HANDLE hThisObject
);
typedef ANSC_STATUS
(*PFN_DKUO_CLOSE)
(
ANSC_HANDLE hThisObject
);
typedef ANSC_STATUS
(*PFN_DKUO_ENABLE)
(
ANSC_HANDLE hThisObject,
BOOL bEnable
);
typedef ANSC_STATUS
(*PFN_DKUO_RECV)
(
ANSC_HANDLE hThisObject,
PVOID buffer,
ULONG ulSize
);
typedef ANSC_STATUS
(*PFN_DKUO_SEND)
(
ANSC_HANDLE hThisObject,
PVOID buffer,
ULONG ulSize,
ANSC_HANDLE hReserved
);
/*
* Udp-based Internet Servers have extremely high requirements on performance, processing delay,
* reliability, and scalability. While the base Ansc Socket Object is OK for most Udp-based client
* applications and even some low-end server applications, it's not suitable for high-end Internet
* server applications. The Daemon Udp Object MUST operate in a multi-tasking capable environment.
* It opens a Udp socket and accepts incoming connection requests. Although some functionalities
* it provides are already available in the base socket object, this object is NOT derived from
* the base Ansc Socket Object.
*/
#define ANSC_DAEMON_SOCKET_UDP_CLASS_CONTENT \
/* duplication of the base object class content */ \
ANSCCO_CLASS_CONTENT \
/* start of object class content */ \
ANSC_SOCKET Socket; \
ANSC_IPV4_ADDRESS HostAddress; \
USHORT HostPort; \
ANSC_IPV4_ADDRESS PeerAddress; \
USHORT PeerPort; \
ULONG HashIndex; \
\
ANSC_HANDLE hDaemonServer; \
ANSC_HANDLE hDaemonEngine; \
ANSC_HANDLE hClientContext; \
ANSC_HANDLE hPacket; \
ULONG RecvBytesCount; \
ULONG SendBytesCount; \
ULONG LastRecvAt; \
ULONG LastSendAt; \
BOOL bClosed; \
BOOL bRecvEnabled; \
BOOL bSendEnabled; \
ANSC_LOCK OpLock; \
\
PFN_DKUO_GET_ADDRESS GetPeerAddress; \
PFN_DKUO_SET_ADDRESS SetPeerAddress; \
PFN_DKUO_GET_PORT GetPeerPort; \
PFN_DKUO_SET_PORT SetPeerPort; \
\
PFN_DKUO_GET_CONTEXT GetDaemonServer; \
PFN_DKUO_SET_CONTEXT SetDaemonServer; \
PFN_DKUO_GET_CONTEXT GetDaemonEngine; \
PFN_DKUO_SET_CONTEXT SetDaemonEngine; \
PFN_DKUO_GET_CONTEXT GetClientContext; \
PFN_DKUO_SET_CONTEXT SetClientContext; \
PFN_DKUO_GET_CONTEXT GetPacket; \
PFN_DKUO_SET_CONTEXT SetPacket; \
PFN_DKUO_RETURN Return; \
PFN_DKUO_RESET Reset; \
\
PFN_DKUO_FINISH Finish; \
PFN_DKUO_OPEN Open; \
PFN_DKUO_CLOSE Close; \
PFN_DKUO_ENABLE EnableRecv; \
PFN_DKUO_ENABLE EnableSend; \
\
PFN_DKUO_RECV Recv; \
PFN_DKUO_SEND Send; \
/* end of object class content */ \
typedef struct
_ANSC_DAEMON_SOCKET_UDP_OBJECT
{
ANSC_DAEMON_SOCKET_UDP_CLASS_CONTENT
}
ANSC_DAEMON_SOCKET_UDP_OBJECT, *PANSC_DAEMON_SOCKET_UDP_OBJECT;
#define ACCESS_ANSC_DAEMON_SOCKET_UDP_OBJECT(p) \
ACCESS_CONTAINER(p, ANSC_DAEMON_SOCKET_UDP_OBJECT, Linkage)
#endif
|
package util
import (
"testing"
appsv1 "k8s.io/api/apps/v1"
v1 "k8s.io/apimachinery/pkg/apis/meta/v1"
)
func Test_isStatefulSetReady(t *testing.T) {
tests := []struct {
name string
sts *appsv1.StatefulSet
want bool
}{
{
name: "StatefulSet is ready",
sts: statefulSet(3, 3, 1, 1),
want: true,
},
{
name: "Not all the replicas are updated in StatefulSet ",
sts: statefulSet(2, 3, 1, 1),
want: false,
},
{
name: "Not all the replicas are ready in StatefulSet ",
sts: statefulSet(3, 1, 1, 1),
want: false,
},
{
name: "StatefulSet of the older generation should not be ready",
sts: statefulSet(3, 3, 2, 1),
want: false,
},
}
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
if got := isStatefulSetReady(tt.sts, 3); got != tt.want {
t.Errorf("isStatefulSetReady() = %v, want %v", got, tt.want)
}
})
}
}
func statefulSet(updatedReplicas int32, readyReplicas int32, observedGeneration int64, generation int64) *appsv1.StatefulSet {
return &appsv1.StatefulSet{
Status: appsv1.StatefulSetStatus{
UpdatedReplicas: updatedReplicas,
ReadyReplicas: readyReplicas,
ObservedGeneration: observedGeneration,
},
ObjectMeta: v1.ObjectMeta{
Generation: generation,
},
}
}
|
The House and Senate Armed Services Committees have reached an agreement on the fiscal year 2014 National Defense Authorization Act (NDAA).
As approved by the committees, the text of the latest iteration of the bill is derived from H.R. 1960, which passed the House on June 14 by a vote of 315-108 and S. 1197, a version passed by a Senate committee by a vote of 23-3, later that same day.
House and Senate leaders hurried to hammer out a mutually acceptable measure so as to get the whole package passed before the end of the year.
Reading the mainstream (official) press, one would believe that the NDAA is nothing more nefarious than a necessary replenishing of Pentagon funds. Readers of The New American know, however, there is much more than budget issues contained in the legislation.
For two years, the NDAA included provisions that purported to authorize the president of the United States to deploy the U.S. military to apprehend and indefinitely detain any person (including an American citizen) who he believes “represent[s] an enduring security threat to the United States.”
Such an immense grant of power is not only unconscionable, but unconstitutional, as well.
Regardless of promises to the contrary made every year since 2011 by President Obama, the language of the NDAA places every citizen of the United States within the universe of potential “covered persons.” Any American could one day find himself or herself branded a “belligerent” and thus subject to the complete confiscation of his or her constitutional civil liberties and to nearly never-ending incarceration in a military prison.
Finally, there is in the NDAA for 2014 a frightening fusion of the federal government’s constant surveillance of innocent Americans and the assistance it will give to justifying the indefinite detention of anyone labeled an enemy of the regime.
Section 1071 of the version of the 2014 NDAA approved by the House and Senate committees this week expands on the scope of surveillance established by the Patriot Act and the Authorization for the Use of Military Force (AUMF).
Section 1071(a) authorizes the secretary of defense to "establish a center to be known as the 'Conflict Records Research Center.’” According to the text of the latest version of the NDAA, the center's task would be to compile a “digital research database including translations and to facilitate research and analysis of records captured from countries, organizations, and individuals, now or once hostile to the United States.”
In order to accomplish the center’s purpose, the secretary of defense will create an information exchange in cooperation with the director of national intelligence.
Key to the functioning of this information exchange will be the collection of “captured records.” Section 1071(g)(1), defines a captured record as "a document, audio file, video file, or other material captured during combat operations from countries, organizations, or individuals, now or once hostile to the United States.”
When read in conjunction with the provision of the AUMF that left the War on Terror open-ended and the prior NDAAs’ classification of the United States as a battleground in that unconstitutional war, and you’ve got a powerful combination that can knock out the entire Bill of Rights.
Finally, when all the foregoing is couched within the context of the revelations regarding the dragnet surveillance programs of the NSA, it becomes evident that anyone’s phone records, e-mail messages, browsing history, text messages, and social media posts could qualify as a “captured record.”
After being seized by the NSA (or some other federal surveillance apparatus), the materials would be processed by the Conflict Records Research Center created by this bill. This center's massive database of electronic information and its collaboration with the NSA converts the United States into a constantly monitored holding cell and all its citizens and residents into suspects. All, of course, in the name of the security of the homeland.
Although the outlook is dire, there are those willing to stand and oppose the threats to liberty posed by the NDAA.
For example, libertarian icon and former presidential candidate Ron Paul recently interviewed Daphne Lee, a lady who calls herself “just a mom” but who made an impassioned speech in Nevada against the indefinite detention provisions of the 2012 NDAA. After talking to Lee, Paul announced that he would work to fight enforcement of unconstitutional provisions of the NDAA nationwide.
Additionally, the People Against the NDAA (PANDA) organization is promoting passage of anti-NDAA legislation in towns, counties, and states. On a website devoted to chronicling these efforts, PANDA lists 27 cities, 17 counties, and 25 states that have enacted or are considering bills or resolutions refusing to execute any element of the NDAA that violates the constitutionally protected liberties of its citizens.
While these bills are at various spots along the process of becoming laws, one state recently signed on to thwart the abuse of power authorized by the NDAA.
On October 1, Governor Jerry Brown announced that he had signed AB 351 into law.
The new statute, called the California Liberty Preservation Act, outlaws the participation of any agency of the state of California, any political subdivision of the state, employee of a state or local agency, or member of the California National Guard from
knowingly aiding an agency of the Armed Forces of the United States in any investigation, prosecution, or detention of a person within California pursuant to (1) Sections 1021 and 1022 of the National Defense Authorization Act for Fiscal Year 2012 (NDAA), (2) the federal law known as the Authorization for Use of Military Force, enacted in 2001, or (3) any other federal law, except as specified, if the state agency, political subdivision, employee, or member of the California National Guard would violate the United States Constitution, the California Constitution, or any law of this state by providing that aid....
... knowingly using state funds and funds allocated by the state to those local entities on and after January 1, 2013, to engage in any activity that aids an agency of the Armed Forces of the United States in the detention of any person within California for purposes of implementing Sections 1021 and 1022 of the NDAA or the federal law known as the Authorization for Use of Military Force, if that activity would violate the United States Constitution, the California Constitution, or any law of this state, as specified.
Interpreted broadly, the Liberty Preservation Act would outlaw state cooperation in any federal act which violates the state or federal constitutions. Although Governor Brown almost certainly didn’t intend the provisions of the law to be applied this liberally, the black letter could arguably be used to protect citizens of California from deprivation of a wide panoply of fundamental rights, including the right to keep and bear arms.
It will be worth watching court dockets in California to see if anyone relies on this language to fight the state's infamous disarmament statutes.
Originally sponsored by State Assemblyman Tim Donnelly, a conservative Republican (now running for governor), the bill’s senate sponsor was one of that body’s “most liberal lawmakers,” Mark Leno.
"Indefinite detention, by its very definition, means that we are abrogating, suspending, just throwing away the basic foundations of our Constitution and of our nation," Leno said.
After being warned by some of his fellow Democrats that siding with Donnelly was tantamount to political suicide, Leno stood firm in defense of liberty. "It doesn't matter where one finds oneself on the political spectrum," he said. "These two sections of this national defense act are wrong, unconstitutional and never should have been included.”
Then, in November, a similar bill was introduced to the Ohio State House of Representatives by state Representatives Jim Butler and Ron Young. This concurrent resolution condemns “Section 1021 of the National Defense Authorization Act for Fiscal Year 2012” and urges “the Attorney General of the State of Ohio to bring suit to challenge the constitutionality of Section 1021 of the National Defense Authorization Act for Fiscal Year 2012.”
While neither the California law nor the Ohio resolution is a perfect example of absolute nullification of an unconstitutional federal act, both stand as examples to other state legislatures of attempts to heed the counsel given by James Madison to states that want to resist federal consolidation of all power.
In The Federalist, no. 46, Madison recommended that an effective way to thwart federal overreach is for agents of the states to refuse “to cooperate with officers of the Union.”
In order for Fiscal Year 2014 NDAA to become the “law,” the House of Representatives must pass the bill this week and the Senate would have to follow suit by the end of next week. This gives Americans only a few short days to contact their federal representatives and senators and encourage them to reject any version of the NDAA that infringes on the timeless civil liberties protected by the Constitution.
Joe A. Wolverton, II, J.D. is a correspondent for The New American and travels frequently nationwide speaking on topics of nullification, the NDAA, and the surveillance state. He is the host of The New American Review radio show that is simulcast on YouTube every Monday. Follow him on Twitter @TNAJoeWolverton and he can be reached at This email address is being protected from spambots. You need JavaScript enabled to view it. |
How Should We Organize Care for Patients With Human Immunodeficiency Virus and Comorbidities? A Multisite Qualitative Study of Human Immunodeficiency Virus Care in the United States Department of Veterans Affairs Supplemental Digital Content is available in the text. Background: With human immunodeficiency virus (HIV) now managed as a chronic disease, health care has had to change and expand to include management of other critical comorbidities. We sought to understand how variation in the organization, structure and processes of HIV and comorbidity care, based on patient-centered medical home (PCMH) principles, was related to care quality for Veterans with HIV. Research Design: Qualitative site visits were conducted at a purposive sample of 8 Department of Veterans Affairs Medical Centers, varying in care quality and outcomes for HIV and common comorbidities. Site visits entailed conduct of patient interviews (n=60); HIV care team interviews (n=60); direct observation of clinic processes and team interactions (n=22); and direct observations of patient-provider clinical encounters (n=45). Data were analyzed using a priori and emergent codes, construction of site syntheses and comparing sites with varying levels of quality. Results: Sites highest and lowest in both HIV and comorbidity care quality demonstrated clear differences in provision of PCMH-principled care. The highest site provided greater team-based, comprehensive, patient-centered, and data-driven care and engaged in continuous improvement. Sites with higher HIV care quality attended more to psychosocial needs. Sites that had consistent processes for comorbidity care, whether in HIV or primary care clinics, had higher quality of comorbidity care. Conclusions: Provision of high-quality HIV care and high-quality co-morbidity care require different care structures and processes. Provision of both requires a focus on providing care aligned with PCMH principles, integrating psychosocial needs into care, and establishing explicit consistent approaches to comorbidity management. |
Sometimes, psychological research acts as an enlightening agent, delving into the depths of the mind and illuminating obscure elements of our unconscious. Other times… well, its discoveries reveal information we already knew – like the fact that cannabis consumers, in some ways, are just different from other people.
As it turns out, the average cannabis user is significantly higher in the trait of “Openness to Experience” than their cannabis-free counterparts. Openness is a trait in the Big Five personality test found to have an enormous association with many facets of life, from IQ to political preference.
Make sure to check out the quiz below to find out how open you are!
The Big Five and Openness
The Big Five personality test (also called the “Five Factor Model”) is divided into five traits:
Openness
Conscientiousness
Extraversion
Agreeableness
Neuroticism
Each trait is independently measured and represents a different aspect of the personality. This means that people don’t just fall into one of five “types” – instead, we look at every trait individually to get a broad overview of the person’s temperament.
Openness is a reflection of a person’s curiosity, ability to learn, and artistic interest. People high in Openness tend to be more engaged with the world around them, more appreciative of natural and artistic beauty, and more prone to imagination and fantasy.
Knowing what we know about the trait, it doesn’t come as a surprise that cannabis consumers tend to score significantly higher in “Openness” than the general population. This finding isn’t just derived from a single study either – it’s a well-documented correlation dating back to the 70’s. In fact, the original description of “Openness to Experience” was created to describe the personality of cannabis consumers.
Open people tend to be more creative, experience more career success in both the arts and sciences, and have higher IQs compared to those with lower levels of Openness. There is also a well-established link between Openness and political progressivism.
Being open has its downsides, too. People high in this trait should be careful not to let curiosity run their lives. If they also happen to be low in Conscientiousness (characterized by hard work and organization), they can often find themselves too interested in the world around them to be able to focus on a specific task for a long period of time.
Cannabis, Curiosity and Creativity
Countless creatives and intellectuals use cannabis – in fact, we know that some of the greatest thinkers ever to have lived were avid partakers in the plant: The Beatles, Carl Sagan, and Hunter S. Thompson, just to name a few. So what exactly is it about cannabis that attracts more of this specific personality type?
One explanation lies in the hypothesis that Openness (along with Extraversion) is an exploratory mechanism. Open people are naturally curious, so they probably have a heightened incentive to try new things.
We’re still unsure whether cannabis has an effect on Openness after being consumed. Famed scientist Carl Sagan even once stated that “The cannabis experience has greatly improved my appreciation for art, a subject which I had never much appreciated before.” This could be an indication that using the plant helped to increase his level of Openness.
Sagan’s claim still needs to be taken with a grain of salt – as far as we know, there has been no scientific study designed to measure levels of Openness before and after cannabis use. If cannabis does have an impact on Openness though, it wouldn’t be the first substance found to have this effect. An experiment at Johns Hopkins University found that people who took psilocybin had significantly higher levels of Openness more than a year later. This was quite an impressive phenomenon, considering the fact that personality traits tend to remain fairly constant throughout life.
Though a lot of research still needs to be done to better understand the correlation between cannabis use and Openness, it’s clear that this information carries with it some pretty important implications.
Curious about your level of Openness? Take the test below and post your results in the comments!
[os-widget path=”/leafly1/how-open-are-you”] |
Photobleaching of endogenous fluorochroms in tissues in vivo during laser irradiation The photobleaching phenomenon has been previously first of all widely studied for exogenous photosensitizers applied in photodynamic therapy (PDT). The present paper deals with detailed investigation of photobleaching of endogenous fluorochroms in tissues in vivo during its laser irradiation at 532, 633 and 670 nm at different powers. The fluorescence decay curves during skin irradiation in vivo at these wavelengths have been obtained and analyzed. The similarity in bleaching behavior of endogenous fluorochroms and exogenous photosensitizers used in PDT gave us the reasons to imply the possible connection between native fluorochroms and low intensity laser therapy effects. The results obtained may be applied for tissue diagnostics and therapy control. |
As Bart and Lisa scramble for the school bus to Springfield Elementary on “The Simpsons” this fall, millions of American kids are repeating the same ritual. More than 25 million children ride school buses each day in the United States, and virtually all of them have a safe ride. In fact, of all the ways your child can get to school, none is safer than the school bus.
But that doesn’t mean parents should be complacent. The most risky part of the school bus ride isn’t the ride itself – it’s getting on and off the bus, according to the National Highway Traffic Safety Administration (NHTSA). Drivers in a hurry who speed around parked school buses are a serious danger to schoolchildren at bus stops, who often dart or step into traffic unexpectedly. But recent studies and reports show that many drivers are not only ignoring warning lights, they’re flaunting the law by texting on their cell as they roll through school zones.
Most state laws require drivers to stop when a school bus displays its flashing red lights, but these laws vary and are routinely violated. A 2014 survey of more than 97,000 school bus drivers found more than 75,000 drivers illegally zoomed past school buses dropping off or picking up passengers in a single day.
Distracted driving is another enormous problem in school zones, according to auto insurance carriers and consumer groups. In an in-depth study on distracted driving in school zones by SafeKids.org, researchers found that of 41,426 drivers observed traveling through a school zone on a given day, one in six was distracted. Slightly more women were distracted than men (187 women out of every thousand, compared to 154 men). Women were also more likely to be distracted by cell phones and electronics and grooming, while men and women were equally distracted while driving by eating, reaching behind the seat and reading.
Ironically, schools that required a lower driving speed in the school zone were more likely to be plagued by distracted drivers than schools that didn’t change the speed limit, the study showed.
A few other intriguing factoids from the school zone study: Drivers of big vehicles, such as SUVs, pickup trucks and minivans, were more distracted than people driving cars. Regardless of gender, drivers who weren't wearing seat belts were 35% more likely to be distracted. And drivers in states that restricted the use of cell phones and other handheld devices in drivers of all ages were 13% less likely to be distracted than those in other states.
All this contributes to schoolchildren being hurt or even killed. Fewer than 0.05 deaths since 1998 are related to school transportation, according to federal safety agencies, but this still means that an average of 19 school children die needlessly each year. Of those children, fourteen are pedestrians, nearly half of whom are between 5 and 7 years old. And 24% are killed by cars or trucks rather than school buses.
Don't rush. Plan things so you get to the bus stop five minutes early: That way you don’t have to hurry. Arrange to walk to and from the bus stop with a friend or someone in your family. And never run for the bus.
No roughhousing. Stand at least six feet from the approaching school bus while you’re waiting at the bus stop. Insurers also warn kids not to push, wrestle or rough-house in line – someone could slip and fall in the street.
Wait to move toward the bus until it stops. Never walk toward the bus until the bus driver has stopped it completely, opened the door and turned on the flashing safety lights.
Be aware of the school bus “danger zone.” Be especially careful in the area 10 to 12 feet encircling the bus. It’s dangerous because the bus driver – and other drivers – may find it hard to see your child there.
Treat the bus ride like any other ride. “It’s like riding the tram at an amusement park,” Hanshew says. “Stay in your seat until the school bus comes to a complete stop.” According to HealthyKids.org, most kids are seriously injured or worse when they’re in a hurry to get on or off the bus and don’t pay attention to surrounding traffic.
Stay in your seat. Find a seat and stay put. If the bus has seat belts, buckle up. Do what the driver tells you to do, and talk quietly so he or she can concentrate. No matter how tempting, don’t stick anything out of a bus window.
When you exit, keep your eyes on the bus and oncoming cars. When getting off the bus, cross the street in front of the bus and make sure the bus driver sees you. Always watch for surrounding traffic, and never cross a street without checking both ways for traffic (tell your child to look left, then right, then left again). If you have a cell phone, don’t make calls, listen to music or text when you’re crossing the street. And – very important -- if you drop something near the bus, always notify the driver before you try to pick it up.
--Mary Purcell contributed to this report. |
Optimized Outcomes Using a Standardized Approach for the Treatment of Patients with Carcinoid Heart Disease Background: Carcinoid heart disease (CHD) is common in patients with carcinoid syndrome (CS). Surgical treatment improves the poor prognosis of CHD, although the reported peri-operative mortality is high (∼17%). We attempted to improve outcomes by implementation of a protocol for the management of patients with CHD at a UK Neuroendocrine Centre of Excellence and report our experience. Methods: All patients treated for CHD between 2008 and 2015 were included. Peri-operative treatment included surgical features such as invasive pulmonary valve (PV) inspection and preservation of the tricuspid subvalvular apparatus. Results: A total of 11 patients were treated; the median age was 63 years (IQR: 56-70). Ten patients underwent both pulmonary valve replacement (PVR) and tricuspid valve replacement (TVR); 1 patient underwent isolated TVR. One patient had additional aortic valve replacement (AVR), another one coronary artery bypass grafting. Bioprostheses (BP) were used in all patients, stented for TVR and AVR, stentless for PVR. Invasive PV inspection caused unplanned PVR in 3 cases (27.3%). All patients were discharged home. One patient (9.1%), who had had previous TVR by another surgeon, had right heart failure (RHF) during follow-up. One death occurred due to progression of CS (day 346). The carcinoids' primary was resected in 5 patients (45.5%) 10 months (4.5-19.5) after cardiac surgery. Conclusion: Excellent results were achieved in patients with CHD. PV stenosis can be underestimated by echocardiography; therefore, intraoperative inspection is recommended. Right ventricular geometry should be respected to prevent RHF. BP should be used, as these patients are likely to undergo future non-cardiac surgeries. |
Online monitoring and diagnosis of batch processes: empirical model-based framework and a case study An empirical model-based framework for monitoring and diagnosing batch processes is proposed. With the input of past successful and unsuccessful batches, the off-line portion of the framework constructs empirical models. Using online process data of a new batch, the online portion of the framework makes monitoring and diagnostic decisions in a real-time basis. The proposed framework consists of three phases: monitoring, diagnostic screening, and diagnosis. For monitoring and diagnosis purposes, the multiway principal-component analysis (MPCA) model and discriminant model are adopted as reference models. As an intermediate step, the diagnostic screening phase narrows down the possible cause candidates of the fault in question. By analysing the MPCA monitoring model, the diagnostic screening phase constructs a variable influence model to screen out unlikely cause candidates. The performance of the proposed framework is tested using a real dataset from a PVC batch process. It has been shown that the proposed framework produces reliable diagnosis results. Moreover, the inclusion of the diagnostic screening phase as a pre-diagnostic step has improved the diagnosis performance of the proposed framework, especially in the early time intervals. |
Depression in Older People: What does the Future Hold? Demographic, economic and social changes over the next few decades are likely to have a significant effect on the care of the elderly. This personal view, from a United Kingdom perspective, examines some of these changes and assesses the impact they may have on the prevalence and treatment of depression in older people. Strategies for minimizing the marginalization of this group are discussed. |
<filename>src/main/java/com/jalasoft/minions/queue/Queue.java
package com.jalasoft.minions.queue;
import com.jalasoft.minions.linkedlist.LinkedList;
/**
* Created by DCABEROB on 5/22/2019.
*/
public class Queue<T> {
LinkedList<T> queue;
public Queue(){
queue = new LinkedList<>();
}
/**
* Method to add in the queue
* @param value - value.
*/
public void add(T value) {
queue.add(value);
}
/**
* Method to remove the first in the queue
*/
public void remove() {
queue.removeFirst();
}
public boolean isEmpty() {
return queue.isEmpty();
}
public T getIndex(int index){
return queue.getIndex();
}
}
|
/* antimicro Gamepad to KB+M event mapper
* Copyright (C) 2015 <NAME> <<EMAIL>>
*
* This program 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.
* This program 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 this program. If not, see <http://www.gnu.org/licenses/>.
*/
#include <QHash>
#include <QHashIterator>
#include "quicksetdialog.h"
#include "ui_quicksetdialog.h"
#include "setjoystick.h"
#include "buttoneditdialog.h"
QuickSetDialog::QuickSetDialog(InputDevice *joystick, QWidget *parent) :
QDialog(parent),
ui(new Ui::QuickSetDialog)
{
ui->setupUi(this);
setAttribute(Qt::WA_DeleteOnClose);
this->joystick = joystick;
this->currentButtonDialog = 0;
setWindowTitle(tr("Quick Set %1").arg(joystick->getName()));
SetJoystick *currentset = joystick->getActiveSetJoystick();
currentset->release();
joystick->resetButtonDownCount();
QString temp = ui->joystickDialogLabel->text();
temp = temp.arg(joystick->getSDLName()).arg(joystick->getName());
ui->joystickDialogLabel->setText(temp);
for (int i=0; i < currentset->getNumberSticks(); i++)
{
JoyControlStick *stick = currentset->getJoyStick(i);
QHash<JoyControlStick::JoyStickDirections, JoyControlStickButton*> *stickButtons = stick->getButtons();
QHashIterator<JoyControlStick::JoyStickDirections, JoyControlStickButton*> iter(*stickButtons);
while (iter.hasNext())
{
JoyControlStickButton *stickbutton = iter.next().value();
if (stick->getJoyMode() != JoyControlStick::EightWayMode)
{
if (stickbutton->getJoyNumber() != JoyControlStick::StickLeftUp &&
stickbutton->getJoyNumber() != JoyControlStick::StickRightUp &&
stickbutton->getJoyNumber() != JoyControlStick::StickLeftDown &&
stickbutton->getJoyNumber() != JoyControlStick::StickRightDown)
{
connect(stickbutton, SIGNAL(clicked(int)), this, SLOT(showStickButtonDialog()));
}
}
else
{
connect(stickbutton, SIGNAL(clicked(int)), this, SLOT(showStickButtonDialog()));
}
if (!stickbutton->getIgnoreEventState())
{
stickbutton->setIgnoreEventState(true);
}
}
}
for (int i=0; i < currentset->getNumberAxes(); i++)
{
JoyAxis *axis = currentset->getJoyAxis(i);
if (!axis->isPartControlStick() && axis->hasControlOfButtons())
{
JoyAxisButton *naxisbutton = axis->getNAxisButton();
JoyAxisButton *paxisbutton = axis->getPAxisButton();
connect(naxisbutton, SIGNAL(clicked(int)), this, SLOT(showAxisButtonDialog()));
connect(paxisbutton, SIGNAL(clicked(int)), this, SLOT(showAxisButtonDialog()));
if (!naxisbutton->getIgnoreEventState())
{
naxisbutton->setIgnoreEventState(true);
}
if (!paxisbutton->getIgnoreEventState())
{
paxisbutton->setIgnoreEventState(true);
}
}
}
for (int i=0; i < currentset->getNumberHats(); i++)
{
JoyDPad *dpad = currentset->getJoyDPad(i);
QHash<int, JoyDPadButton*>* dpadbuttons = dpad->getButtons();
QHashIterator<int, JoyDPadButton*> iter(*dpadbuttons);
while (iter.hasNext())
{
JoyDPadButton *dpadbutton = iter.next().value();
if (dpad->getJoyMode() != JoyDPad::EightWayMode)
{
if (dpadbutton->getJoyNumber() != JoyDPadButton::DpadLeftUp &&
dpadbutton->getJoyNumber() != JoyDPadButton::DpadRightUp &&
dpadbutton->getJoyNumber() != JoyDPadButton::DpadLeftDown &&
dpadbutton->getJoyNumber() != JoyDPadButton::DpadRightDown)
{
connect(dpadbutton, SIGNAL(clicked(int)), this, SLOT(showDPadButtonDialog()));
}
}
else
{
connect(dpadbutton, SIGNAL(clicked(int)), this, SLOT(showDPadButtonDialog()));
}
if (!dpadbutton->getIgnoreEventState())
{
dpadbutton->setIgnoreEventState(true);
}
}
}
for (int i=0; i < currentset->getNumberVDPads(); i++)
{
VDPad *dpad = currentset->getVDPad(i);
if (dpad)
{
QHash<int, JoyDPadButton*>* dpadbuttons = dpad->getButtons();
QHashIterator<int, JoyDPadButton*> iter(*dpadbuttons);
while (iter.hasNext())
{
JoyDPadButton *dpadbutton = iter.next().value();
if (dpad->getJoyMode() != JoyDPad::EightWayMode)
{
if (dpadbutton->getJoyNumber() != JoyDPadButton::DpadLeftUp &&
dpadbutton->getJoyNumber() != JoyDPadButton::DpadRightUp &&
dpadbutton->getJoyNumber() != JoyDPadButton::DpadLeftDown &&
dpadbutton->getJoyNumber() != JoyDPadButton::DpadRightDown)
{
connect(dpadbutton, SIGNAL(clicked(int)), this, SLOT(showDPadButtonDialog()));
}
}
else
{
connect(dpadbutton, SIGNAL(clicked(int)), this, SLOT(showDPadButtonDialog()));
}
if (!dpadbutton->getIgnoreEventState())
{
dpadbutton->setIgnoreEventState(true);
}
}
}
}
for (int i=0; i < currentset->getNumberButtons(); i++)
{
JoyButton *button = currentset->getJoyButton(i);
if (button && !button->isPartVDPad())
{
connect(button, SIGNAL(clicked(int)), this, SLOT(showButtonDialog()));
if (!button->getIgnoreEventState())
{
button->setIgnoreEventState(true);
}
}
}
connect(this, SIGNAL(finished(int)), this, SLOT(restoreButtonStates()));
}
QuickSetDialog::~QuickSetDialog()
{
delete ui;
}
void QuickSetDialog::showAxisButtonDialog()
{
if (!currentButtonDialog)
{
JoyAxisButton *axisbutton = static_cast<JoyAxisButton*>(sender());
currentButtonDialog = new ButtonEditDialog(axisbutton, this);
currentButtonDialog->show();
connect(currentButtonDialog, SIGNAL(finished(int)), this, SLOT(nullifyDialogPointer()));
}
}
void QuickSetDialog::showButtonDialog()
{
if (!currentButtonDialog)
{
JoyButton *button = static_cast<JoyButton*>(sender());
currentButtonDialog = new ButtonEditDialog(button, this);
currentButtonDialog->show();
connect(currentButtonDialog, SIGNAL(finished(int)), this, SLOT(nullifyDialogPointer()));
}
}
void QuickSetDialog::showStickButtonDialog()
{
if (!currentButtonDialog)
{
JoyControlStickButton *stickbutton = static_cast<JoyControlStickButton*>(sender());
currentButtonDialog = new ButtonEditDialog(stickbutton, this);
currentButtonDialog->show();
connect(currentButtonDialog, SIGNAL(finished(int)), this, SLOT(nullifyDialogPointer()));
}
}
void QuickSetDialog::showDPadButtonDialog()
{
if (!currentButtonDialog)
{
JoyDPadButton *dpadbutton = static_cast<JoyDPadButton*>(sender());
currentButtonDialog = new ButtonEditDialog(dpadbutton, this);
currentButtonDialog->show();
connect(currentButtonDialog, SIGNAL(finished(int)), this, SLOT(nullifyDialogPointer()));
}
}
void QuickSetDialog::nullifyDialogPointer()
{
if (currentButtonDialog)
{
currentButtonDialog = 0;
emit buttonDialogClosed();
}
}
void QuickSetDialog::restoreButtonStates()
{
SetJoystick *currentset = joystick->getActiveSetJoystick();
for (int i=0; i < currentset->getNumberSticks(); i++)
{
JoyControlStick *stick = currentset->getJoyStick(i);
QHash<JoyControlStick::JoyStickDirections, JoyControlStickButton*> *stickButtons = stick->getButtons();
QHashIterator<JoyControlStick::JoyStickDirections, JoyControlStickButton*> iter(*stickButtons);
while (iter.hasNext())
{
JoyControlStickButton *stickbutton = iter.next().value();
if (stickbutton->getIgnoreEventState())
{
stickbutton->setIgnoreEventState(false);
}
disconnect(stickbutton, SIGNAL(clicked(int)), this, 0);
}
}
for (int i=0; i < currentset->getNumberAxes(); i++)
{
JoyAxis *axis = currentset->getJoyAxis(i);
if (!axis->isPartControlStick() && axis->hasControlOfButtons())
{
JoyAxisButton *naxisbutton = axis->getNAxisButton();
if (naxisbutton->getIgnoreEventState())
{
naxisbutton->setIgnoreEventState(false);
}
JoyAxisButton *paxisbutton = axis->getPAxisButton();
if (paxisbutton->getIgnoreEventState())
{
paxisbutton->setIgnoreEventState(false);
}
disconnect(naxisbutton, SIGNAL(clicked(int)), this, 0);
disconnect(paxisbutton, SIGNAL(clicked(int)), this, 0);
}
}
for (int i=0; i < currentset->getNumberHats(); i++)
{
JoyDPad *dpad = currentset->getJoyDPad(i);
QHash<int, JoyDPadButton*>* dpadbuttons = dpad->getButtons();
QHashIterator<int, JoyDPadButton*> iter(*dpadbuttons);
while (iter.hasNext())
{
JoyDPadButton *dpadbutton = iter.next().value();
if (dpadbutton->getIgnoreEventState())
{
dpadbutton->setIgnoreEventState(false);
}
disconnect(dpadbutton, SIGNAL(clicked(int)), this, 0);
}
}
for (int i=0; i < currentset->getNumberVDPads(); i++)
{
VDPad *dpad = currentset->getVDPad(i);
if (dpad)
{
QHash<int, JoyDPadButton*>* dpadbuttons = dpad->getButtons();
QHashIterator<int, JoyDPadButton*> iter(*dpadbuttons);
while (iter.hasNext())
{
JoyDPadButton *dpadbutton = iter.next().value();
if (dpadbutton->getIgnoreEventState())
{
dpadbutton->setIgnoreEventState(false);
}
disconnect(dpadbutton, SIGNAL(clicked(int)), this, 0);
}
}
}
for (int i=0; i < currentset->getNumberButtons(); i++)
{
JoyButton *button = currentset->getJoyButton(i);
if (button && !button->isPartVDPad())
{
if (button->getIgnoreEventState())
{
button->setIgnoreEventState(false);
}
disconnect(button, SIGNAL(clicked(int)), this, 0);
}
}
currentset->release();
}
|
Purified Tea (Camellia sinensis (L.) Kuntze) Flower Saponins Induce the p53-Dependent Intrinsic Apoptosis of Cisplatin-Resistant Ovarian Cancer Cells Ovarian cancer is currently ranked at fifth in cancer deaths among women. Patients who have undergone cisplatin-based chemotherapy can experience adverse effects or become resistant to treatment, which is a major impediment for ovarian cancer treatment. Natural products from plants have drawn great attention in the fight against cancer recently. In this trial, purified tea (Camellia sinensis (L.) Kuntze) flower saponins (PTFSs), whose main components are Chakasaponin I and Chakasaponin IV, inhibited the growth and proliferation of ovarian cancer cell lines A2780/CP70 and OVCAR-3. Flow cytometry, caspase activity and Western blotting analysis suggested that such inhibitory effects of PTFSs on ovarian cancer cells were attributed to the induction of cell apoptosis through the intrinsic pathway rather than extrinsic pathway. The p53 protein was then confirmed to play an important role in PTFS-induced intrinsic apoptosis, and the levels of its downstream proteins such as caspase families, Bcl-2 families, Apaf-1 and PARP were regulated by PTFS treatment. In addition, the upregulation of p53 expression by PTFSs were at least partly induced by DNA damage through the ATM/Chk2 pathway. The results help us to understand the mechanisms underlying the effects of PTFSs on preventing and treating platinum-resistant ovarian cancer. Introduction Ovarian cancer is currently ranked at fifth in cancer deaths that affect females. According to the American Cancer Society estimates, about 13,940 ovarian cancer patients will die among 21,750 new women patients in the United States in 2020. The chemotherapy resistance and the inability to detect ovarian cancer in its earliest stages are the biggest problems that doctors face today. Cisplatin is widely used as a chemotherapy medicine to treat ovarian cancer. However, it might cause tumor recurrence and chemotherapy resistance in patients, which are the main reasons for treatment failure in most patients with metastatic disease. In ovarian cancer, key apoptotic regulators, such as the Akt family, p53, and death-receptor family, mediate the cell response to cisplatin, which plays a vital role in the induction of resistance of cancer cells to chemotherapeutic agents. It was reported that the over-expression of anti-apoptotic Bcl-2 family members, including Bcl-2 and Bcl-xL or down-regulation or mutation of pro-apoptotic factors, such as Bax and caspases, are (UPLC-Q-TOF/MS/MS) analysis. Three main bioactive compounds were found to be present (Figure 1a). The molecular formulas of these saponins were identified by high-resolution mass measurements, deprotonated molecular ions −, and fragment ions ( Table 1). The three saponins were Chakasaponin IV, Chakasaponin I and Floratheasaponin A (Figure 1b) (allocating 8.1%, 80.1%, and 1.4%, respectively), based on published fragmentation data and nominal mass calculated from known structures. PTFSs Inhibit Cell Viability and Cell Proliferation in A2780/CP70s and OVCAR-3s The cytotoxicity of PTFSs on A2780/CP70s, OVCAR-3s and IOSE -364s was analyzed using the MTS assay. Cells were treated with various concentrations (0-3.5 g/mL) of PTFSs for 24 h. As shown in Figure 2a, PTFSs inhibited the growth of all three cell lines in a dose-dependent manner, but had a lower cytotoxic effect against normal epithelial ovarian cell line IOSE-364 cells than that against ovarian cancer cell lines A2780/CP70 and OVCAR-3. The cell viability decreased from 100% to 20.28% for A2780/CP70, from 100% to 6.52% for OVCAR-3, and from 100% to 53.64% for IOSE-364 cells after being treated with PTFSs (0-3.5 g/mL) for 24 h. The IC50 of PTFSs was 2.71 g/mL for A2780/CP70, 2.58 g/mL for OVCAR-3, and 3.90 g/mL for IOSE-364. To verify whether PTFSs inhibit cell proliferation, we also checked the protein expression of proliferating cell nuclear antigen (PCNA) in both ovarian cancer cell lines by Western blotting. We found that PTFSs significantly decreased its PTFSs Inhibit Cell Viability and Cell Proliferation in A2780/CP70s and OVCAR-3s The cytotoxicity of PTFSs on A2780/CP70s, OVCAR-3s and IOSE -364s was analyzed using the MTS assay. Cells were treated with various concentrations (0-3.5 g/mL) of PTFSs for 24 h. As shown in Figure 2a, PTFSs inhibited the growth of all three cell lines in a dose-dependent manner, but had a lower cytotoxic effect against normal epithelial ovarian cell line IOSE-364 cells than that against ovarian cancer cell lines A2780/CP70 and OVCAR-3. The cell viability decreased from 100% to 20.28% for A2780/CP70, from 100% to 6.52% for OVCAR-3, and from 100% to 53.64% for IOSE-364 cells after being treated with PTFSs (0-3.5 g/mL) for 24 h. The IC 50 of PTFSs was 2.71 g/mL for A2780/CP70, 2.58 g/mL for OVCAR-3, and 3.90 g/mL for IOSE-364. To verify whether PTFSs inhibit cell proliferation, we also checked the protein expression of proliferating cell nuclear antigen (PCNA) in both ovarian cancer cell expression in the treatment group compared with control group (Figure 2b,c). These results confirm that PTFSs inhibit cell viability and cell proliferation in ovarian cancer cells. Effects of PTFSs (0, 1, 2, 3 g/mL) on the expression of proliferating cell nuclear antigen (PCNA) were determined by Western blotting. GAPDH was utilized for an endogenous reference to standardize protein levels. All values were expressed as mean ± standard error of mean (SEM) of three independent experiments. * p < 0.05 versus control. PTFSs Induce Apoptosis in A2780/CP70s and OVCAR-3s To study if PTFSs inhibit cell viability and proliferation through inducing apoptosis in ovarian cancer cells, the changes in nuclear morphology of A2780/CP70s and OVCAR-3s treated with PTFSs (0,1,2,3 g/mL) for 24 h were analyzed by the Hoechst 33342 DNA staining and observed under a fluorescence microscope. As shown in Figure 3a, 24h PTFS treatment could cause obvious chromatin condensation and nucleus shrinkage in A2780/CP70s and OVCAR-3s. Such results are further verified by the results of flow cytometry assay. The results show that PTFSs significantly decreased the live cell percentage and increased the apoptotic cell one in both ovarian cancer cell lines in a concentration-dependent manner (Figure 3b). The total apoptotic cells increased from 19.53% to 48.34% among A2780/CP70 cells and increased from 18.53% to 66.03% among OVCAR-3 cells. In addition, the JC-1 aggregates/JC-1 monomers ratio in A2780/CP70 cells was quantitated by fluorescence spectrophotometer. The results show that the mitochondrial membrane potential collapsed after being treated with 3 and 4 g/mL PTFSs for 24 h (Figure 3c), which suggested the early appearance of apoptosis. We further checked the protein levels of cleaved PARP and cytochrome c, which served as a marker and executor during cell apoptosis, respectively. The results show that PTFSs increased the protein levels of cleaved PARP and cytochrome c in A2780/CP70s and OVCAR-3s, and the protein level of full length PARP were decreased (Figure 3d, e, f). All above data suggest that PTFS treatment induced mitochondrial membrane potential collapsing and apoptosis in ovarian cancer cells, and the inhibition of PTFSs on ovarian cancer cell viability might be due to inducing apoptosis. Effects of PTFSs (0, 1, 2, 3 g/mL) on the expression of proliferating cell nuclear antigen (PCNA) were determined by Western blotting. GAPDH was utilized for an endogenous reference to standardize protein levels. All values were expressed as mean ± standard error of mean (SEM) of three independent experiments. * p < 0.05 versus control. PTFSs Induce Apoptosis in A2780/CP70s and OVCAR-3s To study if PTFSs inhibit cell viability and proliferation through inducing apoptosis in ovarian cancer cells, the changes in nuclear morphology of A2780/CP70s and OVCAR-3s treated with PTFSs (0, 1, 2, 3 g/mL) for 24 h were analyzed by the Hoechst 33342 DNA staining and observed under a fluorescence microscope. As shown in Figure 3a, 24 h PTFS treatment could cause obvious chromatin condensation and nucleus shrinkage in A2780/CP70s and OVCAR-3s. Such results are further verified by the results of flow cytometry assay. The results show that PTFSs significantly decreased the live cell percentage and increased the apoptotic cell one in both ovarian cancer cell lines in a concentration-dependent manner (Figure 3b). The total apoptotic cells increased from 19.53% to 48.34% among A2780/CP70 cells and increased from 18.53% to 66.03% among OVCAR-3 cells. In addition, the JC-1 aggregates/JC-1 monomers ratio in A2780/CP70 cells was quantitated by fluorescence spectrophotometer. The results show that the mitochondrial membrane potential collapsed after being treated with 3 and 4 g/mL PTFSs for 24 h (Figure 3c), which suggested the early appearance of apoptosis. We further checked the protein levels of cleaved PARP and cytochrome c, which served as a marker and executor during cell apoptosis, respectively. The results show that PTFSs increased the protein levels of cleaved PARP and cytochrome c in A2780/CP70s and OVCAR-3s, and the protein level of full length PARP were decreased (Figure 3d-f). All above data suggest that PTFS treatment induced mitochondrial membrane potential collapsing and apoptosis in ovarian cancer cells, and the inhibition of PTFSs on ovarian cancer cell viability might be due to inducing apoptosis. (d-f) Protein levels of full length PARP, cleaved PARP, and cytochrome c were detected by Western blotting. GAPDH was utilized for an endogenous reference to standardize protein levels. All data were expressed as mean ± SEM of three independent experiments. * p < 0.05 compared with the control group. Above four treatments concentration of PTFSs were 0, 1, 2, and 3 g/mL respectively. PTFSs Induce Apoptosis via Caspase-3/7 and -9 Activation in A2780/CP70s Caspases, as a family of cysteine proteases, play an important role in apoptotic responses. To demonstrate whether the PTFS-induced apoptosis in ovarian cancer cells was associated to the activation of caspases, we examined the activities of caspase-3/7, -8 and -9 in A2780/CP70 cells after being treated with PTFSs for 24 h, as shown in Figure 4a. Treatment with PTFSs enhanced the caspase-3/7 and -9 activities by 1.97-and 1.42-fold compared to that in controls, respectively, while it did not affect the activity of caspase-8. In addition, the protein expression of caspase-3, -7, -8 and -9 were also examined by Western blotting (Figure 4b,c). The results show that the protein expressions of cleaved caspase-3, -7 and -9, and procaspase-9 were significantly increased; contrarily, that of procaspase-3 was decreased, while the protein expressions of procaspase-7, -8 and cleaved caspase-8 were not affected after treated with PTFSs. All the above data suggest that PTFS-induced apoptosis is related to the activation of caspase-3, -7 and -9, but not related to the activation of caspase-8. Int. J. Mol. Sci. 2020, 21, x FOR PEER REVIEW 6 of 20 a double-staining method with FITC-conjugated Annexin V and PI. (c) Quantification histograms represent the ratio of JC-1 aggregates to JC-1 monomers (ratio of 590:530 nm emission intensity) of A2780/CP70 cells, which reveals m dissipation after 24 h treatment with various concentrations PTFSs. (d, e, f) Protein levels of full length PARP, cleaved PARP, and cytochrome c were detected by Western blotting. GAPDH was utilized for an endogenous reference to standardize protein levels. All data were expressed as mean ± SEM of three independent experiments. * p < 0.05 compared with the control group. Above four treatments concentration of PTFSs were 0, 1, 2, and 3 g/mL respectively. PTFSs Induce Apoptosis via Caspase-3/7 and -9 Activation in A2780/CP70s Caspases, as a family of cysteine proteases, play an important role in apoptotic responses. To demonstrate whether the PTFS-induced apoptosis in ovarian cancer cells was associated to the activation of caspases, we examined the activities of caspase-3/7, -8 and -9 in A2780/CP70 cells after being treated with PTFSs for 24 h, as shown in Figure 4a. Treatment with PTFSs enhanced the caspase-3/7 and -9 activities by 1.97-and 1.42-fold compared to that in controls, respectively, while it did not affect the activity of caspase-8. In addition, the protein expression of caspase-3, -7, -8 and -9 were also examined by Western blotting (Figure 4b,c). The results show that the protein expressions of cleaved caspase-3, -7 and -9, and procaspase-9 were significantly increased; contrarily, that of procaspase-3 was decreased, while the protein expressions of procaspase-7, -8 and cleaved caspase-8 were not affected after treated with PTFSs. All the above data suggest that PTFS-induced apoptosis is related to the activation of caspase-3, -7 and -9, but not related to the activation of caspase-8. Effects of PTFSs (0, 1, 2, 3 g/mL) on the expression of caspase-3, -7, -8 and -9 were assayed by Western blotting. GAPDH was utilized for an endogenous reference to standardize protein levels. All values are shown as mean ± SEM of three independent experiments. * p < 0.05 versus control. Cells treated with culture medium containing 0.01% dimethyl sulfoxide (DMSO) was used as the control. PTFSs Induce Apoptosis through the Intrinsic Rather than Extrinsic Apoptotic Pathway in A2780/CP70s As caspase-8 and -9 are respectively the initiators of the extrinsic and intrinsic apoptotic pathway, we further investigated whether the intrinsic or extrinsic pathway participated in the PTFS-induced apoptosis of A2780/CP70s. We first checked the expressions of pro-apoptotic proteins Bax and Bad, and anti-apoptotic proteins Bcl-2 and Bcl-xL, which belong to the Bcl-2 families, related to the intrinsic apoptotic pathway. (Figure 5a,b). The results show PTFSs significantly upregulated Bax and Bad proteins, and downregulated Bcl-xL and Bcl-2 proteins. In other hand, we also examined the protein expression of extrinsic apoptotic pathway-related death receptors DR5, Fas, and FasL and the results show that PTFSs had no effects on these proteins (Figure 5c,d). These results suggest that PTFSs may induce apoptosis in A2780/CP70s through the intrinsic pathway rather than extrinsic pathway, which is accordant with the above results of caspase activities. caspase 3/7, -8, -9 activities of control cells were expressed as 100%. (b, c) Effects of PTFSs (0, 1, 2, 3 g/mL) on the expression of caspase-3, -7, -8 and -9 were assayed by Western blotting. GAPDH was utilized for an endogenous reference to standardize protein levels. All values are shown as mean ± SEM of three independent experiments. * p < 0.05 versus control. Cells treated with culture medium containing 0.01% dimethyl sulfoxide (DMSO) was used as the control. PTFSs Induce Apoptosis through the Intrinsic Rather than Extrinsic Apoptotic Pathway in A2780/CP70s As caspase-8 and -9 are respectively the initiators of the extrinsic and intrinsic apoptotic pathway, we further investigated whether the intrinsic or extrinsic pathway participated in the PTFSinduced apoptosis of A2780/CP70s. We first checked the expressions of pro-apoptotic proteins Bax and Bad, and anti-apoptotic proteins Bcl-2 and Bcl-xL, which belong to the Bcl-2 families, related to the intrinsic apoptotic pathway. (Figure 5a,b). The results show PTFSs significantly upregulated Bax and Bad proteins, and downregulated Bcl-xL and Bcl-2 proteins. In other hand, we also examined the protein expression of extrinsic apoptotic pathway-related death receptors DR5, Fas, and FasL and the results show that PTFSs had no effects on these proteins (Figure 5c,d). These results suggest that PTFSs may induce apoptosis in A2780/CP70s through the intrinsic pathway rather than extrinsic pathway, which is accordant with the above results of caspase activities. The changes in the protein levels induced by PTFSs were expressed as quantification histograms with error bars. GAPDH protein expression was detected by Western blotting and utilized for an endogenous reference to standardize protein levels. * p < 0.05 versus control. Cells treated with culture medium containing 0.01% DMSO were used as the control. PTFS-Induced Intrinsic Apoptosis Is p53-Dependent in Ovarian Cancer Cells PTFSs significantly increased the p53 protein expression and phosphorylation at Ser15 of p53 in our study (Figure 6a-c), which prompted us to explore the role that p53 played in PTFS-induced apoptosis in ovarian cancer cells. We firstly pre-incubated cells with p53 specific inhibitor pifithrin- (PFT-) (20 M) for 24 h and performed the Hoechst 33342 and JC-1 assays after the PTFS treatment to check the cell apoptosis. The caspase activity assays were also performed to determine whether the activities of caspase-3/7 and -9 increased by PTFSs being alleviated by the pre-incubated with PFT-. The results show that pre-incubation of PFT- could decrease the apoptotic rate (Figure 6d,e) and JC-1 ratio (Figure 6f) and reverse the PTFS-induced activities of caspase-3/7 ( Figure 6g) and caspase-9 (Figure 6h), which suggests that p53 is involved in PTFS-induced apoptosis in A2780/CP70 cells. Next, the results of western blotting also show the effects of pre-incubation of 20 M PFT- on the intrinsic apoptosis related protein levels in the cells treated with PARP (Figure 6i,j). The overexpression in A2780/CP70 cells of Apaf-1, Bax, cleaved caspase-3, cleaved caspase-9, full length PARP and cleaved PARP after PTFS treatment was decreased by PFT- pre-incubation compared with the control group. Meanwhile, the inhibition of Bcl-xL expression in A2780/CP70 cells after the PTFS treatment was reversed by PFT- pre-incubation compared with control group. Moreover, we knocked out p53 gene by p53 siRNA and then checked intrinsic apoptotic-related proteins. The results show that knockdown of p53 resulted in significant inhibition of p53 overexpression after PTFS treatment. This p53 depletion attenuated the decreasing by PTFS treatment of Bcl-2, and Bcl-xL protein expressions and reversed the PTFS-induced increasing of Bad, Bax and cytochrome c protein expressions. (Figure 6k,l) According to all the results above, it was suggested that p53 is a pivotal mediator of PTFS-induced intrinsic apoptosis in ovarian cancer cells. Meanwhile, it is also hinted that Bcl-2 family proteins, caspase family proteins, DNA damage-related protein p-Histone H2A.X, apoptosis executor cytochrome c and Apaf-1 were regulated in the process of PTFS-induced intrinsic apoptosis through p53 pathway in A2780/CP70 cells. (PFT-) (20 M) for 24 h and performed the Hoechst 33342 and JC-1 assays after the PTFS treatment to check the cell apoptosis. The caspase activity assays were also performed to determine whether the activities of caspase-3/7 and -9 increased by PTFSs being alleviated by the pre-incubated with PFT-. The results show that pre-incubation of PFT- could decrease the apoptotic rate (Figure 6d,e) and JC-1 ratio (Figure 6f) and reverse the PTFS-induced activities of caspase-3/7 (Figure 6g) and caspase-9 (Figure 6h), which suggests that p53 is involved in PTFS-induced apoptosis in A2780/CP70 cells. Next, the results of western blotting also show the effects of pre-incubation of 20 M PFT- on the intrinsic apoptosis related protein levels in the cells treated with PARP (Figure 6i,j). The overexpression in A2780/CP70 cells of Apaf-1, Bax, cleaved caspase-3, cleaved caspase-9, full length PARP and cleaved PARP after PTFS treatment was decreased by PFT- pre-incubation compared with the control group. Meanwhile, the inhibition of Bcl-xL expression in A2780/CP70 cells after the PTFS treatment was reversed by PFT- pre-incubation compared with control group. Moreover, we knocked out p53 gene by p53 siRNA and then checked intrinsic apoptotic-related proteins. The results show that knockdown of p53 resulted in significant inhibition of p53 overexpression after PTFS treatment. This p53 depletion attenuated the decreasing by PTFS treatment of Bcl-2, and Bcl-xL protein expressions and reversed the PTFS-induced increasing of Bad, Bax and cytochrome c protein expressions. (Figure 6k,l) According to all the results above, it was suggested that p53 is a pivotal mediator of PTFSinduced intrinsic apoptosis in ovarian cancer cells. Meanwhile, it is also hinted that Bcl-2 family proteins, caspase family proteins, DNA damage-related protein p-Histone H2A.X, apoptosis executor cytochrome c and Apaf-1 were regulated in the process of PTFS-induced intrinsic apoptosis through p53 pathway in A2780/CP70 cells. PTFSs Induce DNA Damage in Ovarian Cancer Cells The p53 protein plays an important role response to DNA damage. To investigate the relationship of DNA damage and the up-regulation of p53 by PTFSs, DNA damage-related proteins, including p-histone H2A.X, ATM, p-ATM, Chk2, p-Chk2 and MDM2, were determined in A2780/CP70s and OVCAR-3s. The phosphorylation of histone H2A.X at Ser139 suggests DNA double-strand breaks. As shown in Figure 7, PTFSs significantly upregulated the protein expressions of p-histone H2A.X and p-ATM, but had no effect on the protein level of total ATM. Chk2 can be overexpressed in cancer cells and phosphorylated by p-ATM. PTFSs significantly increased the protein level of p-Chk2, reduced that of MDM2, but had no significant effect on that of total Chk2. These results suggest that PTFS treatment could induce DNA damage in both A2780/CP70s and OVCAR-3s. The p53 protein can be phosphorylated by ATM at Ser15, promoting both the activation and accumulation of p53 in response to DNA damage. As PTFSs significantly increase the protein levels of p-p53 (Ser15) (Figure 5a), it could be reasonable deduced that PTFSs induce DNA damage by activating the ATM-Chk2 pathway, and sequentially up-regulate the p53 expression as well as phosphorylate p53 at Ser15. PTFSs Induce DNA Damage in Ovarian Cancer Cells The p53 protein plays an important role response to DNA damage. To investigate the relationship of DNA damage and the up-regulation of p53 by PTFSs, DNA damage-related proteins, including p-histone H2A.X, ATM, p-ATM, Chk2, p-Chk2 and MDM2, were determined in A2780/CP70s and OVCAR-3s. The phosphorylation of histone H2A.X at Ser139 suggests DNA double-strand breaks. As shown in Figure 7, PTFSs significantly upregulated the protein expressions of p-histone H2A.X and p-ATM, but had no effect on the protein level of total ATM. Chk2 can be overexpressed in cancer cells and phosphorylated by p-ATM. PTFSs significantly increased the protein level of p-Chk2, reduced that of MDM2, but had no significant effect on that of total Chk2. These results suggest that PTFS treatment could induce DNA damage in both A2780/CP70s and OVCAR-3s. The p53 protein can be phosphorylated by ATM at Ser15, promoting both the activation and accumulation of p53 in response to DNA damage. As PTFSs significantly increase the protein levels of p-p53 (Ser15) (Figure 5a), it could be reasonable deduced that PTFSs induce DNA damage by activating the ATM-Chk2 pathway, and sequentially up-regulate the p53 expression as well as phosphorylate p53 at Ser15.. (b, d) The changes in the protein levels induced by PTFSs are expressed as quantification histograms with error bars in A2780/CP70 and OVCAR-3 cells. GAPDH was utilized for an endogenous reference to standardize protein levels. Results are expressed as mean ± SEM from three independent experiments. * p < 0.05 versus control. Cells treated with culture medium containing 0.01% DMSO was used as the control. Discussion Ovarian cancer is a lethal gynecological cancer that affect females all around the world. Patients who have undergone cisplatin-based chemotherapy can experience adverse effects or become resistant to treatment, which is a major impediment for ovarian cancer treatment. Natural products have received new interest as a rich source for drug discovery. Saponins have appeared as one of the most promising anti-cancer treatments, as shown by several recent studies. Discussion Ovarian cancer is a lethal gynecological cancer that affect females all around the world. Patients who have undergone cisplatin-based chemotherapy can experience adverse effects or become resistant to treatment, which is a major impediment for ovarian cancer treatment. Natural products have received new interest as a rich source for drug discovery. Saponins have appeared as one of the most promising anti-cancer treatments, as shown by several recent studies. The biofunctions of saponins isolated from the flower bud extracts of C. sinensis in Japan and China, and C. sinensis var. assamica in India have been reviewed. Our previous work demonstrated that saponins isolated from tea (Camellia sinensis (L.) Kuntze) flowers which contain 14 triterpenoid saponins exclusive of Chakasaponin I and Chakasaponin IV can induce apoptosis in ovarian cancer A2780/CP70 cells. In the present study, we firstly purified tea (Camellia sinensis (L.) Kuntze) flower saponins (PTFSs) which mainly consist of Chakasaponin I (80.1%) and Chakasaponin IV (8.1%) (Figure 1). PTFSs were tested for anti-cancer abilities in A2780/CP70 and OVCAR-3. We discovered that PTFSs powerfully inhibit cell viability and cell proliferation of A2780/CP70s and OVCAR-3s, and demonstrated that these inhibitory effects were likely due to the induction of intrinsic apoptosis by regulation of p53 protein expression. The MTS assay showed that PTFSs were cytotoxic to A2780/CP70s and OVCAR-3s, with relatively lower cytotoxicity in the normal ovarian surface epithelial cell line IOSE-364 (Figure 2a). Expression and synthesis of PCNA were reported to be linked with cell proliferation. PTFS treatment significantly decreased protein expression of PCNA of A2780/CP70s and OVCAR-3s (Figure 2b,c), which suggested the inhibition on the cell proliferation by PTFSs. The results are accordance with our previous study. A potential mechanism of chemotherapy drug and radiation resistance in cells might be their resistance to apoptosis. Brightly stained and condensed nuclei were considered to be apoptotic, which were shown by Hoechst 33342 staining in PTFS-treated A2780/CP70s and OVCAR-3s (Figure 3a). The results are in accordance with those from the flow cytometry assay, which show a significantly increased proportion of apoptotic cells in the PTFS-treated groups (Figure 3b). These results suggest that PTFSs induce apoptosis in ovarian cancer cells. In healthy cells, JC-1 is further concentrated within the mitochondria, stimulated through the mitochondrial transmembrane potential (∆m), where it develops into red-emitting aggregates, compared to that in the cytosol, where it presents as a green-fluorescent monomer. Therefore, the red to green JC-1 fluorescence ratio may be utilized as a sensitive quantification of ∆m. A reduction in red fluorescence and an increase in green fluorescence can signal the disruption of ∆m, which is generally thought to be a "point-of-no-return" in the actions that lead up to apoptosis. Our results demonstrate that PTFS treatment decreased the JC-1 ratio (red/green), which leads to the dissipation of ∆m in A2780/CP70s (Figure 3c). Through permeabilizing the mitochondrial outer membrane, intermembrane space proteins, including the apoptotic signaling molecules cleaved PARP and caspase activator cytochrome c, are released into the cytosol. We detected cleaved PARP and cytochrome c protein, which indicates that PTFS treatment significantly increases levels of both proteins and decreases expression of full length PARP (Figure 3d-f) in A2780/CP70s and OVCAR-3s. These data further confirm the stimulation of apoptosis in ovarian cancer cells by PTFS treatment. Apoptosis could be separated into three separate phases including initiation, integration/decision and execution/degradation. Initiation is dependent on the type of the fatal signal, which can develop due to either the intracellular (intrinsic) microenvironment or the extracellular (extrinsic pathway). Caspases are initiators and executioners in apoptosis-related signaling pathways. Initiator caspases include caspase-8, which is necessary for the extrinsic apoptosis pathway, caspase-9 (necessary for the intrinsic one), and executioner caspases (i.e., caspase-3 and caspase-7). The effects of PTFSs on caspase-3/7, -8 and -9 in A2780/CP70 cells were investigated in our study. PTFSs increased the caspase-3/7 and -9 activities and had no effect on the activity of caspase-8 (Figure 4a). At the same time, we detected levels of procaspase-3, -7, -8, -9 and cleaved caspase-3, -7, -8, -9 proteins. The results indicate that PTFSs heighten cleaved caspase-3, -7, and -9, while having no influence on that of cleaved caspase-8 (Figure 4b,c). Caspase activities and Western blotting assay both demonstrated that PTFSs may stimulate apoptosis in A2780/ CP70s by the caspase-9-initiated intrinsic pathway while being independent of the caspase-8-initiated extrinsic pathway. In addition to the damage to the mitochondrial transmembrane potential and releasing caspase activators cytochrome c, participation of the Bcl-2 family of proteins is a vital event in the intrinsic apoptotic pathway. Thus, we further studied the influence of PTFSs on the Bcl-2 family in ovarian cancer cells, including the pro-apoptotic proteins Bax and Bad, anti-apoptotic proteins Bcl-2 and Bcl-xL. On the other hand, for the extrinsic pathway, tumor necrosis factor-related apoptosis-inducing ligand (TRAIL), including Apo2L/TRAIL and Fas ligand (Fas L), would engage their respective death receptors, DR4/DR5 or Fas. Thus, we checked the DR5, Fas L and Fas to further confirm if extrinsic apoptotic pathway signaling participated in PTFS-stimulated apoptosis. As depicted in Figure 5, PTFSs significantly upregulated Bax and Bad proteins, reduced Bcl-xL and Bcl-2, and had no influence on that of DR5, Fas L and Fas. These results suggest that PTFSs induced apoptosis through Bcl-2 family proteins linked to the intrinsic pathway rather than the extrinsic pathway in A2780/CP70s. p53 is a multifunctional tumor suppressor that governs gene expression, cell cycle arrest, DNA repair, apoptosis, oxidative stress glucose metabolism and angiogenesis. It has been reported that p53 stimulates apoptosis through the intrinsic pathway by regulating the Bcl-2 family and caspase family of proteins. The association of cisplatin sensitivity, apoptotic induction and p53 pathway alterations is of popular interest in ovarian cancer therapy. A number of prior studies in vitro and in vivo have shown that p53 alterations were associated with the failure of radiotherapy and chemotherapy in a range of cancers through loss-of-function, dominant-negative activity, or gain of oncogenic function. Additionally, many reports have focused on the p53-dependent anti-cancer effect of natural products [43,. Recently, it has been reported that platinum chemotherapy could induce additional mutations in TP53 and further increase platinum resistance in ovarian cancer. In this study, we suggested that p53 was up-regulated and phosphorylated at Ser 15 by PTFSs and played a vital function in intrinsic apoptosis stimulated through PTFSs. Overall, the results are in agreement with previous reports. We further evaluated the effect on other proteins related with DNA damage such as ATM, p-ATM, Chk2, p-Chk2, and p-Histone H2A.X. A large number of serine and threonine residues of p53 protein are phosphorylated by many different protein kinases. This is, in part, due to the response process from DNA damage to the p53 protein. Activation of ATM by autophosphorylation at Ser 1981 takes place as a response to exposed DNA double-stranded breaks, which is a sensor of DNA damage. Chk2 acts downstream of ATM and is phosphorylated at Thr68 by ATM in responsive to DNA damage. Rapid phosphorylation of histone H2A.X at Ser139 by ATM could result from DNA damage. ATM phosphorylates p53 at Ser15, impairs the capability of MDM2 to bind p53 and encourages the buildup and stimulation of p53 as a response to DNA damage. Our study shows that PTFSs increase the phosphorylation of ATM at Ser1981, histone H2A.X at Ser139, and Chk2 at Thr68 (Figure 7), which suggests that a DNA double-stranded break developed in ovarian cancer cells. Overall, p53 proteins and phosphorylation at Ser15 were significantly heightened in a concentration-dependent way (Figure 6a,b). Prior analyses have indicated that stimulation of ATM/Chk/p53 axis may encourage apoptosis in human pancreatic cancer cells and in ovarian cancer cells. These data suggest that this pathway could be a part of PTFS-stimulated ovarian cancer cell apoptosis. Kitagawa et al. reported that Chakasaponin I, Chakasaponin II and Floratheasaponin A, which were extracted from tea flowers, could inhibit the proliferation of human digestive tract carcinoma cell line HSC-2, HSC-4, MKN-45 and Caco-2 cells with the IC 50 being 4.4-40.6 M, and could induce apoptosis and cell cycle arrest in HSC-2 cells. Our previous paper reported that saponins extracted from tea flowers, which contain 14 triterpenoid saponins exclusive of Chakasaponin I and Chakasaponin IV, could efficiently induce apoptosis in ovarian cancer cells via both intrinsic and extrinsic apoptotic pathways and also induce cell cycle arrest. The present study is the first evidence of anti-ovarian cancer function of Chakasaponin I in a relatively high purity. All three studies demonstrated the anti-cancer functions of tea flower saponins, and suggested that inducing apoptosis might be one of important underlying mechanisms. In addition, the present study reveals that it is through the intrinsic but not the extrinsic pathway that PTFSs induce apoptosis, dependent on p53 in ovarian cancer cells, which might be a more specific and pivotal mechanism, and is unlike the results of our previous paper. This difference in apoptotic mechanisms might be due to the different saponins we used. It is reported that differences in saponin structure, which include the type, position, and number of sugar moieties attached by a glycosidic bond at different positions of the rings, can characteristically influence biological responses, especially for the antitumor activity. More studies focused on the anti-cancer mechanisms using one specific monomer are needed to explore the structure-function relationship of saponins in the anti-cancer mechanisms. In addition, several papers have reported the anti-cancer functions of triterpenoid saponins via different mechanisms. Momordin Ic has been reported to suppress liver cancer cell invasion via COX-2 inhibition and PPAR activation. Saikosaponin D extracted from Radixbupleuri has been reported to induce apoptosis and to block autophagic degradation in breast cancer cells. Raddeanin A extracted from Anemone raddeana has been reported to suppress the angiogenesis and growth of human colorectal tumor by inhibiting VEGFR2 signaling. All the above mechanisms are worth exploring in future research for anti-cancer functions of saponins extracted from tea flowers. As many saponins exhibit pharmacological properties, a review of the absorption, disposition, and pharmacokinetics of saponins had been done. Recently, the pharmacokinetics and excretion studies of sapindoside B in rat were conducted and reported the pharmacokinetic profiles of sapindoside B in the range of 2.5 to 12.5 mg/kg dosage-dependently, and only 2% intravenous dose of sapindoside B was excreted in its parent form over 48 h. More research is needed to explain the contradiction between the low bioavailability of saponins and their potential anticancer activity in vivo. New drug delivery systems for saponins may be an effective strategy for developing these compounds into medicinal products. In conclusion, our results indicate that PTFSs have a strong and preferential inhibition of cellular proliferation on A2780/CP70s and OVCAR-3s compared to IOSE-364s via intrinsic apoptosis rather than the extrinsic pathway. We also found that p53 protein has a crucial function in intrinsic apoptosis induced by PTFS treatment, and downstream proteins like Bcl-2 families and caspase families were successively regulated. Additionally, these activities were at least partly induced by DNA damage through the ATM/Chk2/p53 axis. Our data help us to understand the mechanisms through which PTFSs may lead to preventing and treating platinum-resistant ovarian cancer. PTFSs were composed of Chakasaponin I and IV. Chakasaponin I may be a potential natural compound for treating platinum-resistant ovarian cancer, and more anti-cancer bioactivities of saponins of tea (Camellia sinensis (L.) Kuntze) flower are worth exploring in the future. Preparation of Purified Tea Flower Saponins (PTFSs) Dried tea (Camellia sinensis (L.) Kuntze) flowers (Zhejiang Yilongfang Co., Quzhou, China) were ground and then extracted with 70% methanol through refluxing at 60 C for 2 h. The methanol isolate was then divided into ethyl-acetate and n-butanol soluble layers. The n-butanol layer was vacuum concentrated, collected, underwent D101 column chromatography, and eluted through 0%, 15%, 30%, 45%, 60%, 75% and 90% ethanol-aqueous solution (v/v) at a rate of 2 bed volume (BV, 2.5 L)/h each for 1 BV. The 90% ethanol-eluted fraction was collected, concentrated and purified by reversed-phase preparative middle-low pressure liquid chromatography system (GE KTA purifier100, Uppsala, Sweden) alongside a SinoChrom ODS-BP column (5 m, 250 mm 10.0 mm i.d., Elite, Dalian, China). Elution was conducted using a mobile phase composed of formic acid, water and acetonitrile (0.1:49.9:50.0, v/v/v) at 210 nm at 1.5 mL/min to acquire five fractions. The first one was additionally filtered using a Waters XBridge Shield RP18 column (Waters, Milford, MA, USA) to produce PTFSs. Mobile phase A included formic acid and water (0.1:99.9, v/v), and B included formic acid and acetonitrile (0.1:99.9, v/v). The elution gradient was 42% B for 13 min, 45% B for 20 min and 47% B for 15 min. The rate of the mobile phase was 1.5 mL/min. After separation, PTFSs were collected and measured by UPLC coupled with electrospray ionization quadrupole time-of-flight mass spectrometry (UPLC-Q-TOF/MS/MS), as previously explained. PTFSs were dissolved in DMSO to prepare a solution with the final concentration of 20 mg/mL, and was kept at −20 C DMSO was bought from Sigma-Aldrich (Sigma, St. Louis, MO, USA). Cell Culture A2780/CP70 and OVCAR-3, two human ovarian cancer cell lines, were provided by Dr. Jiang at West Virginia University. IOSE-364, a normal ovarian surface epithelial cell line, was kindly supplied by Dr. Auersperg at the University of British Columbia. Cell lines were maintained in RPMI 1640 with 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin solution. Cells were maintained in an incubator with 5% CO 2 at 37 C RPMI-1640 and bovine serum albumin (BSA) was bought through Sigma-Aldrich (Sigma, St. Louis, MO, USA). Penicillin-streptomycin solution was bought through Thermo Scientific (Waltham, MA, USA). FBS and phosphate-buffered saline (PBS) were bought through Life Technologies (Invitrogen, Grand Island, NY, USA). Cell Viability The MTS assay was performed to measure cell viabilities of A2780/CP70, OVCAR-3 and IOSE-364. Briefly, cells were seeded in 96-well plates (1 10 4 cells per well) for 24 h. Then, various dosages of PTFSs (0-3.5 g/mL) were added with an equivalent volume of DMSO as control. Cells were washed twice with PBS after incubation for 24 h, and 100 L of fresh Aqueous One Solution was placed into each well. After incubating at 37 C for 1 h, the absorbance of cells was read at 490 nm using a microplate reader (BioTek, Winooski, VT, USA). Cell viability was expressed as a proportion of controls. CellTiter 96 AQueous One Solution Cell Proliferation Assay was bought through Promega (Madison, WI, USA). Apoptosis Analysis by Flow Cytometry A2780/CP70s and OVCAR-3s were seeded in 60 mm plates (1 10 6 cells/dish). After cells adhered for 24 h, various doses of PTFSs (0, 1, 2, 3 g/mL) were added for another 24 h and harvested for additional studies. For apoptosis examination, an Alaxa Fluor 488 Annexin V/Dead Cell Apoptosis kit (Thermo Scientific, Waltham, MA, USA) was utilized. Treated cells were harvested and cleaned twice using cold PBS, which was followed by the resuspension of cells in PI and annexin V buffer, as per the manufacturer's established protocol. Samples were assessed utilizing a FACSCaliber flow cytometry system (BD Biosciences, San Jose, CA, USA). Cellular Caspase Activity Assay A2780/CP70s were seeded into 96-well plates (1 10 4 cells per well). After cells adhered for 24 h, different concentrations of PTFSs (0-3 g/mL), with or without 20 M PFT-, were added and the cells were cultured for another 24 h. An equivalent volume of DMSO was added to the control groups. Caspase-Glo-3/7, -8 and -9 (Promega, Madison, WI, USA) regents were supplemented and incubated for 30 min. Luminescence was measured through the use of a microplate reader (BioTek, Winooski, VT, USA). Analyses were performed three times and results were conveyed as a proportion of controls. Western Blotting Proteins were determined by specific monoclonal antibodies in Western blot analysis. PTFSs were added to A2780/CP70s and OVCAR-3s, with or without 20 M PFT- or p53 siRNA, for 24 h as previously shown. Protein was separated using Mammalian Protein Extraction Reagent and 1% protease inhibitor cocktail. As per our prior study, the total protein concentration was measured by a BCA protein assay kit (Pierce, Rockford, IL, USA). Equivalent levels of protein were loaded onto SDS-PAGE and transferred onto nitrocellulose membranes. The membrane was placed in blocking buffer at room temperature for 1 h, and incubated with targeted primary antibodies at 4 C overnight. The next day, membranes were incubated with secondary antibodies. After washing, the protein was visualized using the ChemiDocTM MP System (Bio-Rad, Hercules, CA, USA). GAPDH was used to normalize relative values of each protein. Primary antibodies targeting Apaf-1, Bax, Bcl-XL, Bcl-2, cytochrome c, cleaved caspase 3, caspase 3, caspase 7, caspase 8, caspase 9, DR5, FasL, Fas, p53, p-p53 (Ser15), PARP, PCNA, p-Histone H2A.X(Ser139), MDM2 and secondary antibodies were bought through Cell Signaling Inc. (Danvers, MA, USA). Primary antibodies targeting p-Chk2 (Thr68), Chk2 (H-300), Bad, and GAPDH were bought from Santa Cruz Biotechnology (Dallas, TX, USA). Statistical Analysis The data were exhibited as mean ± standard error of mean (SEM) from at least 3 independent experiments, and at least 3 biological replications for each independent experiment have been done. Data were assessed by SPSS (SPSS, Chicago, IL, USA) using the Shapiro-Wilk normality test, one-way analysis of variance (ANOVA) and post-hoc test (2-sided Dunnett's test). p-values < 0.05 represent statistical significance. of the National Institutes of Health (NIH) and its contents are solely the responsibility of the authors and do not necessarily represent the official view of NIGMS or NIH. This study was also supported by COBRE grant GM102488/RR032138, ARIA S10 grant RR020866, FORTESSA S10 grant OD016165 and INBRE grant GM103434. Acknowledgments: We thank Kathy Brundage from the Flow Cytometry Core at the West Virginia University for providing technical help on apoptosis analysis. We would also like to extend our thanks to Zhiwei Ge at the Analysis Center of Agrobiology and Environmental Sciences, Zhejiang University for his help on LC-MS analysis. Conflicts of Interest: The authors declare no conflict of interest. |
Dynamic input-dependent encoding of individual basal ganglia neurons Computational models are crucial to studying the encoding of individual neurons. Static models are composed of a fixed set of parameters, thus resulting in static encoding properties that do not change under different inputs. Here, we challenge this basic concept which underlies these models. Using generalized linear models, we quantify the encoding and information processing properties of basal ganglia neurons recorded in-vitro. These properties are highly sensitive to the internal state of the neuron due to factors such as dependency on the baseline firing rate. Verification of these experimental results with simulations provides insights into the mechanisms underlying this input-dependent encoding. Thus, static models, which are not context dependent, represent only part of the neuronal encoding capabilities, and are not sufficient to represent the dynamics of a neuron over varying inputs. Input-dependent encoding is crucial for expanding our understanding of neuronal behavior in health and disease and underscores the need for a new generation of dynamic neuronal models. Characterizing the computational properties of individual neurons is crucial for understanding their contribution to normal brain function and its breakdown during different pathologies. The basic unit of the computation the neuron performs on its inputs into action potentials is a transformation typically termed "neural encoding". An essential set of tools for studying this encoding are computational models, which are a simplified representation of the neurons. By modeling the relationship between stimuli (e.g., visual scenes) and the spiking responses of neurons, computational models aim to account for the mechanisms underlying neuronal computation 1,2. Modeling this relationship serves to predict neuronal responses under different conditions, and facilitate the study of the fundamental properties of neuronal encoding in the normal state as well as changes in different pathological states 3,4. A variety of computational models have been put forward to approximate the activity of experimentally recorded neurons at different levels of abstraction while maintaining biological plausibility. This abstraction level ranges from simple models, such as the integrate-and-fire model, to biologically detailed models. The choice of abstraction level depends on the particular goal of the model, and the question addressed by the model. Computational models can be divided into deterministic and statistical models. Deterministic models include most biophysical models, with a broad range of complexity, and typically map the model's parameters to biophysical properties. In contrast to deterministic models in which the same input will always result in the same output, statistical models reflect the stochastic response of neurons to similar repetitive input. Statistical models do not typically utilize parameters which have a biophysical meaning but instead represent conceptual properties related to the neuron's computation 1. A common statistical model is the generalized linear model (GLM), a generalization of linear regression which relates a linear model to observations from any probability model in the exponential family via a non-linear linking function. The GLM framework has been adapted to point process data, which are often referred to as point process GLM (PP-GLM) 8,9, and has become a standard way to analyze and model spike train data. Here, for simplicity, we use the general term GLM to refer to PP-GLM. The GLM typically incorporates a linear stimulus filter which accounts for stimulus encoding, a spike history function which captures effects such as refractory periods, bursting and other non-Poisson features of spike train statistics, and a bias term which reflects tonic firing. These filters are combined to generate an input to a Poissonian neuron. GLMs have been shown to outperform other models, both for predicting spike trains (neural encoding) and for performing stimulus reconstruction (neural decoding) 12,13. GLMs have been applied to neurons in sensory systems, where they associate known Results Modeling cellular integration of input using GLMs. We recorded the activity of globus pallidus (GP) (n = 57), entopeduncular nucleus (EP) (n = 29) and substantia nigra pars reticulata (SNr) (n = 32) neurons during in-vitro whole-cell recordings following the blockade of GABAergic and glutamatergic transmission. We injected a "frozen noise" current, which was generated by convolving white noise with an alpha function, into the soma of the neurons repeatedly over multiple trials (30-50 trials) separated by periods without noise injection (see methods). This current approximates natural inputs received by the neurons in vivo 21. The spiking activity was identified from the recorded intracellular potential of the neurons and was used to generate a point process (Fig. 1A). The neuronal encoding of the stimulation was estimated using a generalized linear model (GLM) utilizing a non-linear function (exponent) over the linear filters of the incoming stimulus, the post spike filter and a constant bias term (Fig. 1B, see methods). The parameters of the GLM were fitted to the neuronal responses to maximize the fitting to the training set (using the initial 80% of the stimulation period). The resulting GLM was then used to generate simulated spike trains, and the smoothed PSTH was compared to that of the experimental neuron using the test set (last 20% of the stimulation period) and scored using Pearson's correlation coefficient (PCC). The GLM accurately reproduced the responses of the recorded cell, and the resulting firing rate provided an estimation of the in-vitro recorded firing rate (Fig. 1C). Overall, the GLM representation of the neurons demonstrated highly correlated responses to the experimental neuronal responses (Median PCC: GP: 0.86, EP: 0.83, SNr: 0.83) (Fig. 1D). the parameters of GLMs are dependent on the input statistics. We next examined the "invariance assumption" underlying most studies which use GLMs 2,12,13,22. This assumption is that the neuron's fitted GLM parameters are invariant to the input statistics, such that different inputs to the same neuron will result in the generation of similar GLMs. The essence of this assumption implies that the neuron's computational properties are stable across input conditions. We changed the mean and variance of the frozen noise injected to the neuron and compared the resulting GLMs. We varied the constant direct currents (DCs) added to the fluctuating noise, and this affected the mean firing rate ( Fig. 2A), and both the fitted stimulus filter (Fig. 2B) and the post spike filter (Fig. 2C) of the generated GLMs, such that different filters were generated based on the baseline current. In the example neuron, the increase in the rate correlated with an increase in the peak magnitude of the stimulus filter and a shortening of the inhibitory period (filter values ≪1) of the post spike filter. An increase in the variance (between 10 to 50 pA) of the frozen noise while maintaining the same baseline DC led to similar increases in the firing rate (Fig. 2D), and variations in both the stimulus filter (Fig. 2E) and the post spike filter (Fig. 2F). In the example neuron, the increase in variance correlated with a decrease in peak magnitude of the stimulus filter and a shortening of the inhibitory period of the post spike filter. In order to rule out the possibility that these results were due to the relatively short stimulation durations used for generating the GLMS leading to a limited number of spikes, we recorded neuronal activity while injecting a long (30 second) current of frozen noise. We then generated GLMs by first taking only the first 5 seconds of the recording (Fig. 2G, top), and then using the whole recording duration (Fig. 2G, bottom). The longer recordings only slightly improved the divergence of the GLM parameters over different baseline firing rates and resulted primarily in a smoother representation of the GLM parameters. Using filters of longer durations did not change the results as well, as the filters converge to 0 for longer latencies. Thus, using short stimulation periods sufficed to generate GLM parameters, and was not the cause of their variations across different input properties. Adding a firing rate filter which captured the previous mean firing rate of the neuron produced similar results, where the shape of both the stimulus filters and the post spike filters varied with the change in the firing rate of the neuron (Fig. 2H). GLMs fitted over different input statistics are not interchangeable. The dependence of the GLM fitted parameters on the properties of the input is not merely a case of different parameter combinations that lead to similar model performance. The different GLMs are not interchangeable; i.e., when a model neuron receives the same input, with two different DC levels that lead to two different baseline firing rates, the two GLMs generated using each condition do not only look different (Fig. 3A), but also generate different predicted spike trains and form different PSTHs (Fig. 3B). In the example, the spike trains generated from the GLM that was fitted with the different baseline current were less sparse and more accurate in time than the experimental spike trains, as well as those generated with the GLM created with the same current baseline as the test data. Therefore, when comparing the predicted PSTH to the real PSTH of the test data, only the GLM that was fitted using the same recording conditions as the test data generated responses that were highly correlated to the responses of the www.nature.com/scientificreports www.nature.com/scientificreports/ experimental neuron to the incoming current. Using the GLMs of the same neuron that were generated from recordings with different current properties led to responses that differed from the real responses (Fig. 3C). We tested 42 neurons recorded over multiple sessions, where in each session, we injected the same noise with a www.nature.com/scientificreports www.nature.com/scientificreports/ different baseline current. The inability to generalize GLMs from different firing rates of the same neuron was a prominent feature and was expressed in most of the sessions pairs as a reduced correlation between the responses (Fig. 3D). This trend of reduced generalization was smaller when the firing rate of the original neuron was higher than the tested neuron, which indicates that some generalization exists in this case. GLM parameters are dependent on the firing rate. These results suggest that neurons' computational properties can differ as a function of input properties. To quantify the systematic changes in these properties, we analyzed all the GLMs that we generated from the experimental neuron recordings. In some of these neurons, we recorded activity during multiple sessions while changing the DC, which yielded a dataset with diverse firing rates. We analyzed 119 sessions from GP neurons, 61 sessions from EP neurons, and 51 sessions from SNr neurons. In each of the nuclei, the GLM parameters were highly dependent on the firing rate of the neuron. The post spike filter exhibited a strong negative correlation between the duration of inhibition and the firing rate (Fig. 4A). To analyze the stimulus filter, we calculated the total coincidence detection window; i.e., the duration of the positive part of the filter where the neuron sums the arriving inputs, and the effective coincidence detection Generalization of GLMs calculated using different firing rates of the same cell leads to reduced accuracy. The GLM generated using a specific DC, which resulted in a specific firing rate, was then used as a model neuron to generate responses to a novel stimulus with different constant DCs. The model PSTHs were tested against the real responses of the cell using the correlation measure as in B. (D) Same as c, but for the whole population. GLMs were generated using a specific firing rate of a neuron, and their prediction accuracy was assessed (the mean of the whole population represented as a firing rate difference of zero, blue bar). The GLM was then used as a model neuron to predict responses of other firing rates of the neuron (represented as ± firing rate difference); the PSTHs correlations were measured, and their mean for each firing rate difference is presented. Significance is measured in relation to the mean PCCs of the original GLMs (firing rate difference of zero), and only significant mean correlations are presented. Black solid lines represent the exponential fitting functions for the positive (R 2 = 0.48, p < 0.001) and negative (R 2 = 0.32, p < 0.01) firing rate differences. 10:5833 | https://doi.org/10.1038/s41598-020-62750-0 www.nature.com/scientificreports www.nature.com/scientificreports/ window which measures the effective duration in the positive part, when the shape of the stimulus filter displays a sharp decay to zero. Both the total and the effective coincidence detection windows were significantly negatively correlated with the firing rates in all the nuclei, such that higher firing rates narrowed the coincidence detection windows (Fig. 4B,C). We further analyzed the spectral selectivity of the stimulus filters by looking at the characteristic frequency of the Fourier transform of the filter (Fig. 4D). Similar to the coincidence detection windows results, the spectral results presented a positive correlation between the firing rates and the main frequency, and indicated that neurons that fired at low rates acted as low pass filters, whereas at higher firing rates the same neurons tended to display a high pass filtering activity. This phenomenon could also be seen at the single neuron level using wavelet analysis: when recording the same neuron over different firing rates, the neuron's stimulus filters changed their spectral selectivity depending on the firing rates (Fig. 4E). www.nature.com/scientificreports www.nature.com/scientificreports/ Verification of experimental results using model neurons. We tested the results we collected experimentally by comparing them to simulations using a detailed biophysical neuronal model (implemented with the NEURON modeling environment). This served to verify a wide range of input parameters (mean and variance) and specifically long-term activity. The GLM representation of the spike trains generated by the detailed biophysical neuronal model exhibited the same dependence on the firing rates, such that the shape of both the stimulus filters and the post spike filters were dependent on the baseline firing rate (Fig. 5A,B). Furthermore, as in the experimental neurons, the filters were not interchangeable (Fig. 5C). The fidelity and accuracy of the neuronal transformation depends on the firing rate. Next, we examined whether the changes in the computational properties of the experimental neurons during different firing rates affected the information transmission of those neurons. A key property of information processing is the maintenance of reliability; i.e., a particular stimulus will lead to similar neuronal activity each time it is presented. Another feature of information processing is the temporal precision of the responses, which is the timescale of the jitter between responses on different trials to the same input. The precision and the reliability were derived from the smoothed PSTH of the responses of the neuron to multiple injections of the same frozen noise, thus representing the instantaneous firing rate. We then identified elevations in the PSTH, which we termed events. The temporal precision was defined as the mean widths of the event, and the reliability was defined as the fraction of spikes that occurred within the event. The precision of single sessions was negatively correlated with the firing rate in all the nuclei (Fig. 6A), and the reliability had a significant asymptotic relation with the firing rate (Fig. 6B). The combination of the dependence of these features on the firing rate was expressed as the distance between pairs of spike trains from the same session where pairs of spike trains with higher firing rates are expected to be more similar. We computed the distance using the Victor-Purpura metric (q = 0.15) 23,24. In order to rule out the effect of the firing rate, we used a normalized version of this distance 25. For each session, the distance was defined as the mean distance between all pairs of spike trains within the session. As expected, in all the nuclei, there was a significant negative relationship between spike train distance and firing rate (Fig. 6C). Similarity of neuronal encoding of pairs of neurons within and between nuclei. Finally, we examined the relative effect of the firing rate of a neuron relative to other factors that determine neuronal encoding properties. We analyzed the similarity between neuronal responses of different neurons, within and between nuclei that fired at the same baseline firing rate and received similar input. The assumption was that responses of neurons within the same nucleus would be more similar to each other than to those of different nuclei, as a result of the common morphology and biophysical features of neurons within the same nucleus. If the responses of neurons in different nuclei are similar to responses of neurons in the same nucleus, this would suggest that the firing rate of the neuron influences its computation more than its biological properties. We applied the Victor-Purpura distance to characterize the similarity and dissimilarity of neuronal responses of pairs of neurons receiving the same input noise and having the same baseline firing rates during the recording. Our results showed that in some of the cases, the neuronal responses of pairs of neurons within the same nucleus were as similar as pairs of neuron between nuclei (Fig. 7). This may indicate that the firing rate of the neuron plays a significant role in the computation and the encoding properties of neurons. Discussion Context independent modeling studies assume that a fixed set of parameters can model a neuron and that this model is invariant to input statistics. In this study, we challenged this assumption, using PP-GLMs as a representative model for this type of statistical models. We fitted GLMs to the in-vitro whole-cell responses of BG neurons to repeated stimulation. In order for our results to be generalized to in vivo neurons, the injected current should simulate the in vivo environment, i.e., inputs from active pre synaptic neurons. Thus, we injected currents that mimic those inputs received in vivo. The input statistics were varied across sessions, influencing the baseline firing www.nature.com/scientificreports www.nature.com/scientificreports/ rate of the neurons, and forming their internal state. The GLMs accurately reproduced the responses of the experimentally recorded cells in single sessions. However, these static models were not able to reproduce the neuronal spiking activity in different sessions and different states of the neurons. Furthermore, the parameters of the GLMs were found to be highly sensitive to the firing rate of the neurons. Using the GLMs, we quantified the encoding of individual neurons and demonstrated how various encoding properties are dependent on the baseline firing rate. We further demonstrated that the information processing properties of single neurons vary with their firing rates. We found similar results for input dependent computation in detailed and simplified biophysical models as apparent in the experimental dataset. This indicates that the input statistics have a major influence on the computation properties of neurons, such that there is no single "real" static encoding function of a neuron, but rather an input dependent encoding scheme. This study combined modeling and experimental approaches to better understand the encoding properties of individual neurons. In terms of modeling, we used GLMs, which incorporate a linear stimulus filter, a spike history filter, and a bias term. The stimulus filter represents computational properties that are related to the input to the neuron, the spike history filter represents the influence of previous spikes on the current response, and the bias term reflects the tonic firing of the neuron. Together, these filters are commonly assumed to represent the computational properties of the neuron to a large extent. The simplicity of the GLM allows for direct interpretation of the encoding properties of the neuron from the model parameters. Using the GLMs, we demonstrated how static models are insufficient to computationally describe single neurons since computational properties change with the input statistics. In order to rule out the possibility that simply adding complexity to the model would resolve the input dependent computation changes illustrated by the GLMs, we tested our results using an extension of the GLM with a firing rate filter. However, this model did not qualitatively change the overall results, thus hinting that merely adding more complexity to the model cannot overcome this limitation of the static model. Specifically, these changes fail to produce a model which is invariant for a wide range of inputs, because this phenomenon is a property of the neuron and not a side effect of a specific model. It is also not a property derived from the duration of the recording session since models fitted on short trials exhibited similar properties to those fitted on much longer trials. Thus, model parameters that are fitted with a specific input are unfit for reproducing the responses of the same neuron to other inputs with different statistics. Instead, these myriads of static models provide valuable qualitative insights into the wide range of possible encodings by single neurons. These findings are critical for modeling neurons and investigating their computational principles. The firing rates of real neurons typically fluctuate over multiple time scales and over different behavioral states 26,27. The resulting computation during these different states is not static but rather changes as a function of conditions. Thus, in order to capture the variety of the neuron's computational properties, experiment must include different states of the neuron, instead of recording the neuron in a single state. Modeling and interpreting the encoding of a neuron in one state might not be pertinent to other states for the same neuron. Previous studies have attempted to overcome these limitations by including the information of the instantaneous membrane potential 28. These, for example, include models which utilize the state space of the voltage 29, and models that incorporate the non-linear dynamics of the firing rate threshold into the model 19,30,31. Variations in spike threshold might affect neuronal computation mechanisms; however, using model neurons, our results suggest that the mechanisms underlying these input dependent computations are more fundamental and complex. Here, input dependent computation was demonstrated in detailed biophysical neurons. This links the underlying mechanisms to some complex www.nature.com/scientificreports www.nature.com/scientificreports/ mechanisms, which are characteristics of detailed biophysical models. These may include voltage dependent active membrane properties, such as activation of slow potassium currents, sodium inactivation, afterhyperpolarization currents, and changes in spike generation mechanisms due to various voltage dependent changes. The computation of single neurons has been extensively investigated, and in this framework, different inputs are already known to produce varying firing rates and varying responses in single neurons. In some neural systems (e.g. motor control, visual system), the relationship between stimuli and neural responses has been studied in depth 32. The computational properties in these conditions have been examined in part in the context of gain modulation; i.e., the change in the input-output function of a single neuron. In addition to the somatic computation which we studied, neuronal computation also consists of dendritic computation, which has been shown to be more than simple transmitting devices, in that their nonlinear summation of inputs provide the neuron with a variety of computational functions. Here, we explored the influence of the input statistics on the computation inside the soma itself, after receiving and integrating all the inputs from the dendrites. Thus, we blocked all synaptic inputs while injecting current into the soma of the neurons, thus excluding the contribution of the dendrites to this computation, The blockage of synaptic inputs also excluded the influence of neuronal circuits and functional connectivity, both of which are known to change dynamically and influence neuronal computation 39. Thus, the reconfiguration of computation over different inputs was not influenced by external network connections. Previous studies have shown how neuronal computation depends on the statistical properties of the input 18,40,41. In this study, the use of GLMs enabled a direct interpretation of the underlying computational principles, since mapping from the GLM filters to computational meaning was straightforward. We showed the dominant dependency of the computation on the internal state of the neuron, such that when the neuron is in a higher firing rate mode, it tends to trend towards functioning as a coincidence detector, while at lower firing rates it has integrator properties. These features were also reflected in the spectral properties of the model filters, where at high firing rates neurons tended to display high pass filtering activity compared to low pass filtering in lower firing rates. Thus, neurons should not be considered as either integrators or coincidence detectors 42. Our results indicate that these features may vary over states in individual neurons. These properties were consistent across neurons in the different nuclei of the BG. Similar results were observed in the analysis of interneurons in the cortex ( Supplementary Fig. 1), which hints that this phenomenon is not limited to BG neurons. These changes also impact information transmission in that neurons tend to transmit information more reliably and precisely at high firing rates than at lower firing rates. These findings are applicable to any change in neurons' baseline firing rates, which can occur as a result of changes in behavioral state, and in different pathological conditions that influence the firing rates of the neuron. An example of such a pathology is Parkinson's disease which leads to firing rate changes throughout the BG 43,44. Our results imply that this input-dependent computation might contribute to the deficiency in the computational capabilities of these neurons in these different disorders. Subsequent to the changes in the baseline firing rates, the computational properties of the neurons shift to a baseline region which is different from their original baseline, thus influencing their mean computational performance. Finally, we showed how the responses of pairs of neurons in different nuclei were as similar as pairs of neurons within the same nucleus. This finding indicates that the firing rate of the neuron has a considerable influence on its responses, and in some cases is more influential than the morphology and the biological features of the neuron. Overall, our results suggest that neurons have a repertoire of computational properties which are influenced by the input properties and the baseline firing rate of the neuron. Moreover, information processing is dynamic and dependent on the state of the network. Thus, before modeling any neuron, scientists must understand the complexity of the physiological environment. Then, novel dynamic models are needed to provide a better understanding of neuronal behavior in health and disease. Methods In vitro slice preparation. Brain slices were obtained from 14 to 21-day-old Wistar rats as previously described. Rats were killed by rapid decapitation according to the guidelines of the Bar-Ilan University Animal Welfare Committee. This procedure was approved by the National Committee for Experiments on Laboratory Animals at the Israeli Ministry of Health. The brain was quickly removed and placed in ice-cold artificial cerebrospinal fluid (ACSF) containing (in mM): 2.5 KCl, 125 NaCl, 25 glucose, 1.25 Na 2 HPO 4, 2 CaCl 2, 15 NaHCO 3, 1 MgCl 2 and 0.5 Na-ascorbate (pH 7.4 with 95% O 2 /5%CO 2 ). In all experiments the ASCF contained APV (50 M), CNQX (15 M) and GABAzine (20 M) to block NMDA, AMPA and GABA receptors, respectively. Thick sagittal slices (320-350 m) were cut on an HM 650 V Slicer (Microm International, Walldorf, Germany) and transferred to a submersion-type chamber where they were maintained for the remainder of the day in ACSF at room temperature. Experiments were carried out at 37 °C, and the recording chamber was constantly perfused with oxygenated ACSF. In vitro electrophysiology. In vitro recordings form GP, EP and SNr neurons were done as previously described 46,48,49. Individual GP, EP and SNr neurons were visualized by infrared differential interference contrast microscopy using an Olympus BX51WI microscope with a 60x water immersion objective. Electrophysiological recordings were performed in the whole-cell configuration of the patch-clamp technique under visual control using a CCD camera. Recordings were obtained from the soma of GP, EP and SNr neurons using patch pipettes (4-8M) pulled from thick-walled borosilicate glass capillaries (2.0 mm outer diameter, 0.5 mm wall thickness, Hilgenberg, Malsfeld, Germany). The standard pipette solution contained (in mM): 0.5 EGTA, 10 HEPES, 140 K-gluconate, 4 MgATP, 10 NaCl, 5 L-glutathione, 0.05 Spermin, and 0.4 GTP (pH 7.2 with KOH; Sigma, St Louis, MO, USA). Under these conditions, the Nernst equilibrium potential for chloride was calculated to be -69 mV. The reference electrode was an Ag-AgCl pellet placed in the bath. Voltage signals were amplified by an 10:5833 | https://doi.org/10.1038/s41598-020-62750-0 www.nature.com/scientificreports www.nature.com/scientificreports/ Axopatch-200B amplifier or Axopatch-700B (Axon Instruments, Union City, CA, USA), filtered at 5 kHz and sampled at 10 kHz. The 10-mV liquid junction potential measured under these ionic conditions was not corrected for. The recordings were conducted while injecting a filtered white noise current stimulus. The noise traces were a 5 second white noise current convolved with an alpha function with a 3 ms rise time, and each stimulus was presented for ~50 repetitions, separated by periods without noise injection. This stimulus is known to produce reliable spiking 13,50. In some of the sessions we injected the fluctuating current plus a steady state current to induce variable firing rates. The dataset only included recordings displaying stable input resistance, a stable I-F curve, and stable firing patterns across trials throughout the experiment. Generalized linear model. A GLM was fitted to each recording session. The models consisted of a temporal stimulus filter k, a post-spike history filter h, and a constant bias term b 13,15. The stimulus and history filters were represented with 15 and 25 boxcars basis functions of 4 ms, respectively, spaced equally in time. The time-dependent firing rate of each neuron was modeled as (t) = exp(k x + h r + b), where x was the stimulus and r was the neuron's spike train history. The stimulus filter and spike history filter were defined as vectors whereas the bias term was a scalar. Before fitting, the stimuli were down-sampled to 1 KHz and normalized by subtracting the DC component and dividing by the amplitude of the stimulus noise, as in standard procedures 12. The model was trained using 80% of the stimulus presentations from all the trials, and the validation was done on the remaining 20%. Following maximum likelihood fitting, the GLM converges to a global error minimum 51. The quality of the fit was assessed by comparing the PSTH of simulated spike trains to that of experimentally recorded spike trains computed from the test data, using Pearson's correlation coefficient. The PSTHs were smoothed using an 11-ms Gaussian window. Detailed biophysical neurons simulations. We simulated detailed biophysical neurons from detailed neuronal model descriptions provided by the Neocortical Microcircuit Collaboration Portal 52,53. These neuron models were constructed to correspond to experimental cortical neurons from layers 2/3 and layer 5. We used the NEURON simulation environment 54 to generate voltage responses to current step injections in these models. Data availability The data sets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request. code availability Data analysis was performed using MATLAB (Mathworks) and python. The code is available from the corresponding author upon request. |
A growing number of secure mobile solution providers say the answer to BYOD is not to control the device, but to control the data.
When it comes to things that keep CIOs up at night, mobility, particularly bring your own device (BYOD), is at the top of the list or near it. Mobile device management (MDM) products and services are often the reflexive response to the need for more secure mobile computing, but in many ways that's like using a chainsaw rather than a scalpel to perform surgery. A growing number of secure mobile solution providers say the answer to BYOD is not to control the device, but to control the data.
"It's appropriate to manage the device if you own that device," says Alan Murray, senior vice president of products at Apperian, a provider of a cloud-based mobile application management (MAM) solution. "If the corporation owns the device, it should manage that device. When is it valid to manage the application? Always."
Smartphones are now in the hands of hundreds of millions of employees around the world, and other mobile devices like tablets are a growing phenomenon as well. This influx of consumer-owned devices into the enterprise environment has sparked data loss fears within many IT organisations. And if you think it's not happening in your company, think again.
"Even if you don't think you're doing BYOD, you're doing BYOD," Murray says. "It's a matter of whether you're doing it formally or like an ostrich."
For the most part, organisations are adjusting to the new reality. According to the State of Mobility Survey 2012 by Symantec, 59% of the 6,275 respondents reported that their organisations are making line-of-business applications accessible from mobile devices, and 71% said their organisation is looking at implementing a corporate "store" for mobile applications.
It's not hard to see why. Organizations believe embracing mobile computing increases the efficiency and effectiveness of their workforces. Symantec's survey found that 73% of respondents expected to increase efficiency through mobile computing, and all of them did realise that increased efficiency.
"Four or five years ago, it was all about the mobile elite," says John Herrema, senior vice president of corporate strategy for Good Technology, provider of secure mobile solutions. "They had company-owned devices to do some pretty basic things around email, browsers and PIM. Apps never really took off on that platform for a variety of reasons. But what we're seeing now is these BYOD devices have a ton of corporate use. Users are self-reporting that they're doing the equivalent of an extra week of work a month on their mobile devices by doing things like checking their email before they go to bed. The devices are out there. The users want this access. The more you give them access, the harder and longer they work. If you can't find a way to overcome the security concerns, you are leaving massive amounts of productivity on the table."
Several different strategies are emerging to help organisations control their data in a mobile environment. One of the more popular strategies is MAM, often associated with the creation of curated enterprise app stores. The idea behind MAM is to focus enterprise resources on managing what's really important to the business-its data-by taking charge of the apps that can access that data while leaving employees in control of the devices they own. MAM allows organisations to mandate encryption, set and enforce role-based policies for applications including how they store and share documents and even remove data and deprovision apps when an employee leaves the company (or loses a device). In other words, you can ensure that sensitive data never leaves your customer relationship management app without preventing salespeople from playing Angry Birds on their own devices during their own time.
"I'm not going to access proprietary data by opening Angry Birds," says Brian Duckering, senior manager of Enterprise Mobility at Symantec, which has also adopted the MAM approach. "So do I need to manage Angry Birds? Probably not."
"We've always believed that ultimately security and compliance boils down to being able to control the data," adds Herrema. "Trying to control the device, in a lot of cases, is neither necessary nor sufficient. A lot of the typical device management methods don't work anymore in a BYOD world. You can't tell a BYOD user who owns an iPhone 4S that they can't use Siri or iCloud or that they can't use the App Store. At the end of the day, if you have control of your own data and make sure that your data isn't leaking off into personal applications and services, you don't have to touch the rest of the device. I don't have to tell the user that you can't use Dropbox. I just have to make sure that none of my sensitive corporate documents wind up in Dropbox."
"In many cases, you actually have great control over protecting that data than you would with a general MDM solution," Symantec's Duckering notes.
It should be noted that even when you manage applications rather than devices, special care is necessary for certain high-risk application types. For instance, in addition to providing the ability to manage internally developed apps and third party apps, Good also provides its own secure email app and secure browser app.
"The reason we have a secure email app and a secure browser app is that the native apps on these devices are inherently leaky," says Good's Herrema. "If you can't actually secure and manage the core browser and the core address book and core email app, you're still going to have data loss."
Instead of MAM, Red Bend Software takes an alternative approach that is more reminiscent of MDM. It uses type 1 hypervisors on particular Android handsets to create what is essentially two virtual phones running simultaneously on the same physical hardware. One phone is the standard consumer device for use with Facebook and Twitter and other consumer-facing applications. The other is a phone running a dedicated Android operating system geared for the enterprise.
"We allow the enterprise to completely manage that part of the phone," says Morten Grauballe, executive vice president of Corporate Development and Strategy at Red Bend.
Grauballe explains that by leveraging a type 1 hypervisor, Red Bend is able to achieve excellent performance because it runs directly on the phone's hardware (as opposed to a type 2 hypervisor, which runs as a software layer above a device's operating system). And, he adds, Red Bend achieves significantly better security because it doesn't run inside the same OS as the other consumer-facing applications.
"The usability goes both ways," he says. "It gives the IT organisation better control, but gives the user the privacy and freedom they would like."
One drawback of Red Bend's type 1 hypervisor approach is that it can't be implemented on just any smartphone. It requires the handset manufacturer or chipset manufacturer to architect the device to support bare metal virtualisation. Red Bend is attacking that problem aggressively.
"We're working with our customers, who are all the mobile device manufacturers-chipset manufacturers to ODMs and OEMs-to actually change the architecture and how the next generation of mass-market devices are designed and built so they are enterprise ready from the beginning," explains Lori Sylvia, executive vice president of Marketing at Red Bend.
Red Bend is not alone. Virtualisation juggernaut VMware has launched a similar project, called Horizon Mobile Virtualisation, to allow the enterprise to deploy its own secure virtual phone images to employee-owned smartphones.
Desktop as a Service (DaaS) specialist Desktone is also using virtualisation to solve the BYOD puzzle, but with an approach that differs from Red Bend's. Rather than virtualising the phone, Desktone is virtualising users' desktop computers and deliver them as a service, giving them the ability to access that virtual desktop via different devices, from a physical desktop or laptop to a tablet or smartphone.
"Rather than managing devices, it's more about managing users," says Danny Allan, CTO of Desktone and former directory of security research for IBM.
Desktone's solution allows organisations to set policies for how services can be accessed and with which devices. For instance, it could allow a user to access a certain service from an iPad while on the road but not while in the office, or vice versa.
In the end, whichever strategy you adopt for dealing with BYOD, the vendors all agree that the key is to secure your sensitive data while still providing the end user the freedom and flexibility to use devices to enhance their productivity. If your solution is too onerous to use, end users won't use your apps and you'll fail to recognise the productivity gains mobile computing offers.
"If the solution that you apply is too restrictive, then as much as everyone wants BYOD, it's simply not going to be a practical solution because no one will use it," Duckering says. |
3D vertebrae labeling in spine CT: an accurate, memory-efficient (Ortho2D) framework Purpose. Accurate localization and labeling of vertebrae in computed tomography (CT) is an important step toward more quantitative, automated diagnostic analysis and surgical planning. In this paper, we present a framework (called Ortho2D) for vertebral labeling in CT in a manner that is accurate and memory-efficient. Methods. Ortho2D uses two independent faster R-convolutional neural network networks to detect and classify vertebrae in orthogonal (sagittal and coronal) CT slices. The 2D detections are clustered in 3D to localize vertebrae centroids in the volumetric CT and classify the region (cervical, thoracic, lumbar, or sacral) and vertebral level. A post-process sorting method incorporates the confidence in network output to refine classifications and reduce outliers. Ortho2D was evaluated on a publicly available dataset containing 302 normal and pathological spine CT images with and without surgical instrumentation. Labeling accuracy and memory requirements were assessed in comparison to other recently reported methods. The memory efficiency of Ortho2D permitted extension to high-resolution CT to investigate the potential for further boosts to labeling performance. Results. Ortho2D achieved overall vertebrae detection accuracy of 97.1%, region identification accuracy of 94.3%, and individual vertebral level identification accuracy of 91.0%. The framework achieved 95.8% and 83.6% level identification accuracy in images without and with surgical instrumentation, respectively. Ortho2D met or exceeded the performance of previously reported 2D and 3D labeling methods and reduced memory consumption by a factor of ∼50 (at 1 mm voxel size) compared to a 3D U-Net, allowing extension to higher resolution datasets than normally afforded. The accuracy of level identification increased from 80.1% (for standard/low resolution CT) to 95.1% (for high-resolution CT). Conclusions. The Ortho2D method achieved vertebrae labeling performance that is comparable to other recently reported methods with significant reduction in memory consumption, permitting further performance boosts via application to high-resolution CT. |
How and when to investigate syncope. Recurrent syncope is a distressing phenomenon and may be very difficult to treat where the cause remains obscure. A methodical approach to investigation will often be fruitful when proper emphasis is given to thorough clinical assessment and simple investigations. Where these are normal, head-up tilt testing will often diagnose 'malignant vasovagal syndrome'. |
/*FUNCTION**********************************************************************
*
* Function Name : FTM_DRV_UpdatePwmPeriodDither
* Description : This function will use in the PWM period dithering. This value
* is added to an internal accumulator at the end of each PWM period. The value is
* updated with its write buffer value according to the register synchronization.
*
* Implements : FTM_DRV_UpdatePwmPeriodDither_Activity
*END**************************************************************************/
status_t FTM_DRV_UpdatePwmPeriodDither(uint32_t instance,
uint8_t newModFracVal,
bool softwareTrigger)
{
DEV_ASSERT((instance == 1U) || (instance == 2U));
DEV_ASSERT(newModFracVal <= 32U);
FTM_Type * ftmBase = g_ftmBase[instance];
FTM_DRV_SetModFracVal(ftmBase, newModFracVal);
if (softwareTrigger)
{
FTM_DRV_SetSoftwareTriggerCmd(ftmBase, true);
}
return STATUS_SUCCESS;
} |
/*
* Copyright 2013 <NAME>
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package org.antkar.syn.internal.parser;
import java.io.PrintStream;
import org.antkar.syn.internal.Checks;
import org.antkar.syn.internal.CommonUtil;
import org.antkar.syn.internal.lrtables.ParserProduction;
import org.antkar.syn.internal.lrtables.ParserState;
/**
* Nonterminal stack element. Contains sub-elements.
*/
final class NonterminalParserStackElement extends ParserStackElement {
private final ParserProduction production;
/** The top of the stack of sub-elements. */
private final ParserStackElement subElements;
NonterminalParserStackElement(
ParserStackElement prev,
ParserState state,
ParserProduction production,
ParserStackElement subElements)
{
super(prev, state);
this.production = Checks.notNull(production);
this.subElements = Checks.notNull(subElements);
}
@Override
IParserNode createParserNode() {
IParserAction action = production.getAction();
IParserNode result = action.execute(subElements);
return result;
}
@Override
AmbiguityNode createAmbiguityNode() {
int length = production.getLength();
if (length == 0) {
return AmbiguityNode.NULL;
}
AmbiguityNode[] subAmbiguityNodes = new AmbiguityNode[length];
ParserStackElement element = subElements;
for (int i = 0; i < subAmbiguityNodes.length; ++i) {
subAmbiguityNodes[i] = element.createAmbiguityNode();
element = element.getPrev();
}
return new AmbiguityNode(subAmbiguityNodes);
}
@Override
void print(PrintStream out, int level) {
CommonUtil.printIndent(out, level);
out.println(production.getNonterminal().getName());
ParserStackElement[] array = getSubElementsArray();
for (ParserStackElement element : array) {
element.print(out, level + 1);
}
}
/**
* Creates an array of sub-elements.
*/
private ParserStackElement[] getSubElementsArray() {
int length = production.getLength();
if (length == 0) {
return EMPTY_ARRAY;
}
ParserStackElement[] result = new ParserStackElement[length];
ParserStackElement element = subElements;
for (int i = length - 1; i >= 0; --i) {
result[i] = element;
element = element.getPrev();
}
return result;
}
}
|
package com.example.jianfeng.cmsbusiness_android.loginInfo;
import android.content.Context;
import android.database.Cursor;
import android.database.sqlite.SQLiteDatabase;
import android.database.sqlite.SQLiteOpenHelper;
import android.widget.Toast;
import com.example.jianfeng.cmsbusiness_android.contacts.CMSContactsVO;
import java.io.ByteArrayInputStream;
import java.io.ByteArrayOutputStream;
import java.io.ObjectInputStream;
import java.io.ObjectOutputStream;
import java.util.ArrayList;
import java.util.List;
/**
* Created by jianfeng on 19/1/13.
*/
public class CMSSQLManager extends SQLiteOpenHelper {
private static String name = "CMSBusinessTest2.db"; //库名
private static String ContactsListTabName = "CMSContactsList2"; //联系人表名
private static int version = 1;
private Context myContent;
private static CMSSQLManager instance = null;
/**
* integer：整形
* real：浮点型
* text：文本类型
* blob：二进制类型
* PRIMARY KEY将id列设置为主键
* AutoIncrement关键字表示id列是自动增长的
*/
//db.execSQL("create table info (_id integer primary key autoincrement,name varchar(20))");
//private static final String CREATE_BOOK = "CREATE TABLE book ("
// + "id integer PRIMARY KEY Autoincrement ,"
// + "author text ,"
// + "price real ,"
// + "pages integer,"
// + "name text )";
private String ContactTabCreateSQL = "create table if not exists " + ContactsListTabName + " (ContactId integer primary key autoincrement, UserID text, Contacts blob)";
/** --------------------单列的创建----------------- */
public static CMSSQLManager shared(Context context) {
if (instance == null) {
instance = new CMSSQLManager(context);
}
return instance;
}
private CMSSQLManager(Context context) {
/** 建库 */
super(context, name, null, version);
myContent = context;
}
@Override
public void onCreate(SQLiteDatabase db) {
/** 建表 not create and not execute if system has tab */
db.execSQL(ContactTabCreateSQL);
Toast.makeText(myContent, "数据库和表创建成功", Toast.LENGTH_SHORT).show();
}
/* onUpgrade:数据库版本发生变化时调用该方法,该方法特别适合做表结构的修改
* public void onUpgrade(SQLiteDatabase db, int oldVersion, int newVersion) {
* 执行sql语句做表结构的修改 数据库版本也需要同时变更 才会起作用
* db.execSQL("alter table user add phone varchar(20)");
* System.out.println("旧版本:" + oldVersion + "新版本+" + newVersion);
*/
@Override
public void onUpgrade(SQLiteDatabase db, int oldVersion, int newVersion) {
onCreate(db);
}
/**
* 更新
* @param contacts
*/
public void updateContacts(List<CMSContactsVO> contacts) {
SQLiteDatabase db = getWritableDatabase();
try {
for(CMSContactsVO VO:contacts) {
String userID = VO.UserID;
ByteArrayOutputStream arrayOutputStream = new ByteArrayOutputStream();
ObjectOutputStream objectOutputStream = new ObjectOutputStream(arrayOutputStream);
objectOutputStream.writeObject(VO);
objectOutputStream.flush();
byte data[] = arrayOutputStream.toByteArray();
objectOutputStream.close();
arrayOutputStream.close();
Object[] obj = new Object[] { userID, data };
String sql = "UPDATE " + ContactsListTabName + " SET UserID = ?, Contacts = ? WHERE id IN (SELECT id FROM student ORDER BY id ASC LIMIT 0,1000";
db.execSQL(sql,obj);
}
} catch(Exception e) {
e.printStackTrace();
}
db.close(); // 关闭数据库
}
/**
* 保存 ， 插入
* @param contacts
*/
public void saveContacts(List<CMSContactsVO> contacts) {
try {
SQLiteDatabase database = getWritableDatabase();
for(CMSContactsVO VO:contacts){
String userID = VO.UserID;
ByteArrayOutputStream arrayOutputStream = new ByteArrayOutputStream();
ObjectOutputStream objectOutputStream = new ObjectOutputStream(arrayOutputStream);
objectOutputStream.writeObject(VO);
objectOutputStream.flush();
byte data[] = arrayOutputStream.toByteArray();
objectOutputStream.close();
arrayOutputStream.close();
Object[] obj = new Object[] { userID, data };
String sql = "insert into " + ContactsListTabName + "(UserID, Contacts) values(?,?)";
database.execSQL(sql, obj);
}
database.close();
} catch (Exception e) {
e.printStackTrace();
}
}
/**
* 数据库查询所以联系人
* @param
*/
public List<CMSContactsVO> loadContacts() {
SQLiteDatabase database = getReadableDatabase();
ArrayList<CMSContactsVO> list = new ArrayList<CMSContactsVO>();
Cursor cursor = database.rawQuery("select * from " + ContactsListTabName, null);
if (cursor!=null && cursor.getCount()>0) {
try {
while (cursor.moveToNext()) {
String userId = cursor.getString(cursor.getColumnIndex("UserID"));
byte[] contactsByte = cursor.getBlob(cursor.getColumnIndex("Contacts"));
ByteArrayInputStream byteArrayInputStream = new ByteArrayInputStream(contactsByte); // 创建ByteArrayInputStream对象
ObjectInputStream objectInputStream = new ObjectInputStream(byteArrayInputStream); // 创建ObjectInputStream对象
CMSContactsVO object = (CMSContactsVO) objectInputStream.readObject(); // 从objectInputStream流中读取一个对象
byteArrayInputStream.close(); // 关闭输入流
objectInputStream.close(); // 关闭输入流
list.add(object);
}
cursor.close();
} catch (Exception e) {
e.printStackTrace();
}
database.close();
}
return list;
}
}
|
<reponame>TrainingITCourses/aplicaciones-pwa-ManuelGO<filename>speed/src/app/list-result/list-result.component.ts
import { Component, OnInit, Input, ChangeDetectionStrategy } from '@angular/core';
@Component({
changeDetection: ChangeDetectionStrategy.OnPush,
selector: 'app-list-result',
templateUrl: './list-result.component.html',
styleUrls: ['./list-result.component.css']
})
export class ListResultComponent implements OnInit {
@Input() public list;
constructor() {
}
ngOnInit() {
}
}
|
Critical geragogy and foreign language learning: An exploratory application ABSTRACT This article proposes an exploratory application of the principles of critical geragogy (Formosa, 2002, 2011, 2012) to foreign language (FL) education (i.e., L2 learning in the L1 community). Critical geragogy is an educational, practical framework intended to empower older adults and lead them to emancipate from age strictures (Glendenning & Battersby, 1990, as cited in Formosa, 2012, p. 74). These strictures permeate the FL classroom and cause many older learners to adopt self-defeating attitudes and beliefs toward their abilities to learn an FL (Andrew, 2012; Ramrez Gmez, 2014). The limited research on FL learning in older adults (60 years old and over) either confirming, exploring, or rejecting any claimsuggests that these beliefs and attitudes are not supported by rigorous empirical evidence. Rather, they are mostly based on generally held preconceptions and impressionistic data (Ramrez Gmez, 2014, 2015). Drawing from these notions and qualitative research from various sources, I propose an exploratory set of principles for what I have called critical foreign language geragogy. In this article, I argue that a critical view of social preconceptions of aging is fundamental to transform older learners attitudes toward their own FL learning potential and render them more realistic and constructive. I believe that such an approach to FL geragogy may contribute to improving older learners' expectations, goal setting, and ultimate accomplishment. |
//
// ========================================================================
// Copyright (c) 1995-2014 Mort Bay Consulting Pty. Ltd.
// ------------------------------------------------------------------------
// All rights reserved. This program and the accompanying materials
// are made available under the terms of the Eclipse Public License v1.0
// and Apache License v2.0 which accompanies this distribution.
//
// The Eclipse Public License is available at
// http://www.eclipse.org/legal/epl-v10.html
//
// The Apache License v2.0 is available at
// http://www.opensource.org/licenses/apache2.0.php
//
// You may elect to redistribute this code under either of these licenses.
// ========================================================================
//
package org.eclipse.jetty.spdy.server.http;
import org.eclipse.jetty.spdy.http.HTTPSPDYHeader;
import org.eclipse.jetty.util.Fields;
import org.eclipse.jetty.util.ssl.SslContextFactory;
public class SPDYTestUtils
{
public static Fields createHeaders(String host, int port, short version, String httpMethod, String path)
{
Fields headers = new Fields();
headers.put(HTTPSPDYHeader.METHOD.name(version), httpMethod);
headers.put(HTTPSPDYHeader.URI.name(version), path);
headers.put(HTTPSPDYHeader.VERSION.name(version), "HTTP/1.1");
headers.put(HTTPSPDYHeader.SCHEME.name(version), "http");
headers.put(HTTPSPDYHeader.HOST.name(version), host + ":" + port);
return headers;
}
public static SslContextFactory newSslContextFactory()
{
SslContextFactory sslContextFactory = new SslContextFactory();
sslContextFactory.setEndpointIdentificationAlgorithm("");
sslContextFactory.setKeyStorePath("src/test/resources/keystore.jks");
sslContextFactory.setKeyStorePassword("<PASSWORD>");
sslContextFactory.setTrustStorePath("src/test/resources/truststore.jks");
sslContextFactory.setTrustStorePassword("<PASSWORD>");
sslContextFactory.setProtocol("TLSv1");
sslContextFactory.setIncludeProtocols("TLSv1");
return sslContextFactory;
}
}
|
#include<iostream>
using namespace std;
int main()
{
char i[81];
for (int a = 0; a<80; a++)
cin >> i[a];
char a[11][11];
for (int i = 0; i<10; i++)
for (int j = 0; j<10; j++)
cin >> a[i][j];
for (int j = 0; j<8; j++)
for (int m = 0; m<10; m++)
{
int x = 0;
for (int l = 0; l<10; l++)
if (i[l + j * 10] == a[m][l])
x++;
if (x == 10)
cout << m;
}
} |
What they think of us: A study of teaching medical specialists attitude towards psychiatry in India Context: Attitudes of teaching medical specialists are important in shaping medical students' attitudes toward psychiatry. Data on attitudes of teaching medical specialists of India toward psychiatry are limited. Aims: The aim was to study the attitude of teaching medical specialists of an academic medical center in East India toward psychiatry. Settings and Design: This was a cross-sectional descriptive study. Materials and Methods: We administered attitude toward psychiatry-30 (ATP 30) scale to teaching medical specialists of the All India Institute of Medical Sciences, Bhubaneswar, based on convenience sampling. Of 104 specialists contacted, 88 returned the completed questionnaire. Statistical Analysis: We carried out descriptive statistical analysis and expressed results in mean and standard deviation. We analyzed the association of demographic characteristics, specialization, and duration of professional experience with total ATP scores using Chi-square test. We used subgroup analysis to compare mean ATP scores in different demographic and professional groups. We used independent t-test and ANOVA for between group comparisons. Results: The response rate was 84.62% with a mean ATP score of 88.60. Female gender and having a family member with mental illness was significantly associated with favorable ATP. Notable findings were that 97% of participants were favorable toward patients with psychiatric illness, 90% felt psychiatric interventions as effective whereas 87% found psychiatry unappealing and 52% said that they would not have liked to be a psychiatrist. Conclusions: While favorable attitudes toward patients with psychiatric illness and psychiatric interventions may mean better patient care; unfavorable attitudes toward psychiatry as a career choice may adversely affect postgraduate recruitment rates. |
<filename>ph_split_train_val.py
import pickle
import numpy as np
from ssd_data import BaseGTUtility
from ph_gt_data import GTUtility
def random_split(self, split=0.8):
gtu1 = BaseGTUtility()
gtu1.gt_path = self.gt_path
gtu1.image_path = self.image_path
gtu1.classes = self.classes
gtu2 = BaseGTUtility()
gtu2.gt_path = self.gt_path
gtu2.image_path = self.image_path
gtu2.classes = self.classes
n = int(round(split * len(self.image_names)))
idx = np.arange(len(self.image_names))
np.random.seed(0)
np.random.shuffle(idx)
train = idx[:n]
val = idx[n:]
gtu1.image_names = [self.image_names[t] for t in train]
gtu2.image_names = [self.image_names[v] for v in val]
gtu1.data = [self.data[t] for t in train]
gtu2.data = [self.data[v] for v in val]
if hasattr(self, 'text'):
gtu1.text = [self.text[t] for t in train]
gtu2.text = [self.text[v] for v in val]
gtu1.init()
gtu2.init()
return gtu1, gtu2
PICKLE_DIR = './pickles/'
# AI-HUB
# PICKLE = 'printed_hangul_word.pkl'
# TRAIN = 'printed_hangul_word_train.pkl'
# VALIDATION = 'printed_hangul_word_val.pkl'
# AIG-IDR
PICKLE = 'hospital_receipt_60000.pkl'
TRAIN = 'hospital_receipt_60000_train.pkl'
VALIDATION = 'hospital_receipt_60000_val.pkl'
with open(PICKLE_DIR + PICKLE, 'rb') as f:
gt_util_cracker = pickle.load(f)
gt_util_train, gt_util_val = random_split(gt_util_cracker)
print(' # Train pkl file saves to %s ...' % TRAIN)
pickle.dump(gt_util_train, open(PICKLE_DIR + TRAIN, 'wb'))
print(' # Done')
print(' # Validation pkl file saves to %s ...' % VALIDATION)
pickle.dump(gt_util_val, open(PICKLE_DIR + VALIDATION, 'wb'))
print(' # Done')
print(len(gt_util_train.image_names))
print(len(gt_util_val.image_names))
|
class possum:
"""
Class for creating spectra
"""
def __init__(self):
self.__c = 2.99e+08 # speed of light in m/s
self.__mhz = 1.0e+06
def _createPOSSUMspectrum(self):
band1 = self._createFrequency(700.,1000.,nchan=300)
band2 = self._createFrequency(1000.,1300.,nchan=300)
band3 = self._createFrequency(1500.,1800.,nchan=300)
self.nu = np.concatenate((band1,band2[1:],band3))
def _createFrequency(self, numin=700., numax=1800., nchan=100.):
"""
Creates an array of evenly spaced frequencies
numin and numax are in [MHz]
To call:
_createFrequency(numin, numax, nchan)
Parameters:
numin
numax
Postcondition:
"""
# ======================================
# Convert MHz to Hz
# ======================================
numax = numax * self.__mhz
numin = numin * self.__mhz
# ======================================
# Generate an evenly spaced grid
# of frequencies and store the array
# ======================================
return np.arange(nchan)*(numax-numin)/(nchan-1) + numin
def _createNspec(self, flux, depth, chi):
"""
Function for generating N faraday spectra
and merging
To call:
createNspec(flux, depth, chi)
Parameters:
flux [float, array]
depth
chi [float, array]
"""
nu = np.asmatrix(self.nu)
flux = np.asmatrix(flux).T
chi = np.asmatrix(chi).T
depth = np.asmatrix(depth).T
P = flux.T * np.exp(2*1j * (chi + depth * np.square(self.__c / nu)))
P /= P.max()
self.Polarization_ = np.ravel(P)
def _addNoise(self, sigma):
pass |
Dabigatran in the treatment and secondary prophylaxis of venous thromboembolism in children with thrombophilia Key Points The efficacy and safety of dabigatran for acute VTE demonstrated noninferiority to standard of care in children with thrombophilia. Dabigatran demonstrated a favorable safety profile in secondary prevention of VTE in children with thrombophilia. Dabigatran demonstrated a favorable safety profile in secondary prevention of VTE in children with thrombophilia. In the phase 2b/3 DIVERSITY trial, 3 suggest dabigatran could be an appropriate long-term anticoagulant for children with thrombophilia. These trials were registered at www.clinicaltrials.gov as #NCT01895777 and #NCT02197416. Introduction In children, venous thromboembolism (VTE; including deep vein thrombosis and pulmonary embolism ) is a severe multifactorial disease. Common clinical risk factors include use of central venous catheters (CVCs), underlying disease, and thrombophilia. For example, a significant association between inherited thrombophilic disorders and VTE onset, as well as recurrence, has been shown in a meta-analysis of observational studies in children 6 ; therefore, screening for inherited thrombophilia defects, which typically include factor V Leiden (FVL), prothrombin (PT) mutations, and deficiencies of antithrombin, protein C, or protein S, could be of clinical value. 5,7 Longer term complications of VTE include recurrence, postthrombotic syndrome (PTS), chronic thromboembolic pulmonary hypertension, and death. 4, The VTE recurrence rate has been reported to be~3% in newborns and 8% in older children, 6,11 with risk increasing to as high as 29% in children with certain inherited thrombophilia traits. 12 VTE-related death in children has been reported to be 0% to 3.7% and PTS has been reported with a frequency of 9.5% to 70%. Anticoagulation is the standard treatment for VTE. 16 The exact duration for optimal anticoagulation therapy has yet to be established in children with acute VTE. Recent evidence-based recommendations suggest longer duration for unprovoked thromboembolic events (6-12 months), regardless of inherited thrombophilia markers, than provoked events (3 months). 16,17 Continuation of treatment is dependent on the benefits of maintaining a reduced risk of VTE recurrence vs the risk of bleeding. 18,19 Standard of care (SOC) anticoagulation is typically low-molecularweight heparin (LMWH) or vitamin K antagonists (VKAs) in pediatric patients with symptomatic VTE. 16 Direct oral anticoagulants (DOACs) have shown superiority or noninferiority to SOC in lowering the prospect of thromboembolic complications, with comparable or diminished bleeding risk, and, therefore, current adult-based recommendations favor their use in the treatment of patients with proximal DVT and nonmassive PE, except for patients with antiphospholipid syndrome. 20 Currently, there is no preferred anticoagulation agent recommended for long-term use, particularly in thrombophilia subgroups; however, use of DOACs might be advantageous because of the requirement for less clinical monitoring and follow-up than SOC treatments, and also, fewer food and drug interactions. 20 The potential benefits of DOAC use, including stability and lower residual thrombus burden, along with a lower bleeding risk, have yet to be fully investigated in children with unprovoked VTE. Two large international, multicenter, pediatric phase 2b/3 studies have examined the efficacy and safety of the DOAC dabigatran etexilate in the treatment of acute VTE and in long-term secondary VTE prevention in children. 21,22 The acute VTE treatment (DIVERSITY, #NCT01895777) open-label, randomized, phase 2b/3 study demonstrated the noninferiority of dabigatran compared with SOC in children from birth to the age of <18 years. 21 The secondary VTE prevention (#NCT02197416) phase 3, single-arm, cohort study showed a favorable safety profile for dabigatran in secondary VTE prevention in children with persistent VTE risk factor(s) from birth to the age of <18 years. 22 Based on the results of these 2 pivotal, global studies, dabigatran etexilate was approved by the European Medicines Agency and US Food and Drug Administration for pediatric indications in January and June 2021, respectively. 23,24 We present herein, a subgroup analysis from these studies of the efficacy and safety of dabigatran in children with thrombophilia vs those with negative/unknown thrombophilia status. Trial designs The findings presented here are from subgroup analyses of children with thrombophilia vs those with negative/unknown thrombophilia status from the acute VTE treatment (DIVERSITY) and secondary VTE prevention studies. Both study designs have been described previously and key exclusion criteria were: "conditions associated with increased bleeding risk, renal dysfunction, hepatic disease, active infective endocarditis, heart valve prosthesis requiring anticoagulation, and children aged 0 to <2 years with gestational age at birth <37 weeks or with bodyweight lower than the third percentile (according to World Health Organization standards)." 21,22,25,26 The studies were conducted in accordance with the Declaration of Helsinki and the principles of Good Clinical Practice and were approved by all investigational site ethics committees. Written informed consent was obtained before participation, according to the International Conference on Harmonisation Good Clinical Practice, and the regulatory and legal requirements of each participating country. Both studies were sponsored by Boehringer Ingelheim. As listed, both studies were registered at www.clinicaltrials.gov. Trial populations For the acute VTE treatment study (DIVERSITY), patients from birth to the age of <18 years with a diagnosis of VTE (ie, DVT, PE, or cerebral venous sinus thrombosis) objectively confirmed by compression ultrasound, computed tomography, or magnetic resonance imaging, and who were initially treated for 5 to 21 days with SOC (eg, unfractionated heparin or LMWH), with parenteral anticoagulation therapy expected to last for ≥3 months, were eligible for inclusion. 21,25 For the secondary VTE prevention study, patients aged 3 months to <18 years with an objectively confirmed diagnosis of VTE, who had been treated for acute VTE with SOC for ≥3 months, or who had completed dabigatran or SOC treatment in the acute VTE treatment study and had an unresolved clinical thrombosis risk factor requiring further anticoagulation for secondary prevention of VTE, were eligible for inclusion. 22,26 In the current analysis, patients were assessed for thrombophilia status and were classified as "thrombophilia positive" or "thrombophilia status negative/unknown." Randomization and treatments In the acute VTE treatment study (DIVERSITY), patients were randomized (2:1) to receive open-label dabigatran or SOC, treated for 3 months from randomization, and followed up for an additional month. 21,25 In the secondary VTE prevention study, patients were treated with dabigatran for up to 12 months. 22,26 Dabigatran was dosed according to an age-and weight-adjusted nomogram derived from estimated renal function to achieve exposure comparable to that of adult populations treated with dabigatran. Outcomes Primary and secondary end points differed according to the individual study. In the acute VTE treatment study, the primary efficacy end point was a composite (centrally adjudicated by an independent, blinded committee) of the proportion of children with complete thrombus resolution, freedom from recurrent VTE (including symptomatic and asymptomatic, contiguous progression or noncontiguous new thrombus, DVT, PE, paradoxical embolism, and thrombus progression), and freedom from VTE-related death. 21,25 Secondary/ other end points included residual thrombotic burden by the end of study treatment (defined as complete or partial thrombus resolution, stabilization of thrombus, or thrombus progression); incidence of bleeding events; major bleeding events (MBEs; defined as fatal bleeding, clinically overt bleeding , retroperitoneal, pulmonary, intracranial, central nervous system bleeding; bleeding requiring surgical intervention) clinically relevant nonmajor (CRNM) bleeding events (defined as overt bleeding for which a blood product was administered and that was not directly attributable to the patient's underlying medical condition, or bleeding that required medical or surgical intervention to restore hemostasis, other than in an operating suite); 28 and other safety/ tolerability/adherence outcomes. 21,25 In the secondary VTE prevention study, all outcomes were considered safety related. 22 Primary end points included recurrence of VTE, mortality, MBEs, CRNM bleeding events, and minor bleeding events; assessed at 3, 6, and 12 months after enrollment. 22 Secondary end points included the occurrence of newly diagnosed or worsening of baseline PTS (per the modified Villalta scale) 29,30 at 3, 6, and 12 months after enrollment. PTS was assessed at 3 months to capture any patients enrolled with a history of VTE who might subsequently present with PTS. At each on-treatment study visit in both studies, adherence was calculated as adherence (%) = (actual number of dabigatran doses taken since last count planned number of dabigatran doses that should have been taken in the same period) 100. The overall average adherence was calculated as the average of adherence at each visit. Patients with thrombophilia subgroup Thrombophilia screening was not a requirement of the clinical trials, therefore, patients' requirement for thrombophilia screening was determined by the attending physician at each site. The study databases were reviewed, and patient thrombophilia status was determined and classified as "thrombophilia positive" or "thrombophilia status negative/unknown"; a further subset of patients with confirmed inherited thrombophilia 32 (those with only FVL and/or PT mutation) was also established. For the thrombophiliapositive subgroup, thrombophilia was additionally categorized as "major" or "minor" based on predefined criteria. Major thrombophilia included patients with homozygous FVL, homozygous PT mutation, compound heterozygous FVL and PT mutations, antithrombin deficiency, protein C deficiency, protein S deficiency, or those who were antiphospholipid antibody (APLA)-and/or lupus anticoagulant (LA)-positive, as well as specified combinations of these or other thrombophilic conditions or mutations (listed in Table 3 footnotes). 31 Minor thrombophilia included coagulation disorders not defined as major thrombophilia: heterozygous FVL, heterozygous PT gene mutation (including those whose zygosity was recorded as unknown), specified combinations of thrombophilic conditions, and specified other mutations and conditions with uncertain thrombophilic significance ( Table 3 footnotes). 31 Statistical analysis For both studies, patient demographics, medical history, and baseline clinical characteristics were analyzed descriptively. Timeto-event analyses were summarized as Kaplan-Meier estimates; other end points, including safety and adverse events (AEs) were analyzed descriptively. In the acute VTE treatment study, noninferiority testing for the composite primary end point was performed on Mantel-Haenszel-weighted rate difference with a margin of 20% used for a 2-sided test at significance level of.05. 33 Study populations The acute VTE treatment study (DIVERSITY) took place across 65 centers in 26 countries. The analysis population comprised 267 patients; 62 patients with documented thrombophilia (32 with confirmed inherited thrombophilia ) and 205 patients with thrombophilia-negative/unknown status ( Figure 1). Within the thrombophilia subgroup, 37 patients were categorized as having major thrombophilia and 25 with minor thrombophilia. A sensitivity analysis of the thrombophilia-negative (n = 145) and thrombophilia-unknown (ie, not tested; n = 60) subgroups found both groups to be largely similar in terms of safety and efficacy outcomes (supplemental Table 1). Therefore, these subgroups were merged into 1 "thrombophilia-negative/unknown" group. The secondary VTE prevention study took place across 60 sites in 22 countries. The analysis population included 213 patients; 106 with documented thrombophilia (44 with confirmed inherited thrombophilia ) and 107 with thrombophilia-negative/unknown status ( Figure 1). Within the thrombophilia subgroup, 73 patients were categorized as having major thrombophilia and 33 with minor thrombophilia. The electronic case report form captured data on thrombophilia and thrombophilia types, therefore, all patients for whom thrombophilia status was not captured were automatically considered "thrombophilia negative/unknown." Of the patients included in the analyses, 91 were previously enrolled in the acute VTE treatment study (47 with thrombophilia and 44 as thrombophilia negative/ unknown); supplemental Tables 2 and 3 detail the 35 patients with thrombophilia who rolled over from the DIVERSITY study to the secondary VTE prevention study. "Other conditions requiring secondary VTE prophylaxis" includes a variety of medical conditions with unclear thrombophilic significance considered to be conditions requiring secondary VTE prophylaxis in accordance with local VTE treatment/prophylactic protocols and comprise residual/unresolved DVT or sinus vein thrombosis, recurrent provoked VTE, strong family history of PE, and implantable medical devices other than central venous or arterial catheters (eg, ports, esndocardial electrodes). Most of these are conditions requiring secondary VTE prophylaxis only in combination with categories or VTE risk factors listed in other rubrics of the table, that is, medical circumstances that increase risk of thrombosis or thrombophilic conditions. Other mutations and conditions with indefinite thrombophilic significance: MTHFR, PAI-1, sustained elevated factor VIII level, factor XII deficiency, ACE mutation, 53 thrombospondin mutations, MTR mutation, 54 GPIA, GPIIIA, integrin A2 pathology, thrombocytopathy, FGB mutation, and fibrinolysis enzyme activity abnormality. Baseline demographics, clinical characteristics, and medical history Patient demographics and clinical characteristics at baseline in each study are described in Table 1 (by thrombophilia status) and supplemental Table 4 (by major or minor thrombophilia classification). In the acute VTE treatment and secondary VTE prevention studies,~23% and~50% of patients had thrombophilia diagnosed, respectively. In both studies, patients with documented thrombophilia were~2.5 years older, on an average, than patients with negative/unknown thrombophilia status. Patients with thrombophilia, and in particular, major thrombophilia, were predominantly male. In the secondary VTE prevention trial, patients with thrombophilia were more likely to have PE as their index VTE event than those with negative/unknown thrombophilia status. In the acute VTE treatment study, a higher proportion of patients had previous VTE in the thrombophilia group than in the thrombophilia-negative/unknown group but rates were similar between groups in the secondary prevention study ( Table 2). Tables 2 and 3, respectively (by thrombophilia status), and supplemental Table 5 (by major or minor thrombophilia classification). Presence of a CVC was recorded for a greater proportion of the thrombophilia-negative/unknown groups. In the secondary VTE prevention study, a greater proportion of patients with thrombophilia than those negative/unknown had other conditions requiring secondary VTE prophylaxis (including recurrent unprovoked VTE or structural venous abnormality). Medical history and details of thrombophilia conditions are shown in In the secondary VTE prevention study, 24 patients (23.1%) with thrombophilia had a history of PTS compared with 12 (11.3%) in the thrombophilia-negative/unknown group. Of the 73 patients with major thrombophilia and 33 patients with minor thrombophilia, 12 (16.9%) and 12 (36.4%) patients, respectively, had a history of PTS. Efficacy and safety Acute VTE treatment study (DIVERSITY) VTE OUTCOMES. By day 84 (or end of therapy ), few patients experienced progression of index thrombus. Thrombus progression was experienced by 8.1% of patients with documented thrombophilia (Table 4), mostly those with major thrombophilia (6.3% of patients with confirmed inherited thrombophilia ; Table 5; supplemental Table 6) and by 2.0% of patients with negative/unknown thrombophilia status ( Table 4). The VTE recurrence rate was 12.9% among patients with thrombophilia (mostly those with major thrombophilia; supplemental Table 6) and 2.9% for those with negative/unknown thrombophilia status (Table 4; Figure 2A). For the thrombophilia group, 5.1% of patients treated with dabigatran and 13.0% treated with SOC experienced thrombus progression, and 7.7% and 21.7%, respectively, experienced VTE recurrence (Table 4). For these patients, partial thrombus resolution was achieved in 43.6% and 34.8% of patients treated with dabigatran and SOC, respectively, and complete thrombus resolution in 35.9% and 21.7% of patients, respectively. For the high-risk group of patients with major thrombophilia, the frequency of thrombus progression was 10.0% for those treated with dabigatran and 11.8% for those with SOC, and the VTE recurrence rate was 15.0% and 23.5%, respectively (supplemental Table 6). The proportion of patients with VTE outcomes was similar between treatment groups within the negative/unknown thrombophilia group. PTS. In the acute VTE treatment study, within the treated set, PTS (newly identified or worsening from baseline) was reported as an AE in 2 out of 62 patients (3.2%) in the thrombophilia group (1 each with minor and major thrombophilia) and 5 out of 204 (2.4%) in the thrombophilia-negative/unknown group ( Figure 4A; by thrombophilia status), although the difference was not statistically significant (P =.66). For patients with confirmed inherited thrombophilia, 26.1% treated with dabigatran and 44.4% treated with SOC experienced bleeding events (Table 5). No patients with thrombophilia had an MBE over the course of the study. Among patients with negative/unknown thrombophilia status, bleeding events were similar between the SOC and dabigatran treatment groups (Table 4; Figure 4A). PTS. In the secondary VTE prevention study, newly identified or worsening of baseline PTS was reported as a study outcome at 12 months by 3 out of 106 patients (2.8%) with thrombophilia and 0 out of 107 patients without thrombophilia (Table 5). *Complete thrombus resolution, freedom from recurrent VTE, and freedom from VTE-related death. Defined as any symptomatic or asymptomatic contiguous progression of the index thrombus. PTS (newly identified or worsening from baseline) was not specified as an outcome in the acute treatment study but was reported as an AE within the treated set. None of the patients with thrombophilia who rolled over from the acute VTE treatment study (22 treated with dabigatran and 13 with SOC) and continued anticoagulation by dabigatran in the long-term secondary VTE prevention study experienced MBEs (supplemental Table 3). AEs In the acute VTE treatment study, rates of AEs, serious AEs, and AEs leading to treatment discontinuation were similar across subgroups (supplemental Table 7). The most common AEs included headache, nasopharyngitis, alopecia, and epistaxis. In the secondary VTE prevention study, rates of AEs, serious AEs, and AEs leading to treatment discontinuation were higher among patients with thrombophilia (supplemental Table 8). The most common AEs included headache, nasopharyngitis, dyspepsia, and upper respiratory tract infection. Adherence Adherence with study medication was routinely high in both studies. In the acute VTE treatment study, adherence exceeded 98% in all subgroups of thrombophilia status, and in the secondary VTE prevention study, adherence exceeded 96% in all subgroups. Discussion Two large studies, 1 phase 2b/3 and 1 phase 3, have demonstrated the effectiveness and safety of dabigatran for the treatment of acute VTE and for secondary prevention of VTE in children. 21,22 This analysis from the same studies examined patient characteristics and outcomes in the subgroup of children with thrombophilia. It should be noted that testing for thrombophilia was not a protocolspecified requirement for either study. However, for those children with known thrombophilia, and also for those newly tested for thrombophilia markers, the assignment for thrombophilia tests and the interpretation of results was the responsibility of the treating pediatrician, who, in most cases, specialized in pediatric hematology and, therefore, was privy to international definitions on inherited and acquired thrombophilia in children. Furthermore, collected and interpreted thrombophilia test results were verified with patients' source data during regular monitoring procedures; 100% source data verification should be conducted according to the monitoring plan of both trials. Therefore, we have a reasonable degree of certainty that the subgroup classified as thrombophilia positive has been correctly identified. Although the frequency of unfavorable VTE outcomes such as recurrence and progression of index thrombus were low in the acute VTE treatment study, the current analysis of data showed a *Complete thrombus resolution, freedom from recurrent VTE, and freedom from VTE-related death. Defined as any symptomatic or asymptomatic contiguous progression of the index thrombus. PTS (newly identified or worsening from baseline) was not specified as an outcome in the acute treatment study but was reported as an AE within the treated set. higher proportion of patients with thrombophilia compared with those with negative/unknown thrombophilia status experienced these outcomes. Consequently, complete thrombus resolution, freedom from recurrent VTE, and freedom from VTE-related death (primary end point) was achieved by a lower proportion of patients in the thrombophilia group (but similarly for major and minor subgroups) than those in the thrombophilia-negative/unknown status group. For the thrombophilia-negative/unknown status group, there was no obvious difference in VTE outcomes between dabigatran and SOC treatment groups. Of note, we observed that a numerically higher proportion of patients who were thrombophilia positive, and particularly those with confirmed inherited thrombophilia with FVL and/or PT mutations, experienced complete thrombus resolution, freedom from recurrent VTE, and freedom from VTE-related death with dabigatran vs SOC. Similarly, numerically lower rates of VTE recurrence and thrombus progression, and higher rates of partial thrombus resolution and complete thrombus resolution were observed in patients from the thrombophilia-positive group treated with dabigatran vs SOC. Although differences between treatment cohorts did not reach statistical significance, this is not unexpected because of the small sample sizes. Irrespective of thrombophilia status, consistency of effect with the overall study results, 21 in terms of noninferiority of dabigatran to SOC, was observed for the primary end point. Similar proportions of patients with and without thrombophilia experienced bleeding events. For the thrombophilia group, numerically fewer patients treated with dabigatran experienced bleeding events, including major and CRNM bleeding events, compared with those treated with SOC, although no statistical differences were observed between treatments. In the negative/unknown thrombophilia group, frequency of bleeding events was similar between treatment groups. In the analysis of data from the secondary VTE prevention study, higher rates of VTE recurrence and any bleeding events were observed among children with thrombophilia compared with children in whom thrombophilia status was negative/unknown. However, rates of major or CRNM bleeding were low, with no clear difference in rates according to thrombophilia status. The findings of greater VTE risk associated with presence of thrombophilia are broadly in line with previous observational studies of VTE recurrence in pediatric populations with so-called major thrombophilia, particularly in non-CVC-related events. Of note, only~15% (dabigatran arm) and~22% (SOC arm) of patients in the DIVERSITY study, and only~3% of patients enrolled in the secondary prophylaxis study, had central linerelated events. 21,22 For instance, in a previous German-wide national pediatric study, the presence of single vs combined prothrombotic defects was both associated with higher odds of VTE recurrence (single defect: odds ratio, −4.6; 95% CI, −2.3 to 9.0; P <.0001; combined defects: odds ratio, −24.0; 95% CI, −5.3 to 108.7; P <.0001). 5 39 for secondary VTE prevention) revealed no difference in symptomatic VTE recurrence/VTE-related deaths between patients with thrombophilia treated with dabigatran or warfarin, with a similar safety profile. 40 More recent adult reports, including a systematic review, confirmed a similar efficacy of DOACs, including dabigatran, as an alternative anticoagulation strategy for patients with thrombophilia, except for cases of heparin-induced thrombocytopenia, APLA (eg, triple positive), and paroxysmal nocturnal hemoglobinuria, not included in our report because of their rarity in children. To date, comparable evaluations of the performance of other DOACs in children with venous thrombotic events associated with thrombophilia are scant but are starting to emerge. A previous randomized trial on rivaroxaban in children, evaluating its efficacy and safety for acute VTE treatment, included 32 patients with inherited thrombophilia (intervention arm, 27 patients ; SOC, 5 patients ). However, no sensitivity analysis could be conducted because of its limited sample size. 44 If the signal for a differential effect of dabigatran vs SOC on residual thrombus burden translates into meaningful differences in outcomes, this would be expected to confer benefits in reducing VTE recurrences and PTS. In adults, persistence of thrombosis despite a course of anticoagulation is a predictor of VTE recurrence and PTS; 45,46 for example, a lack of thrombus resolution was associated with a statistically significant fourfold increase in the odds for PTS in the original pediatric cohort reporting the modified Villalta scale, 1 of the 2 pediatric PTS scales accepted by the International Society on Thrombosis and Haemostasis. 30,47 Nonetheless, a recent cross-sectional follow-up study of adults diagnosed with VTE randomized to receive either dabigatran or warfarin did not identify a difference in prevalence of PTS in either treatment group. 48 The generalizability of this finding to the pediatric population remains to be proven. Consistent with our results, pediatric studies have shown patients with CVC-related DVT to be younger than those with non-CVC-related DVT. 49 PTS predictors included increased residual DVT burden, which is relevant for our study findings. 50 For noncatheter-related DVT, like our findings, patients had thrombophilia more often, along with DVT recurrences. Furthermore, studies in adults have shown an increased risk of persistence of residual thrombus in patients with thrombophilia, highlighting the need to further investigate the role of DOACs in children with thrombophilia to prevent unfavorable VTE-related outcomes. 51 Considering the limited data available on the topic of DOACs in children with thrombophilia, we believe that the data summarized herein provides unique hypothesis-generating findings. We acknowledge that and captured on the medical history page of the electronic case report forms, but there were no requirements to perform any tests concerning thrombophilia (including repeat, genetic, and/or family investigations) after a patient had been enrolled. However, FVL/PT mutations were confirmed by genetic tests (source-verified data were entered into the electronic case report forms) and, therefore, we were able to identify patients with confirmed inherited thrombophilia. Rare coagulopathies or mutations and conditions with indefinite thrombophilic significance were entered in the case report form as free text. Concentrations of antithrombin, protein C/S, and APLA/LA were not captured in a standardized manner, nor were types of APLA (immunoglobulin, subclasses G or M), or whether triple-positive status was present. As mentioned, we derive confidence in the thrombophilia diagnoses because these were based on the judgment of the pediatrician, who, in most cases, specialized in pediatric hematology and was familiar with the concept of developmental hemostasis and age-specific levels of thrombophilic markers. A sample selection bias might be present because testing for thrombophilia status was not protocol mandated. To mitigate this issue, a sensitivity analysis comparing patients with negative vs unknown thrombophilia status did not identify differences on the patient characteristics of both groups. Another limitation is the small number of patients in some of the subgroups analyzed herein; hence, our findings should be considered exploratory and be interpreted with caution. Conclusion Given that thrombophilia is associated with a higher risk of long-term complications of VTE, it is important to establish appropriate and stable treatments for prolonged duration in this patient group. The exploratory findings of this study suggest that dabigatran could potentially have improved efficacy in these higher risk patients with thrombophilia compared with those without thrombophilia, who are at lower risk. Taken together, these data suggest that dabigatran could be appropriate for long-term anticoagulation in pediatric patients with thrombophilia. Future studies might add to the evidence on the treatment effects of DOACs, such as dabigatran vs SOC. |
<reponame>angular-package/type
export interface GuardAre { }
|
MEMPHIS — A teenage boy in Tennessee used hard work, planning and dedication to earn enough money to pay for college before he even starts.
“I just stay away from the negativity and try to always be positive,” Kevuntez King told WHBQ.
King said he grew up in a single-parent home with his mother, whose influence paved his way to success.
“She just taught me how to be independent like she had it, (and) she just wanted me to go get it myself,” he said.
From age 12 to 17, King sold newspapers with one goal: earn enough to pay for his entire college education. And he did just that.
“When it came down to school, my mom didn’t have to come out of pocket to do anything or I didn’t have to take out any loans to go to school,” he told WHBQ.
He earned $200 every Sunday for five years straight and was just accepted into Tennessee State University.
King has this piece of advice to share: “Make sure you surround yourself with people that’s trying to go up in life and not trying to bring you down. Just stay positive and always believe in yourself and push for it.” |
<filename>essem-reporter-protoc2/src/main/java/org/attribyte/essem/reporter/Proto2Builder.java<gh_stars>0
/*
* Copyright 2018 Attribyte, LLC
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
*
* You may obtain a copy of the License at
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing,
* software distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND,
* either express or implied.
*
* See the License for the specific language governing permissions
* and limitations under the License.
*/
package org.attribyte.essem.reporter;
import com.codahale.metrics.MetricRegistry;
import java.net.URI;
import java.net.URISyntaxException;
import java.util.Properties;
public class Proto2Builder extends Builder {
/**
* Creates a builder.
* @param uri The essem endpoint URI.
* @param registry The registry to report.
*/
Proto2Builder(final URI uri, final MetricRegistry registry) {
super(uri, registry);
}
/**
* Creates a builder.
* @param props The properties.
* @param registry The registry to report.
* @throws IllegalArgumentException if a property is invalid.
* @throws URISyntaxException if the report URI is invalid.
*/
Proto2Builder(final Properties props, final MetricRegistry registry) throws IllegalArgumentException, URISyntaxException {
super(props, registry);
}
@Override
public EssemReporter build() {
return new Proto2Reporter(uri, authValue, deflate,
registry, clock, application, host, instance, role, description, statusSupplier, filter, rateUnit, durationUnit,
skipUnchangedMetrics, hdrReport, alertSupplier);
}
} |
<reponame>jonnrb/radarsign
package main
import (
"flag"
"time"
)
var (
addr = flag.String("addr", "0.0.0.0:8080", "an address, port pair the HTTP server will listen on")
downloadTime = flag.Duration("time.download", 10*time.Second, "total time to spend trying to probe download speed")
uploadTime = flag.Duration("time.upload", 10*time.Second, "total time to spend trying to probe upload speed")
configureTimeout = flag.Duration("time.configure", 5*time.Second, "maximum time to spend getting the initial speedtest.net configuration")
throttleTime = flag.Duration("time.throttle", 5*time.Minute, "how long to wait before allowing speed probe metrics to be refreshed")
candidateServers = flag.Int("candidate_servers", 5, "number of servers whose latency gets checked during a speed probe")
)
|
// Called when there is nothing playing anymore
public static void clearNowPlaying()
{
masterColor = PAUSED_COLOR;
masterState = 1;
writeFile(stateFile, "1");
writeFile(masterColorFile, PAUSED_COLOR.getRed() + "," + PAUSED_COLOR.getGreen() + "," + PAUSED_COLOR.getBlue());
for (NowPlayingProviderConsumerService consumerService : NowPlayingProviderServer.connectedConsumerServices)
consumerService.sendNowPlayingInfo();
} |
def merge_npt(cls, npts_flat):
npts = pd.DataFrame(npts_flat)
npts.set_index(0, inplace=True)
npts.index.name = 'pcap'
npts.fillna(-1, inplace=True)
npts = npts.apply(pd.to_numeric, downcast='integer')
(header, max_length) = npts_flat.result
if header is not None:
npts.columns = cls.flatten_columns(header, max_length)
print('nPrints padded to maximum size:', max_length)
print('nPrint features',
'shape:', npts.shape,
'dtypes:', npts.dtypes.unique(),
'size (mb):', npts.memory_usage(deep=True).sum() / 1024 / 1024)
return npts |
Acylation A New Means to Control Traffic Through the Golgi The Golgi is well known to act as center for modification and sorting of proteins for secretion and delivery to other organelles. A key sorting step occurs at the trans-Golgi network and is mediated by protein adapters. However, recent data indicate that sorting also occurs much earlier, at the cis-Golgi, and uses lipid acylation as a novel means to regulate anterograde flux. Here, we examine an emerging role of S-palmitoylation/acylation as a mechanism to regulate anterograde routing. We discuss the critical Golgi-localized DHHC S-palmitoyltransferase enzymes that orchestrate this lipid modification, as well as their diverse protein clients (e.g., MAP6, SNAP25, CSP, LAT, -adrenergic receptors, GABA receptors, and GLUT4 glucose transporters). Critically, for integral membrane proteins, S-acylation can act as new a self-sorting signal to concentrate these cargoes in rims of Golgi cisternae, and to promote their rapid traffic through the Golgi or, potentially, to bypass the Golgi. We discuss this mechanism and examine its potential relevance to human physiology and disease, including diabetes and neurodegenerative diseases. INTRODUCTION A major function of the Golgi is to receive, sort and modify proteins and lipids and to deliver these cargoes to new locations . Impaired coordination of this highly orchestrated assembly line can result in abnormal glycosylation (;Fisher and Ungar, 2016) or a virtual "traffic jam." In neurons, Golgi dysfunction is associated with numerous diseases such as Parkinson's, Alzheimer's, and Huntington's diseases, as well as with cognitive disorders (Glick and Nakano, 2009;Bexiga and Simpson, 2013;Rabouille and Haase, 2015). Altered Golgi function in other cell types may also contribute to disease. Yet, how the mammalian Golgi sorts proteins and lipids remains poorly understood. ORGANIZATION OF THE MAMMALIAN GOLGI In mammalian cells the Golgi looks like a perinuclear crescent or "ribbon" by light microscopy and as a "stack" of pancake-like cisternae by three-dimensional electron microscopy (;Huang and Wang, 2017;), with the cis side closest to the ER. The cisternae are stacked by GRASP proteins (Zhang and Wang, 2015) and individual Golgi "mini"-stacks are laterally linked to form an extensive Golgi ribbon (Wei and Seemann, 2017;). The crescent-shaped ribbon is often several micrometers long and highly convoluted. It contains, on average, five cisternae with a thickness of 50 nm each, which are separated by intercisternal spaces of 5-15 nm (). This organization of the Golgi differs drastically from that found in lower eukaryotes: yeast Golgi is neither stacked nor linked and within the cisternae, the Golgi enzymes exchange dynamically to cause maturation of the enclosed cargo (;Glick and Nakano, 2009). These differences in Golgi structure underscore one of the biggest and most debated riddles in cell biology: How does the Golgi function to sort anterograde traffic? Whereas for retrograde traffic there is a clear consensus that COPI vesicles mediate retrograde flux, for anterograde traffic multiple models have been proposed, including those of vesicular transport, cisternal maturation, and others (;Jackson, 2009;Pfeffer, 2010;Glick and Luini, 2011). Central to this longstanding debate is the question of whether cisternae within the mammalian Golgi ribbon mature (i.e., dynamically exchange their enzymes), or whether the enzyme composition of a given cisterna is stable, with cargo passing through static successive layers via vesicular/tubular carriers (Glick and Luini, 2011). Two longstanding challenges in addressing this question are that current imaging cannot visualize live Golgi dynamics at sufficient resolution to unambiguously distinguish between these models (which may require live cell imaging at tens of nanometers in 3D), and the machinery used for anterograde traffic is debated (e.g., Do COPI vesicles carry only retrograde traffic or do they act in both directions?). Arguably, the only current consensus is that there is no consensus. But might there also be other means to drive cargo forward? A CIS-FACE PROBLEM: HOW TO DISTILL ANTEROGRADE CARGO FROM RETROGRADE PROTEINS AND LIPIDS? A remarkable requirement of Golgi function upstream of intra-Golgi transport lies in the segregation of anterograde from retrograde cargo and lipids (Glick and Nakano, 2009). Indeed, one of the largest remaining questions in Golgi biology is exactly how sorting of anterograde from retrograde cargo (ER resident proteins, lipids) is achieved upon Golgi entry. While glycosylation is a highly coordinated and critical function of the Golgi, there is little evidence that it is a driving force for sorting (Gomez-Navarro and Miller, 2016). But how much sorting is there? This has been addressed by comparing the total amount of anterograde membrane that leaves the ER via COPII vesicles to the amount of new membrane actually required to sustain cell growth (Barlowe and Helenius, 2016;). Notably, about 90% of the membrane has to be recycled to the ER and, in light of this large retrograde backflow, the incoming cargo has to be extensively concentrated to achieve an efficient rate of anterograde net movement. The question remains as to whether anterograde sorting of cargo at the Golgi is an active process (i.e., mediated by unknown signals), or whether a net anterograde flux is achieved simply by the selective retrograde retrieval of lipids and proteins upon entry into the Golgi. More simply, is active sorting required to move cargo forward, or not? More is understood about the retrograde traffic machinery, which has been extensively characterized and appears to consist mainly of COPI vesicles and tubular Rab6-connections. Both act in conjunction with the selective retrieval of retrograde cargo and depend on basic amino acids that serve as concentration and retrieval signals (;Duden, 2003;Barlowe and Helenius, 2016). How sorting of anterograde cargo is achieved at the cis face of the Golgi is much less clear. One obvious concept is that there may be an anterograde amino acid sorting signal. Here, it was recently proposed that acidic residues on the cytoplasmic tail of a model transmembrane cargo, vesicular stomatitis virus glycoprotein (VSV-G) could promote anterograde routing at the Golgi (). Nevertheless, the lack of such residues in the cytoplasmic tails of multiple viral spike proteins (HA, NA, and GP), as well as mammalian integral PM resident proteins, suggests that such signals do not act generally in anterograde sorting, but rather are specific to VSV-G at COPII exit sites in the ER (Votsmeier and Gallwitz, 2001). The Cis-Golgi Is a Hot Spot for Protein Palmitoylation Hints that some type of lipid-based anterograde sorting signal might exist appeared in the 1980s, when it was shown that efficient intra-Golgi cargo transport requires a specific type of activated lipid, palmitoyl-CoA (Glick and Rothman, 1987). Furthermore, it was discovered that the hydrolysis and transfer of palmitoyl-CoA is necessary for both vesicle budding and fission, and that its presence is needed in donor Golgi membranes (;). These data positioned the acylation reaction at the cis-Golgi. However, despite uncovering a potential role of palmitoyl-CoA in anterograde Golgi sorting, the underlying molecular mechanism could not be elucidated. PALMITOYLATION -LIPIDATION OF PROTEINS AS A MOLECULAR SWITCH? In recent years, palmitoyl-CoA has increasingly been recognized for its importance in the fields of developmental and cell biology, in microbiology, and in neuroscience (El-Husseini ;Linder and Deschenes, 2007;Fukata and Fukata, 2010;). In addition to its role in the synthesis of membrane lipids (), palmitate is attached covalently and post-translationally to several hundreds of proteins (). These protein lipidation (or fatty acylation) reactions can occur on three types of different amino acids: on cysteine residues (forming thioester linkages, S-palmitoylation), on serine/threonine residues (forming oxyester linkages, O-palmitoylation), or on primary amino groups (forming amide linkages, N-palmitoylation) (). Among these distinct types of acylation, S-palmitoylation stands out, as the thioester-bond is easily reversible and can thus act as a two-way toggle. Further, acylation has an inherent biophysical propensity to dynamically alter the properties of the modified protein, including the oligomerization state within a membrane or the targeting of soluble proteins to membranes (Smotrys and Linder, 2004). As such, S-acylation is well-positioned to function as a molecular and biophysical switch. For clarity, while the fatty acid S-acylation is typically palmitate, other fatty acids such as stearate (C18) and unsaturated variants can be used. DHHC S-PALMITOYLTRANSFERASES ARE FOUND IN KEY POSITIONS ACROSS THE SECRETORY PATHWAY In mammals, the enzymes responsible for protein S-acylation are a family of 23 proteins, termed protein-acyl transferases or PATs. These integral membrane proteins typically contain four to six transmembrane domains and a cysteine-rich domain (CRD) within a cytoplasmic loop. A conserved Asp-His-His-Cys (DHHC) tetrapeptide motif is located within the CRD and is the active site of each "DHHC" enzyme (Fukata and Fukata, 2010;Greaves and Chamberlain, 2011;De and Sadhukhan, 2018;). The fact that the active site is positioned at the cytoplasmic face dictates that unlike N-palmitoylation, which takes place in the lumen of the Golgi (Buglino and Resh, 2008;), S-acylation occurs exclusively in the cytoplasm. Acylation of a substrate protein is thought to be mediated by a two-stage mechanism: auto-acylation of a single or multiple Cys residue(s) within the CRD of the DHHC enzyme by Acyl-CoA to form an enzyme-acyl thioester moiety, and the Cys residue from a substrate protein attacks the enzyme-acyl intermediate to form a substrate-acyl product (). Notably, integral membrane proteins frequently have Cys residues immediately adjacent to their transmembrane anchors (), and recent crystallographic data position the active site of the PAT near the cytoplasmic leaflet (). Unlike most other posttranslational modifications, S-acylation does not seem to require a consensus motif. Rather, it appears that the juxtamembrane positioning of the substrate Cys residue close to the PAT active site is required for acylation of the substrate (). How such a spatial positioning is achieved to promote the acylation of soluble proteins is even less clear. A separate set of enzymes, called acyl-protein thioesterases (APTs), mediates the de-acylation of S-palmitoylated proteins (;Lin and Conibear, 2015;). Interestingly, APTs are soluble proteins that are recruited to Golgi membranes, in part by S-palmitoylation, and that are further able to undergo auto-depalmitoylation. Recent work shows that one isoform, APT1, resides on mitochondria and that the S-palmitoylation of mitochondrial proteins is dynamically regulated (). Thus in distinct cellular membranes, acylated cargo proteins are positioned in vicinity to enzymes that remove this modification, so that the abundance of acylated proteins may be maintained under homeostatic control (). Given that there are 23 DHHC PAT isoforms in mammals, several important questions naturally arise. Where are these acylation/de-acylation machineries localized? Do PATs exhibit differential expression patterns across different tissues? Are they present in different stages of development or stages of cell proliferation? Ohno et al. set out to address these points and examined both the subcellular location of overexpressed PAT isoforms as well as their expression levels in different tissues (). Strikingly, the majority of DHHC isoforms appeared to localize exclusively or partially to the Golgi, while others exhibited a distinct localization to the ER, the PM, or to vesicular-tubular structures. Furthermore, a tissue-specific expression pattern was observed for multiple isoforms, e.g., DHHCs 11, 19, and 20 appear exclusively in the testis, while many were ubiquitously expressed and only appear to be absent in distinct tissues. S-Palmitoylation Is a Committed Step for Anterograde Transport at the Cis-Golgi A recurring theme in the literature is the correlation between S-palmitoylation and the affinity of membrane proteins for sphingolipid/cholesterol-containing membranes (;;Ernst R. et al., 2018). The observations that DHHC isoforms localize to distinct positions within the secretory pathway and are present in distinct tissues at specific levels suggest that S-palmitoylation serves a purpose beyond that of targeting proteins to cholesterol-rich microdomains. However, the strong concentration of DHHC isoforms in the Golgi could be interpreted as a requirement of protein lipidation to achieve compatibility with the complex (and cholesterol-sphingolipid rich) membranes encountered at the trans-Golgi network and beyond. A gradient of sphingolipids and cholesterol exists within the Golgi, with the trans-Golgi/TGN exhibiting the highest concentrations of both lipid classes (Van ). If S-palmitoylation serve the purpose of sorting anterograde cargo to "raft-like" membranes, this is presumably where DHHC enzymes should be localized. CONCENTRATION OF DHHC ISOFORMS IN THE CIS-GOLGI Recently, the intra-Golgi localization of DHHC PATs was investigated in a quantitative manner using exogenously expressed tagged constructs (Ernst A.M. et al., 2018). 17 out of 23 DHHC PATs exhibited significant overlap with endogenous Golgi markers, and 9 localized to the Golgi exclusively: DHHCs 3,7,9,11,13,15,17,21,and 22. Strikingly, 6 out of 9 DHHCs exhibited strong co-localization with endogenous cis-Golgi markers (DHHCs 3, 7, 13, 17, 21, and 22), while the remaining 3 were positioned at the trans-Golgi (DHHCs 9, 11, and 15). Interestingly, all cis-Golgi-localized DHHCs are expressed in most tissues, while those detected at the trans-Golgi are highly tissue-specific (). If sorting into "raft-type" microdomains might be a main function of protein S-palmitoylation, then the observations prompt the question of why the entire ubiquitously expressed pool of Golgi DHHC PATs is found in the cis . Notably, cell-free system studies of minimal components needed for anterograde Golgi transport demonstrated that palmitoyl-CoA in cis Golgi donor membranes greatly facilitated anterograde cargo transport and budding (Glick and Rothman, 1987;;). Together, this suggests that palmitoylation may do more than only sorting proteins to "raft" membrane domains. S-PALMITOYLATION INDUCES ANTEROGRADE SORTING OF MEMBRANE CARGO Independently, Ernst A.M. et al. employed a clickable analog of palmitate, alkyne-palmitate, to identify in mammalian cells the major site of S-acylation activity. Pulse-chase based metabolic labeling revealed a rapid and specific incorporation of palmitate into the cis-Golgi. This incorporation depended on activation of the probe with CoA and resulted in thioester linkages to proteins within Golgi membranes other than the aforementioned DHHC PATs; specificity was validated by Triacsin C (a competitive inhibitor of Acyl CoA synthetase) and hydroxylamine-sensitivity of the palmitate labeling in situ and of proteins on SDS-PAGE (neutral hydroxylamine cleaves thioester linkages present of S-acylated proteins, but not oxyester linkages formed from incorporation of palmitate into lipids). In pulse-chase experiments, the S-palmitoylated proteins partitioned over time from the cis-to the trans-Golgi, in a strictly anterograde fashion, with no signal appearing in the ER. The cargo then appeared at the PM, but this did not occur if DHHC enzyme activity was impaired. To identify the cis-Golgi PAT responsible for the apparent anterograde cargo routing, candidate Golgi-localized PATs (DHHCs 3,7,9,11,13,15,17,21,22) were overexpressed and probed for catalysis of anterograde routing of S-palmitoylated proteins. Only the closely related DHHCs 3 and 7 were capable of catalyzing the rate and extent of anterograde transport of bulk S-acylated proteins. Concordantly, the model S-acylated substrates VSV-G and transferrin receptor were probed for a differential partitioning through the Golgi as a function of their acylation status (importantly, both cargoes are classical "non-raft" markers, ruling out a sphingolipid/cholesterol-dependency of the sorting). Strikingly, while the rate of entry into the Golgi was identical for wildtype and mutant cargoes, transport through the Golgi was slowed when these cargo proteins were not acylated, strongly suggesting that S-acylation represents a sorting event, routing them efficiently along an anterograde track. These data suggest that the biochemical requirement for palmitoyl-CoA for in vitro reconstitution trafficking assay detected in reports 30 years prior (Glick and Rothman, 1987;) stemmed from S-palmitoylation of the model cargo VSV-G employed in the cell-free system, and that no other unknown cofactors were involved in modulating partitioning of VSV-G form donor to acceptor Golgi membranes. In search for an explanation of how S-palmitoylation modulated the anterograde routing of cargo, alkyne-palmitate-based metabolic labeling was combined with electron tomography. In agreement with an earlier observation that indicated VSV-G preferentially accumulated at the cisternal rims (), the authors found that bulk S-palmitoylated proteins are indeed strongly enriched in the highly curved perimeters of the cisternal rim, which consists of tubules and fenestrated sheet-like elements (). In order to test whether sorting to areas of high curvature results directly from S-acylation of the cargo, model acylated transmembrane peptides were probed for a partitioning between flat and curved membranes in vitro. Strikingly, acylation of the peptides resulted in a strong partitioning into highly curved membranes, to an extent in line with the increase in anterograde transport observed for trafficking of model acylated cargoes through the Golgi. Together, the data strongly support a model whereby S-acylation directly modulates the biophysical property of the cargo, resulting in its increased partitioning to the cisternal rim of cis-Golgi membranes, which in turns facilitates its anterograde routing (see Figure 1). ADDITIONAL EXAMPLES FOR ANTEROGRADE ROUTING OF S-ACYLATED MEMBRANE CARGO Linker for Activation of T Cells (LAT) Hundt et al. investigated the trafficking of LAT, a dually S-palmitoylated protein that is a crucial signaling molecule for T-cell receptor-based stimulation of T-cells (). Anergic T cells lack palmitoylation of LAT, resulting in a reduction in Tyr phosphorylation and activation of PLC1 (). Hundt et al. demonstrated that when LAT is not S-palmitoylated, its coupling to sphingolipid/cholesterol-rich membranes is not affected, but rather results in LAT's accumulation in the Golgi, with negligible levels detected at the PM. Further, LAT is in the family of transmembrane adaptor proteins (TRAPs), and the additional family members LIME and NTAL/LAB also were shown to require S-palmitoylation for efficient export to the PM (). Thus, anterograde routing of TRAPs via S-palmitoylation at the Golgi emerges as a requirement for T-cell function. -Adrenergic Receptors (AR) -adrenergic receptor (AR) isoforms 1-3 are all S-palmitoylated proteins, but these isoforms exhibit different sites of acylation. Recently, Adachi and colleagues investigated mammalian 3 AR, and observed that a distinct site (Cys-153) is crucial for proper targeting of the receptor to the PM, while other sites (Cys-361/363) impact its stability at the PM (). These observations prompt the hypothesis that within the AR family, different sites of S-palmitoylation are employed to toggle between states, and may control AR targeting to different FIGURE 1 | S-palmitoylation acts as a switch to sort anterograde cargo through the Golgi. Membrane cargo entering the Golgi undergoes rapid S-palmitoylation by cis-Golgi resident DHHC protein-acyl transferases (top left). Acylation of the membrane protein results in a change in the spontaneous curvature of cargo (C 0 > 0), generating curvature stress (top right). This can also concentrate the acylated cargoes. Middle panel: effect of S-acylation on cargo transport in the vesicular/tubular transport model (stable cisternae). Acylation causes the cargo to be stabilized in regions of higher curvature, e.g., in tubular intercisternal connections (), the highly curved and fenestrated cisternal rim (), and in tubular or vesicular carriers budding thereof (Glick and Rothman, 1987;;), putatively providing a mechanism for Golgi bypass (arrow). Lower panel: effect of S-acylation on cargo transport in the cisternal progression/maturation model: S-palmitoylation triggers an entrapment of cargo at the cisternal rim. This facilitates the extraction of retrograde lipids and cargo and reduces the inclusion of anterograde cargo in the retrograde carriers. Thioesterase activity allows for cycles of lateral diffusion back to the cisternal center and re-acylation by DHHCs. cellular locations, modulating its surface expression and hence downstream signaling. GABA Type a Receptors (GABA A Rs) GABA receptors are well-established S-acylated PM residents. Based on studies using overexpressed DHHCs, they were postulated to be palmitoylated by DHHCs 3 and 7 (). Subsequently, Kilpatrick et al. identified DHHC3 as a specific PAT of the 2 subunit of GABA A Rs through knockout of either DHHC3 or 7 in mice (). Whereas knockout of DHHC7 had no effect on GABA A R trafficking in neurons, DHHC3 KO neurons exhibited drastically reduced levels of GABA A R 2 at synapses, impacting synaptic function. Most remarkably, and in line with the identification of DHHCs 3 and 7 as ubiquitous regulators of protein sorting at the cis-Golgi (Ernst A.M. et al., 2018), knockout of the individual PATs resulted in only marginal effects, while a double knockout of DHHCs 3 and 7 resulted in perinatal lethality of mouse embryos, emphasizing the importance of these PATs and sorting at the cis-face of the Golgi for general cell function. Glucose Transporter 4 (GLUT4) Insulin stimulates glucose uptake in fat and muscle cells by causing the exocytic translocation of GLUT4 to the PM. Although this had previously been considered exclusively as a post-Golgi process, more recent data make clear that recycled GLUT4 accumulates in a pool of small (∼50 nm diameter) vesicles that reside near the ERGIC and cis-Golgi compartments, in association with TUG, Golgin-160, and ACBD3 (Acyl-CoA Binding domain-containing protein 3, also known as GCP60) (;Bogan, 2012;Orme and Bogan, 2012;). Upon insulin stimulation, these vesicles are mobilized by TUG cleavage, and they are proposed to traffic to the cell surface by an unconventional secretion pathway that bypasses the Golgi stack (;Bogan, 2012;). GLUT4 is S-acylated and the PAT responsible for this modification was recently identified (). Both DHHCs 3 and 7 bound to GLUT4, but only silencing of DHHC7 abolished GLUT4 acylation, leading to the conclusion that DHHC7 acts as a specific PAT for GLUT4. Other proteins that cotraffic with GLUT4 or that regulate this process are also palmitoylated in adipocytes (a). Importantly, the ability of insulin to stimulate GLUT4 translocation was impaired by knockdown or knockout of DHHC7 and by mutation of the palmitoylated Cys residue in GLUT4 (;). Thus, it may be that S-acylation is required to concentrate the GLUT4 in the highly curved membranes of the insulin-responsive vesicles. S-PALMITOYLATION CONFERS MEMBRANE PROTEINS A "SORTING-COMPETENT" STATE How might S-palmitoylation of proteins induce partitioning to the highly curved rims of Golgi cisternae? As noted above, S-palmitoylation occurs exclusively at the cytoplasmic leaflet of endomembranes, frequently on one or two adjacent sites and at juxtamembrane positions of integral membrane proteins. It is known that the cis-Golgi cisternae are stacked by the action of GRASP65 proteins, which leads to a significant flattening of membranes in the stacked area (). Upon S-palmitoylation, the acylated proteins would exhibit a local mass excess in the cytoplasmic leaflet of the membrane, concomitant with an asymmetric hydrophobic Z-profile (i.e., an increase in spontaneous curvature/transition from cylindrical to conical profile); see top right panel of Figure 1. In support of this concept, early studies on erythrocytes demonstrated that drugs intercalating into the cytoplasmic leaflet induced a morphological change from a flat and disc-like to spherical morphology (Sheetz and Singer, 1974). In the Golgi, the tightly stacking proteins in the central disk region would inhibit such a morphological transition -therefore, the local mass excess in the cytoplasmic leaflet due to S-palmitoylation is expected to induce curvature stress, forcing acyl chains in the cytoplasmic leaflet to potentially form locally curved clusters. This stress could presumably induce the observed partitioning toward the positively curved cisternal rims, where S-palmitoylation-induced curvature stress would be released (Figure 1). This reasoning suggests that S-palmitoylation serves as a biophysical switch and sorting signal at the cis Golgi, to extract cargo proteins from planar membranes and concentrate them at the cisternal rims. The cisternal rim is fenestrated and comprises a network of tubular-vesicular elements, referred to as the "non-compact zone/region" (). A significant fraction of PM resident proteins (>15%) are predicted to be S-acylated (Ernst A.M. et al., 2018). Yet, the data nonetheless raise the question how anterograde cargo that lack this sorting signal can be efficiently sorted to the cisternal rim. The diffusive flux along the gradient generated by S-palmitoylated proteins would have the potential to drag along non-acylated proteins. In line with this hypothesis, overexpression of cis-Golgi DHHCs resulted in an increased flux of (soluble) secretory cargo, and even impacted non-acylated cargo to a lower but significant extent (Ernst A.M. et al., 2018). Multiple additional scenarios can also be envisioned, and could contribute to spontaneous curvature-based sorting of proteins at the Golgi: conicity of the hydrophobic moiety (the membrane anchor) acquired through hetero-oligomerization or protein folding, as observed for the polytopic membrane channel KvAP, which exhibits affinity for areas of high curvature without being lipidated (), heterooligomerization of palmitoylated and non-palmitoylated proteins, and coagulation of integral S-palmitoylated proteins and luminal proteins via lectins, e.g., the mannose-specific lectin VIP36 that is present at the cis Golgi (). A further consideration is that S-acylation-induced sorting of membrane proteins may apply to intracellular organelles other than the Golgi. At the ER, notably, S-palmitoylation of the ER-resident Calnexin alters its localization toward mitochondria-associated membranes (MAM), which exhibit a high extent of positive curvature (). Supportively, access of the ER-thioredoxin TMX to MAM also requires S-acylation (). However, S-acylation of Calnexin was also suggested to regulate access to sheet-like structures (), which might not be a contradiction given that a recent study detected nanoholes within ER-sheets (that would give rise to curvature, ). Additionally, the endosome-localized DHHC15 was suggested to concentrate cargo for efficient uptake by retromer (). Thus, S-palmitoylation may be a ubiquitous mechanism to cause the partitioning of integral membrane proteins into regions of high curvature. From an evolutionary standpoint, S-palmitoylation represents a widespread post-translational modification that is conserved from yeast to mammals, plants, and even encountered in multiple parasites (;;;;). The fact that virtually all spike proteins of enveloped viruses are S-palmitoylated, regardless of whether or not their viral exit sites are enriched in sphingolipids/cholesterol (e.g., Influenza A NA, VSV-G, Ebola GP), further emphasizes the evolutionary pressure toward maintaining this putative anterograde sorting signal to maximize the efficiency of their anterograde routing at the Golgi. Possibly, palmitoylation may also induce curvature at the PM, which may be relevant for budding of membrane, in addition to targeting soluble and membrane proteins ;). Controlled Access of Soluble Proteins to the Cis-Golgi S-Palmitoylation Machinery Two members within the DHHC PAT family, the huntingtin-interacting proteins DHHC 13 and 17, particularly stand out because of their architecture. They contain long amino-terminal ankyrin-repeats, which were shown to bind to specific soluble substrate bearing corresponding ankyrin-binding motifs (). Among their substrates are the critical neuronal proteins MAP6 and CSP, but also the ubiquitous t-SNAREs in the SNAP25 family ). Interestingly, Lemonidis et al. demonstrated that binding of these substrates to DHHCs 13 and 17 does not or only marginally induces their acylation, whereas DHHCs 3 and 7 are not able to bind the proteins but efficiently mediate their acylation. The fact that DHHCs 3, 7, 13, and 17 all localize to the cis-Golgi strongly suggest cooperativity among the PATs in this subcompartment, with DHHCs 13 and 17 serving to recruit these proteins to the cis-Golgi, where they are subsequently S-palmitoylated by DHHCs 3 and/or 7 in a concerted, two-stage process. As noted above, several proteins involved in the regulation of GLUT4 translocation are palmitoylated, and some of these may also be recruited as soluble proteins onto highly curved membranes (a,b;). In unstimulated cells, the insulin-responsive vesicles are trapped in association with ACBD3, which binds palmitoyl-CoA (;). The related protein ACBD6 facilitates N-myristoylation of substrate proteins by cooperating with N-myristoyltransferase enzymes (Soupene and Kuypers, 2019). Whether ACBD3 may participate in palmitoylation of soluble proteins is not known, although in general ACBD family members have diverse roles in lipid-modification and metabolism pathways (;;) and in Golgi structure (;). Data also suggest that the accumulation of a pool of small, insulin-responsive vesicles may be impaired during the development of type 2 diabetes, which may result from excess membrane diacylglycerols and sphingolipids (;;Czech, 2017;;Petersen and Shulman, 2018). We speculate that these excess lipids might potentially disrupt the palmitoylation-based sorting mechanisms discussed above and thus contribute to attenuated insulin action. Outlook Enrichment of S-Palmitoylated Cargo in Vesicular-Tubular Carriers Versus Entrapment in Maturing Cisternae at the Golgi S-palmitoylation-induced sorting of cargo to the cisternal rim fits well with the long-proposed role of COPI vesicles as anterograde carriers in addition to their well-established role as retrograde carriers (Brandizzi and Barlowe, 2013). Artificially introducing stable adhesions ("staples") between Golgi cisternae does not impair anterograde transport, showing that anterograde transport primarily occurs from the highly curved rims, where COPI vesicles bud (). When mitochondria are engineered to invade the Golgi by affording them adhesion to its cisternae, these organelles dissect the cisternae apart and immobilize them (). Nonetheless, anterograde transport continues (), implying a small diffusible carrier. Independently, COPI vesicles, visualized by super-resolution microscopy, carry VSV-G protein between separated Golgi areas in fused cells and account for most of the cargo transported by this route (). We propose that palmitoylation-driven anterograde sorting bears the potential to help resolve the long-standing conundrum of how one coat could mediate two fates. Retrograde cargo is selected into COPI vesicles by classic receptor-dependent binding (such as KDEL -KDEL receptor or KKXX -coatomer (;Duden, 2003;Barlowe and Helenius, 2016)). We can envision a physical-chemical, membrane-intrinsic process by which S-palmitoylated anterograde cargo spontaneously clusters within curved regions to the exclusion of other proteins, including retrograde cargo and their receptors. The same COPI coat could then pinch off distinct anterograde and retrograde vesicles () from these differentiated regions of membrane. An alternative interpretation, along the lines of a cisternal maturation model for intra-Golgi transport (;), is that S-palmitoylation-induced sorting of cargo to the rim serves the purposes of redistributing the cargo while the cisternae mature (i.e., enzyme contents change). Potentially through the action of APT-mediated acylation-deacylation cycles, the de-palmitoylated cargo would lose their stable anchoring at the cisternal rim and partition back to the (flat) center of the cisterna. There, the cargo would engage in additional rounds of acylation by DHHCs, and through its partitioning back and forth across the cisterna, its chances of being processed by glycosyltransferases might increase. However, a challenge here is to explain how palmitoylated cargo would traffic through the maturing Golgi faster than non-palmitoylated cargo, as observed (Ernst A.M. et al., 2018). One would need to invoke that only the non-palmitoylated cargo partially traffic retrograde, for which there is currently scant evidence. A third option would be a "fast-track" through the Golgi mediated by intercisternal continuities. VSV-G translocation through the Golgi was reported to induce intercisternal connections (), and additional types of GTPase-dependent tubules were detected on the Golgi (;). S-palmitoylation could here putatively provide a mechanism to allow partitioning into these highly curved connections between cisternae (Figure 1). To address this will require visualization of sufficiently large areas of the Golgi, with specificity, at very high resolution. This may be possible with new advances in super-resolution imaging () and with large volume electron microscopy approaches such as FIB-SEM (). FUTURE STEPS: VISUALIZING DHHC PLATFORMS Acylation at the cis-face of the Golgi emerges as an important mechanism for routing of several proteins to the PM. Given that multiple DHHC PATs are capable of cooperating to achieve S-palmitoylation of specific substrates (;), coupled with the fact that several of these isoforms are present only in distinct tissues (), indicates that screening DHHC libraries to identify "specific" PATs for a protein of interest is not adequate. PATs are emerging as finely adjusted enzyme assemblies rather than isolated entities, and thus their spatial context (location within the same cisterna and across cisternae) has to be studiedideally at endogenous levels and by means of super-resolved light and electron microscopy. These nano-assemblies are expected to vary in composition across different tissues and during different developmental stages/states of cell proliferation. We hypothesize that DHHC platforms have the potential to provide a mechanism for fine-tuning the flux of specific substrates from the Golgi to the PM, thereby controlling the availability and timing of receptors, channels, and other types of cargo at the cell surface. With S-acylation as a marker for anterograde cargo in hand, live-cell and super-resolution microscopy-compatible orthogonal labeling strategies need to be developed to better visualize trafficking of anterograde cargo as a class (as opposed to individual proteins with varying properties) -on the nanoscale and on timescales that allow to observe complete transport/maturation cycles across the mammalian Golgi. These advances would allow investigations to reopen the exploration of putative differences in the mechanism of intra-Golgi transport between lower and higher eukaryotes, and furthermore would enable detailed studies of acylation-controlled protein export from the Golgi in response to external stimuli. Such studies could elucidate important fundamental principles, as well as insights into mammalian physiology and disease. AUTHOR CONTRIBUTIONS AE, DT, and JB researched the relevant literature, and conceived and wrote the manuscript. FUNDING AE and DT thank the Mathers Foundation. JB is grateful to the NIH (DK092661) and the American Diabetes Association for research support. |
Two lawsuits have been filed seeking information about the U.S. military’s plan to use women to “close with and kill” the enemy, based on reports that suggest having women in “tip of the spear” fighting units such as the Rangers and the Navy SEALs may be counterproductive.
The Obama administration announced two years ago it would make female military personnel eligible for assignment to direct ground combat units, including the infantry, beginning in January 2016. Under the military’s structure, women would be ordered into such positions.
Various military agencies and units since then have been analyzing the safety and effectiveness of the strategy along with the Center for Military Readiness, an independent group.
Lawyers with the Thomas More Law Center have submitted a number of Freedom of Information Act requests for details on behalf of CMR and its president, Elaine Donnelly, without effect.
So the law team announced it has filed a FOIA lawsuit against the U.S. Special Operations Command in federal court in the Eastern District of Michigan and a second against the Army in federal court in Washington.
Both are on behalf of Donnelly and seek records “related to the effectiveness of women in direct combat roles.”
Since founding CMR in 1993, Donnelly has researched and reported on social policy in military service. She recently has worked with Erin Mersino, senior trial counsel for Thomas More, on the information requests.
“Adherence to the FOIA is crucial because it allows the public access to our government,” Mersino said. “The documents we requested under FOIA are time sensitive. Permanent decisions regarding women in the infantry are projected to be made as soon as January 2016. The public should be informed of such important matters that directly affect our national security.”
When Obama announced his plan for women to hit the trenches, there were a few women who volunteered for tests, and CMR reported even that number is dwindling.
Read “Deadly Consequences” for the story about how Obama’s military strategy is “dead set on eviscerating our military by pushing women to the front lines.”
“Obtaining the documents asked for in the lawsuits will allow Elaine Donnelly to analyze the safety and effectiveness of allowing women in the infantry and provide its findings and analysis to the public and to the military at a crucial point in time,” the lawyers argue.
“Of particular interest … is the attempt by the Pentagon to insert women into the one of the most grueling training regimens in the entire military establishment, the U.S. Army Rangers. The deep concern now is that the Pentagon will reduce the physical requirements so that women will pass.”
Richard Thompson, the chief counsel for Thomas More, said the “question is not whether women should serve in combat, they already do, and admirably.”
“The question is whether women should purposely be placed in situations where they must close with the enemy in extremes of physical endurance, climate and terrain, brutal and violent death, injury, horror, and fear, just to satisfy the feminist agenda,” he said.
“Too many generals in the Pentagon know better, but they succumb to political pressure acting more like politicians than true military leaders. They already know that the end result will be compromised standards, destruction of the effectiveness of units like the Rangers and Navy SEALs, and disruption of the warrior spirit and ethos so carefully nurtured over the years.”
Most recently, Donnelly cited a report from Britain on the same issue.
The British Ministry of Defense released its report with “conflicted” evidence regarding “combat effectiveness” when women are added to ground troops.
Highlighting the “physiological differences, cohesion, and related factors,” it finds that “gender-integration problems” could be “mitigated.”
But Donnelly’s analysis notes the report admits, that if “the steps necessary to mitigate the risks are grossly disproportionate in terms of time, resources and cost,” that nation’s previous exclusion of women from ground combat troops “may have to remain in place.”
Donnelly succinctly sums up the report: “There are no benefits balancing the weight of costs and risks that detract from combat readiness and effectiveness. … Every use of the word ‘mitigate’ in the MOD report pinpoints a problem, not an advantage.”
She pointed out the simple goal of the military: “To close with and kill the enemy.”
The British study acknowledges that in that work, “violent death, injury, all-pervading concussive noise, horror, fear, blood and high levels of emotion are common themes.”
Significantly, the report found simple physiological differences “disadvantage women in strength-based and aerobic fitness tests by 20 percent to 40 percent, so for the same output women have to work harder than men.”
Size, stamina, injury susceptibility, morbidity, trauma, loads all are factors, the study said.
It reported 4.5 percent of women enlisting in the British Army are able to achieve [physical employment] standards, compared with 90 percent of the men.
WND previously reported on the studies in the U.S.
A Marine study showed for every man who fails a simulated artillery lift-and-carry test, 28 women fail. And for a test simulating moving over a seven-foot high wall, less than 1.2 percent of the men could not get over, compared to 21.32 percent of women.
The results were found in Marine Corps documentation by the Center for Military Readiness, which earlier issued a report called “U.S. Marine Corps Research Findings: Where is the Case for Co-Ed Ground Combat?”
According to CMR, the Obama administration expects the Marine Corps to find a way to assign women to ground combat units without lowering standards.
“In the independent view of CMR, quantitative research done so far indicates that these expectations cannot be met,” the group said
“Androgenic characteristics in men, which are not going to change, account for greater muscle power and aerobic (endurance) capabilities that are essential for survival and mission accomplishment in direct ground combat,” the report said.
According to the CMR study’s executive summary, the Marines obtained information from 409 men and 379 women who volunteered to perform five “proxy” tests to simulate combat demands.
“These capabilities are essential for survival and mission success in direct ground combat,” the study found.
In a pull-up test, women averaged 3.59 while men averaged 15.69 – more than four times as many.
A “clean and press” comprised single lifts of 70, 80, 95 and 115 pounds plus six repetitions of a 65 pound lift. Eighty percent of the men passed the 115 pound test but only 8.7 percent of the women.
In the 120 mm tank loading simulation, participants were asked to lift a simulated round weighing 55 pounds five times in 35 seconds or less. Men failed at a less than 1 percent rate while women failed at a rate of 18.68 percent.
The Marines said nearly one in five women “could not complete the tank loading drill in the allotted time.”
A separate report from Israel National News reported a book, “Lochamot Betzahal,” confirmed the results of 13 years of research on female participation in IDF combat units and declared the feminist experiment in the Israeli military a failure.
The New York Post reported fewer than 8 percent of Army women wanted a combat job. |
<filename>Algorithm.Python/FutureOptionBuySellCallIntradayRegressionAlgorithm.py
# QUANTCONNECT.COM - Democratizing Finance, Empowering Individuals.
# Lean Algorithmic Trading Engine v2.0. Copyright 2014 QuantConnect Corporation.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License
from datetime import datetime, timedelta
import clr
from System import *
from System.Reflection import *
from QuantConnect import *
from QuantConnect.Algorithm import *
from QuantConnect.Data import *
from QuantConnect.Data.Market import *
from QuantConnect.Orders import *
from QuantConnect.Securities import *
from QuantConnect.Securities.Future import *
from QuantConnect import Market
### <summary>
### This regression algorithm tests In The Money (ITM) future option calls across different strike prices.
### We expect 6 orders from the algorithm, which are:
###
### * (1) Initial entry, buy ES Call Option (ES19M20 expiring ITM)
### * (2) Initial entry, sell ES Call Option at different strike (ES20H20 expiring ITM)
### * [2] Option assignment, opens a position in the underlying (ES20H20, Qty: -1)
### * [2] Future contract liquidation, due to impending expiry
### * [1] Option exercise, receive 1 ES19M20 future contract
### * [1] Liquidate ES19M20 contract, due to expiry
###
### Additionally, we test delistings for future options and assert that our
### portfolio holdings reflect the orders the algorithm has submitted.
### </summary>
class FutureOptionBuySellCallIntradayRegressionAlgorithm(QCAlgorithm):
def Initialize(self):
self.SetStartDate(2020, 1, 5)
self.SetEndDate(2020, 6, 30)
self.es20h20 = self.AddFutureContract(
Symbol.CreateFuture(
Futures.Indices.SP500EMini,
Market.CME,
datetime(2020, 3, 20)
),
Resolution.Minute).Symbol
self.es19m20 = self.AddFutureContract(
Symbol.CreateFuture(
Futures.Indices.SP500EMini,
Market.CME,
datetime(2020, 6, 19)
),
Resolution.Minute).Symbol
# Select a future option expiring ITM, and adds it to the algorithm.
self.esOptions = [
self.AddFutureOptionContract(i, Resolution.Minute).Symbol for i in (self.OptionChainProvider.GetOptionContractList(self.es19m20, self.Time) + self.OptionChainProvider.GetOptionContractList(self.es20h20, self.Time)) if i.ID.StrikePrice == 3200.0 and i.ID.OptionRight == OptionRight.Call
]
self.expectedContracts = [
Symbol.CreateOption(self.es20h20, Market.CME, OptionStyle.American, OptionRight.Call, 3200.0, datetime(2020, 3, 20)),
Symbol.CreateOption(self.es19m20, Market.CME, OptionStyle.American, OptionRight.Call, 3200.0, datetime(2020, 6, 19))
]
for esOption in self.esOptions:
if esOption not in self.expectedContracts:
raise AssertionError(f"Contract {esOption} was not found in the chain")
self.Schedule.On(self.DateRules.Tomorrow, self.TimeRules.AfterMarketOpen(self.es19m20, 1), self.ScheduleCallbackBuy)
self.Schedule.On(self.DateRules.Tomorrow, self.TimeRules.Noon, self.ScheduleCallbackLiquidate)
def ScheduleCallbackBuy(self):
self.MarketOrder(self.esOptions[0], 1)
self.MarketOrder(self.esOptions[1], -1)
def ScheduleCallbackLiquidate(self):
self.Liquidate()
def OnEndOfAlgorithm(self):
if self.Portfolio.Invested:
raise AssertionError(f"Expected no holdings at end of algorithm, but are invested in: {', '.join([str(i.ID) for i in self.Portfolio.Keys])}")
|
<gh_stars>1-10
package fr.uem.efluid.services.types;
import fr.uem.efluid.model.entities.CommitState;
import java.util.Collection;
import java.util.concurrent.atomic.AtomicBoolean;
import java.util.concurrent.atomic.AtomicInteger;
import java.util.concurrent.atomic.AtomicLong;
/**
* <p>
* For combined multi-prepared commit
* </p>
*
* @author elecomte
* @version 2
* @since v0.0.8
*/
public class WizzardCommitPreparationResult {
private final long totalDiffSize;
private final int totalTableCount;
private final int totalProjectCount;
private final int totalDomainsCount;
private final PilotedCommitStatus status;
private final CommitState state;
/**
* @param totalDiffSize
* @param totalTableCount
* @param totalProjectCount
* @param totalDomainsCount
*/
public WizzardCommitPreparationResult(long totalDiffSize, int totalTableCount, int totalProjectCount, int totalDomainsCount,
PilotedCommitStatus status, CommitState state) {
super();
this.totalDiffSize = totalDiffSize;
this.totalTableCount = totalTableCount;
this.totalProjectCount = totalProjectCount;
this.totalDomainsCount = totalDomainsCount;
this.status = status;
this.state = state;
}
/**
* @return the totalDiffSize
*/
public long getTotalDiffSize() {
return this.totalDiffSize;
}
/**
* @return the totalTableCount
*/
public int getTotalTableCount() {
return this.totalTableCount;
}
/**
* @return the totalProjectCount
*/
public int getTotalProjectCount() {
return this.totalProjectCount;
}
/**
* @return the totalDomainsCount
*/
public int getTotalDomainsCount() {
return this.totalDomainsCount;
}
/**
* @return the status
*/
public PilotedCommitStatus getStatus() {
return this.status;
}
public CommitState getState() {
return this.state;
}
/**
* @param preps
* @return
*/
public static WizzardCommitPreparationResult fromPreparations(Collection<PilotedCommitPreparation<?>> preps) {
final AtomicLong totalDiffSize = new AtomicLong(0);
final AtomicInteger totalTableCount = new AtomicInteger(0);
final AtomicInteger totalProjectCount = new AtomicInteger(0);
final AtomicInteger totalDomainsCount = new AtomicInteger(0);
final AtomicBoolean allCompleted = new AtomicBoolean(true);
preps.forEach(p -> {
totalDiffSize.addAndGet(p.getSize());
totalTableCount.addAndGet(p.getTotalTableCount());
totalProjectCount.incrementAndGet();
totalDomainsCount.addAndGet(p.getTotalDomainCount());
if (p.getStatus() != PilotedCommitStatus.COMPLETED) {
allCompleted.set(false);
}
});
// For state based display in wizzard. Default is merge
CommitState setState = !preps.isEmpty() ? preps.iterator().next().getPreparingState() : CommitState.MERGED;
return new WizzardCommitPreparationResult(
totalDiffSize.get(),
totalTableCount.get(),
totalProjectCount.get(),
totalDomainsCount.get(),
allCompleted.get() ? PilotedCommitStatus.COMPLETED : PilotedCommitStatus.DIFF_RUNNING,
setState);
}
}
|
Comparing pregnancy outcomes between natural cycles and artificial cycles following frozenthaw embryo transfers Frozen embryo transfer (FET) is increasing in prevalence. In contrast to the amount of research performed on the actual cryopreservation procedure, there are limited data with respect to optimal endometrial preparation in FET cycles. Increasingly artificial cycle (AC) preparation is being adopted over the natural cycle (NC) to facilitate greater access to FET. However, there remains a paucity of data comparing pregnancy outcomes between these two commonly used cycle types. |
Pedagogical competence of sustainable aesthetic cosmetologists Development of self-care in cosmetology goes hand-in-hand with development of the public society. Nowadays, large increase of innovative hardware technologies, more sophisticated procedures for skin care and treating of skin problems are entering the field of Aesthetic Cosmetology. There is an increasing need for highly qualified professionals in the Aesthetic Cosmetology industry who are able to perform these complex procedures professionally and educate their client awareness. The cooperation between the Aesthetic cosmetologist and the client in health care is a mandatory condition, so the specialist's pedagogical is essential in order to inspire and motivate the client for a healthier lifestyle. This is particularly relevant in the context of sustainable development, which focuses on the sustainability of humanity as a whole. Pedagogical competence is already developed during the study course. Student already learns to correctly and professionally transfer his professional knowledge and skills to his clients through healthcare. Professional higher education contributes to improving the pedagogical skills of prospective specialists. At the University of Latvia P. Stradins Medical College, the students are prepared for professional work and attitude of an Aesthetic cosmetologist during their studies in the course Aesthetic cosmetology. Self-motivation of Aesthetic cosmetologists to develop themselves is important throughout their professional work, so they can provide quality and modern healthcare. The aim of the study to find out the views of aesthetic cosmetologists about pedagogical competence. The research methods such as questionnaires, self-evaluation, discussions and experiment were used in the current investigation. The place of the research: Riga, Latvia. Altogether 28 respondents take part in this investigation. The study results demonstrated different results before and after approbation the study course Pedagogy in Aesthetic Cosmetology. After the study course, more and more respondents demonstrate the need for pedagogical competency in the field and plan to educate their clients and transfer professional knowledge. Most of the respondents (98 %) self-assessed their pedagogical competences very high completely sufficient and almost sufficient. |
Call it a geek-culture state fair, a pop-culture spectacle or an entertainment circus.
But whatever moniker you choose to slap on it, you'd better also slap on your sword and superhero boots: Wizard World, one of the biggest pop-culture conventions in North America, is poised to land at Madison's Alliant Energy Center the first weekend in February, much like the Incredible Hulk will land on Tony Starks' Hulkbuster armor in this summer's Avengers sequel.
How big is it, you ask? Wizard World is headlined by William Shatner, who, depending on your generational prism, is either the legendary Captain Kirk from Star Trek or the guy who gets you great rates on hotel rooms and rental cars. The three-day extravaganza features appearances by actors from big-time entertainment properties like Marvel's Agents of S.H.I.E.L.D. (Brett Dalton), The Vampire Diaries (Ian Somerhalder) and True Blood (Racine native Kristin Bauer van Straten). Bruce Campbell, who has parlayed his role as Ash in the classic Evil Dead films into a long and lucrative geek-culture career, will be there taping episodes of Last Fan Standing, a game show centered on geek trivia.
Comic artists and animated films are on the docket. And, of course, plenty of people dressed up in superhero spandex, zombie chic and chainmail will be letting their geek flags fly.
Wizard World has always been perched high in the pop-culture firmament -- the con's annual Chicago event is second only to San Diego's International Comic Convention in terms of size and scope -- but for most of its existence, the event has been held in bigger cities like New York and Philadelphia. As geek culture has expanded to every corner of the entertainment universe over the past five years, Wizard World has seized the opportunity: Last year, the show added seven new midsized locations in places like Tulsa, Richmond and San Antonio, and this year chose Madison to be one of the new cities for 2015.
"Madison deserved a show. The audiences there deserved it," says John Macaluso, the affable 58-year-old CEO of Wizard World Inc., the entertainment company that stages the convention events. The business grew from the ashes of a print magazine that used to track the value of collectible comic books back in the late 1990s and early 2000s.
Macaluso calls Madison a "clear sky" city -- meaning the field is clear of similarly sized convention competitors, and Wizard World will tap into the state's deep geek-culture roots. Milwaukee hosted Gen Con, the granddaddy of tabletop gaming conventions, for 18 years before the event picked up stakes and moved to Indianapolis in 2003. Today, Madison has its own thriving community of smaller-but-growing geek-culture cons, events like the sci-fi focused Odyssey Con, the feminist-focused WisCon, Gamehole Con and Geek.Kon, an event that pulled 2,400 people last year.
Adam Pulver, one of Geek.Kon's primary organizers, says the news that Wizard World would be stopping in Madtown surprised him. "Madison is a bit off the beaten path for them," he says. "That said, Madison has always been a very geek-friendly city, with many game and comic stores, along with several professionals in the industry."
He means people like John Kovalic, the artist who created the sublime Dork Tower comic strip, as well as classic games like Apples to Apples and Munchkin. For decades, Kovalic has enjoyed a front seat at the cosmic convergence between pop and geek culture and fans' growing appetite to indulge. Ten years ago, Kovalic was doing comic strips riffing on the fact that a mere handful of celebs -- cats like Wil Wheaton, from Star Trek: The Next Generation, ex-major league pitcher Curt Schilling and voice-of-Groot Vin Diesel -- were dipping their toes in geekly gaming pursuits.
"Today, there's nothing shocking about celebrities immersing themselves in geek culture," says Kovalic. 'You've got the cast of Downton Abbey playing Cards Against Humanity."
Wizard World's Macaluso agrees. "The genres we go after -- things like horror, superheroes, sci-fi and fantasy -- are growing by leaps and bounds," he says.
He's not just blowing Pym Particles, those subatomic specials that make Ant-Man a possibility. A TV show like AMC's The Walking Dead went from zero to 23 million viewers in an undead heartbeat. Marvel's Guardians of the Galaxy is just the latest in a long line of superhero flicks to conquer year-end profit charts. Even if you've been living under that proverbial rock, a pack of Marvel superheroes, a troop of bloodthirsty zombies and the bloodlusty Lannisters from HBO's Game of Thrones have probably found you and smashed that boulder into so much pixie dust.
According to Judy Frankel at the Greater Madison Convention & Visitors Bureau, Wizard World Madison is expected to bring in more than $1 million in economic benefit to the city. Even though Wizard World events are scheduled later this year in Midwestern cities like Chicago, Des Moines and Minneapolis, fans from those areas are expected to trek here, too.
Local businesses and events that cater to geek culture are banking on more than a short-term economic bang from the teeming mobs of fans, cosplayers and autograph hounds. If Wizard World succeeds here, it may also boost the profiles of events like Pulver's Geek.Kon.
"Much of the fun comes in meeting people who share your interests, seeing some amazing costumes, and just being able to indulge in your fandom for a weekend," he says. "That's something that Geek.Kon and the other conventions in the area provide every year," says Pulver.
Simon Tsang, the owner of the west-side collectibles store Geek Plastiq, is one of the guys who also stand to benefit from the bump. "The thing about con culture is that it's becoming bigger," he says. "I don't see it going away after Wizard World leaves town. Geeks are always going to be there, wanting what's new."
They'll be here, sure -- but will Wizard World? Macaluso talks about "building a foundation in Madison," but won't confirm anything at this point, other than to say "we're working on it." Given that nearly 40 major superhero and sci-fi movies will hit screens in the next five years alone, Madison's geeks are hoping Wizard World will turn out to be more -- maybe even much more -- than a one-time fling.
Michael Rooker began his Hollywood career playing a stone-cold psychopath in Henry: Portrait of a Serial Killer. Now that he's starred as the blue-skinned, scenery-chomping bounty hunter Yondu in Marvel's Guardians of the Galaxy and the hateful Merle Dixon on AMC's The Walking Dead, he's connecting with a huge geek-culture audience, many of whom will be only too happy to mob him when Wizard World comes to Madison. As it turns out, he has an interesting tie to the Dairy State.
Not everyone expected Guardians of the Galaxy to be such a gigantic hit. What's it like to get swept up into the Marvel-movie rollercoaster?
I think you know you've succeeded when you start being harassed on the Internet. People are stabbing me and harassing me -- finally, I've made it in Hollywood.
Have you been to Wisconsin?
I'm familiar with Wisconsin -- I grew up in Chicago. I have a place in Princeton, Wis. It's an old brewery. Don't know what I'm going to do with it yet, but it's kind of a cool little space. It's called Tiger Brew. It went out of business many, many years ago.
Do you, Rebecca Romijn and Jennifer Lawrence have a support group for actors recovering from blue body paint?
You know, mine wasn't bad. It was just my upper torso and my arms; all the rest of my body was covered with wardrobe. A lot of people had it a lot worse than me.
Which was the more bizarre role for you -- Henry, Yondu, or the heavy metal zombie Viking in the videogame Lollipop Chainsaw?
The voiceover stuff is really cool, and I enjoy doing it. I did all the motion-capture work for the other couple of games I was part of. I just finished doing voiceover work for Finelli, a shotgun company. Primarily, I do film work. The Walking Dead, that was huge for me. I don't usually do ongoing characters for TV. It was a wonderful project to get into.
What's the weirdest thing to happen to you at a Wizard World event?
I don't know what's "weird." Wizard World events are very family-oriented. The fans come from all over the place. You don't get a lot of weird stuff. It's fun, the people dress up, cosplay. It's like a state fair atmosphere. We all have a great time. There's food and drink. People just spend their days walking around looking at the artists and celebrities. It's like a big family.
Let's reframe the question. What's the most memorable thing that's happened to you?
I was doing a show -- it wasn't even a Wizard World event -- and literally the lines were out the building and around the block. I thought that was very strange and unusual. I couldn't figure out why there were so many people. Then I realized -- The Walking Dead has so many viewers and became so popular so fast that when it was advertised that I was going to be there, people got out of their living rooms and down to the place.
I bet you sometimes wish you had Yondu's arrow to get you out of these uncomfortable situations.
It's crazy at these places. I don't think Yondu's arrow could have helped out. I'll tell you one thing -- you are absolutely wasted by the end of the day. You are exhausted. You have time to eat, have a nightcap and you're done for.
What's next up for you?
The Marvel guys already announced that Yondu is part of number two. I just finished a movie called King Bolden about the birth of jazz. The story of Buddy Bolden, the grandfather of jazz, the first king of jazz in New Orleans. It's a period piece set in 1905, close to his death. I hope everyone gets a chance to see it when it comes out. |
package SystemDesign.ParkingLot;
import SystemDesign.ParkingLot.Model.*;
import java.util.ArrayList;
import java.util.HashMap;
import java.util.List;
import java.util.Map;
public class ParkingMain {
public static void main(String[] args) throws InterruptedException {
String nameOfParkingLot ="Pintosss Parking Lot";
Address address = Address.builder().city("Bangalore").country("India").state("KA").build();
Map<ParkingSlotType, Map<String,ParkingSlot>> allSlots = new HashMap<>();
Map<String,ParkingSlot> compactSlot = new HashMap<>();
compactSlot.put("C1",new ParkingSlot("C1",ParkingSlotType.Compact));
compactSlot.put("C2",new ParkingSlot("C2",ParkingSlotType.Compact));
compactSlot.put("C3",new ParkingSlot("C3",ParkingSlotType.Compact));
allSlots.put(ParkingSlotType.Compact,compactSlot);
Map<String,ParkingSlot> largeSlot = new HashMap<>();
largeSlot.put("L1",new ParkingSlot("L1",ParkingSlotType.Large));
largeSlot.put("L2",new ParkingSlot("L2",ParkingSlotType.Large));
largeSlot.put("L3",new ParkingSlot("L3",ParkingSlotType.Large));
allSlots.put(ParkingSlotType.Large,largeSlot);
ParkingFloor parkingFloor = new ParkingFloor("1",allSlots);
List<ParkingFloor> parkingFloors = new ArrayList<>();
parkingFloors.add(parkingFloor);
ParkingLot parkingLot = ParkingLot.getInstance(nameOfParkingLot,address,parkingFloors);
Vehicle vehicle = new Vehicle();
vehicle.setVehicleCategory(VehicleCategory.Hatchback);
vehicle.setVehicleNumber("KA-01-MA-9999");
Ticket ticket = parkingLot.assignTicket(vehicle);
System.out.println(" ticket number >> "+ticket.getTicketNumber());
//persist the ticket to db here
Thread.sleep(10000);
double price = parkingLot.scanAndPay(ticket);
System.out.println("price is >>" + price);
}
}
|
<filename>tests/utils.py
import sys
from contextlib import contextmanager
from io import StringIO
import logging
@contextmanager
def captured_output():
new_out, new_err = StringIO(), StringIO()
old_out, old_err = sys.stdout, sys.stderr
try:
sys.stdout, sys.stderr = new_out, new_err
yield sys.stdout, sys.stderr
finally:
sys.stdout, sys.stderr = old_out, old_err
# https://docs.python.org/3/howto/logging-cookbook.html
class LoggingContext(object):
def __init__(self, logger, level=None, handler=None, close=True):
self.logger = logger
self.level = level
self.handler = handler
self.close = close
def __enter__(self):
if self.level is not None:
self.old_level = self.logger.level
self.logger.setLevel(self.level)
if self.handler:
self.logger.addHandler(self.handler)
def __exit__(self, et, ev, tb):
if self.level is not None:
self.logger.setLevel(self.old_level)
if self.handler:
self.logger.removeHandler(self.handler)
if self.handler and self.close:
self.handler.close()
# implicit return of None => don't swallow exceptions
# # Example using the logging context handler:
#
# if __name__ == '__main__':
# logger = logging.getLogger('foo')
# logger.addHandler(logging.StreamHandler())
# logger.setLevel(logging.INFO)
# logger.info('1. This should appear just once on stderr.')
# logger.debug('2. This should not appear.')
# with LoggingContext(logger, level=logging.DEBUG):
# logger.debug('3. This should appear once on stderr.')
# logger.debug('4. This should not appear.')
# h = logging.StreamHandler(sys.stdout)
# with LoggingContext(logger, level=logging.DEBUG, handler=h, close=True):
# logger.debug('5. This should appear twice - once on stderr and once on stdout.')
# logger.info('6. This should appear just once on stderr.')
# logger.debug('7. This should not appear.')
|
def uniprot2genename(self, name):
from bioservices import UniProt
c = UniProt(cache=True)
try:
res = pd.read_csv(StringIO(c.search(name, limit=1)), sep="\t")
return list(res["Gene names"].values)
except:
print("Could not find %s" % name) |
A study of Scientists led by the University of Cambridge and Banaras Hindu University has suggested that Climate change contributed in the fall of the Bronze Age Indus Valley Civilisation (also known as the Harappan Civilisation), which spanned across present India and Pakistan. The study has revealed that the Bronze Age megacities declined during the 21st and 20th centuries BC and never recovered back following a series of droughts that lasted for about 200 years in the zone of Indus valley.
To study on the concept of the collapse the team of British Scientists studied snail shells that were preserved in the sediments of an ancient lake bed Kotla Dahar in Haryana. They anlysed oxygen isotopes of the shell and calculated the amount of rain that happened in the lake thousands of years ago.
The results of the project shed a light on the mystery for sudden disappearance of the major cities of the Indus Valley Civilisation. The study has also linked the decline with the documented global scale climate event and its impact on other civilisation of Old Kingdom in Egypt, the early Bronze Age civilizations of Greece and Crete as well as the Akkadian Empire in Mesopotamia.
Scientists confirmed that climate change as the reason for collapse of the civilization by studying the evidences from Arabian Sea, Oman and Meghalaya and say that unexpected weakening of the summer monsoon affected northwest India 4100 years ago.
The study also involved researchers from Imperial College London, the University of Oxford, the Indian Institute of Technology Kanpur and the Uttar Pradesh State Archaeology Department.
The study was funded by the British Council UK-India Education and Research Initiative to investigate the archeology river systems and climate of north-west India using a combination of archaeology and geo-science. The study was published in the journal Geology.
Background
Ancient Indus Valley Civilization city also known as the Mohenjo Daro or Mound of the Dead flourished between 2600 and 1900 BC. It was one of the first world and ancient Indian cities. The site was discovered in the 1920s and lies in Pakistan's Sindh province. The civilisation is considered almost as old as those of Egypt and Mesopotamia. |
Francis and the Lights - "ETC"
Francis, Francis, Francis. Where have you been?
It’s been almost two years since we have had anything new from Francis and the Lights. The light-on-his-feet-crooner (watch this video right now if you don’t know what I’m talking about) is back with a beautiful taste of what I hope is a lot more to come. “ETC” is simple. Echoing hand-claps and beautiful, airy vocals come together to form the dreamy track. It’s easy to get lost in Mr. Starlight’s voice as it swirls over the beat and he begs the question, “Are we dreaming?” The song has an astral vibe, which couldn’t be more fitting considering the front man’s name. Close your eyes and you’ll gladly lose yourself with “ETC”.
The track is being released by UK indie label Good Years. It is available for stream and download below. |
<reponame>zhangkn/iOS14Header
/*
* This header is generated by classdump-dyld 1.0
* on Sunday, September 27, 2020 at 11:54:43 AM Mountain Standard Time
* Operating System: Version 14.0 (Build 18A373)
* Image Source: /System/Library/PrivateFrameworks/MediaFoundation.framework/MediaFoundation
* classdump-dyld is licensed under GPLv3, Copyright © 2013-2016 by <NAME>.
*/
#import <libobjc.A.dylib/MFStateDumpable.h>
@class NSDictionary;
@interface MediaFoundation.InternalPlaybackStackController : _UKNOWN_SUPERCLASS_ <MFStateDumpable> {
delegate;
currentQueueControllerItem;
maximumPlayerQueueLength;
playerController;
queueAssetLoader;
queueController;
reporter;
errorController;
externalPlaybackController;
backgroundTaskController;
}
@property (readonly,nonatomic) NSDictionary * stateDictionary;
-(NSDictionary *)stateDictionary;
@end
|
Depletion of Kinesin 5B Affects Lysosomal Distribution and Stability and Induces Peri-Nuclear Accumulation of Autophagosomes in Cancer Cells Background Enhanced lysosomal trafficking is associated with metastatic cancer. In an attempt to discover cancer relevant lysosomal motor proteins, we compared the lysosomal proteomes from parental MCF-7 breast cancer cells with those from highly invasive MCF-7 cells that express an active form of the ErbB2 (N-ErbB2). Methodology/Principal Findings Mass spectrometry analysis identified kinesin heavy chain protein KIF5B as the only microtubule motor associated with the lysosomes in MCF-7 cells, and ectopic N-ErbB2 enhanced its lysosomal association. KIF5B associated with lysosomes also in HeLa cervix carcinoma cells as analyzed by subcellular fractionation. The depletion of KIF5B triggered peripheral aggregations of lysosomes followed by lysosomal destabilization, and cell death in HeLa cells. Lysosomal exocytosis in response to plasma membrane damage as well as fluid phase endocytosis functioned, however, normally in these cells. Both HeLa and MCF-7 cells appeared to express similar levels of the KIF5B isoform but the death phenotype was weaker in KIF5B-depleted MCF-7 cells. Surprisingly, KIF5B depletion inhibited the rapamycin-induced accumulation of autophagosomes in MCF-7 cells. In KIF5B-depleted cells the autophagosomes formed and accumulated in the close proximity to the Golgi apparatus, whereas in the control cells they appeared uniformly distributed in the cytoplasm. Conclusions/Significance Our data identify KIF5B as a cancer relevant lysosomal motor protein with additional functions in autophagosome formation. Introduction Lysosomes are membrane-bound dynamic organelles that represent the final destination for endocytic, secretory and autophagic pathways. The physiological importance of lysosomes is highlighted by a number of diseases resulting from defects in the lysosomal biogenesis and function. On the contrary, the enhanced synthesis, trafficking and extracellular release of lysosomal proteases (cathepsins), are important hallmarks of malignancy and associate with the invasive and metastatic capacity of cancer cells. Interestingly, the lysosomal changes associated with immortalization and transformation of cancer cells also sensitize cancer cells to programmed cell death pathways involving lysosomal membrane permeabilization. Once triggered, lysosomal membrane permeabilization results in the release of cathepsins and other lysosomal hydrolases to the cytosol, where they can trigger the mitochondrial outer membrane permeabilization followed by caspase-mediated apoptosis or mediate caspase-independent programmed cell death. Thus, the inhibition of lysosomal trafficking/exocytosis appears as a promising target for cancer therapy. It would not only inhibit the cathepsin-mediated invasion but also obstruct the general trafficking and possibly result in the accumulation of lysosomes destined for secretion and therefore further sensitize cancer cells to lysosomal cell death pathways. This hypothesis is supported by data showing that vincristine, a microtubuledestabilizing anti-cancer drug, not only inhibits lysosome trafficking but also induces a rapid increase in the volume of the lysosomal compartment followed by lysosomal leakage and cathepsin-dependent cell death. Because drugs that disturb the microtubule network show high general toxicity, we speculated that a more specific interference with lysosome trafficking could result in anti-cancer strategies with fewer side effects. Accordingly, we wanted to identify and characterize motor proteins important for lysosome transport in cancer cells. Motor proteins utilizing the cytoskeleton as substrate for movement are divided into myosin motors that move along actin microfilaments and kinesin/dynein motors that use microtubules through the interaction with tubulin for their movement. Motor proteins are powered by the hydrolysis of ATP and convert chemical energy into mechanical work enabling them to move cargo (vesicles, proteins and lipids) over long distances. Microtubule specific motors consist of two basic types of microtubule motors: plus-end motors and minus-end motors, depending on the direction in which they move along the filaments within the cell. The truncated form of the ErbB2 receptor is frequently found over-expressed in breast cancer and its expression and activity correlates with increased invasiveness, motility and poor prognosis. Accordingly, the ectopic expression of DN-ErbB2 in MCF-7 breast cancer cells renders them highly mobile and invasive (Our unpublished observation). Prompted by the finding that the DN-ErbB2-induced invasive phenotype was associated with altered lysosomal trafficking and a several fold increase in the expression and activity of lysosomal proteases, we chose this model system to search for cancer relevant lysosomal motor proteins. We applied a quantitative proteomic analysis on purified lysosomes from DN-ErbB2 MCF-7 and control cells showing that some motor protein levels were significantly up-regulated following DN-ErbB2 induction. Interestingly, we found that DN-ErbB2 increased the expression of kinesin 5B (KIF5B), a motor protein implicated in lysosomal and mitochondrial transport. In line with this, KIF5B mRNA has been reported to be up-regulated in several types of cancer tissues including bladder cancer (GDS1479 record), advanced gastric cancer (GDS1210 record), squamous cell carcinoma (GDS2200 record), sporadic basal-like breast cancer and BRCA1-associated breast cancer (GDS2250 record) (Data obtained from NCBI: http://www.ncbi.nlm.nih. gov/sites/entrez; Gene Expression Omnibus). KIF5B is a Nkinesin (Plus-end motor) belonging to the super family of kinesin-1 molecular motor proteins that together with cytoplasmic dynein is responsible for microtubule-dependent transport of cargo in eukaryotic cells. To elucidate the role of KIF5B in cancer cells we examined its function in various lysosomal pathways including the lysosomal cell death pathway, the resealing response after plasma membrane damage (exocytosis) and macroautophagy. Materials and Methods Cell culture and treatments MCF-7, HeLa and U2OS cells originate from human breast carcinoma, cervix carcinoma and osteosarcoma, respectivly. MCF-7-eGFP-LC3 cell line is a single cell clone of MCF-7 cells expressing a fusion protein consisting of enhanced green fluorescent protein (eGFP) and rat LC3. MCF-7-DNErbB2 and MCF-7-pTRE cell lines are single cell clones of MCF-7 expressing the tetracycline transactivator transfected with pTRE-DNErbB2 and pTRE, respectively. HeLa-LIMP1-eGFP cells are HeLa cells expressing eGFP-tagged lysosome integral membrane protein-1 (LIMP-1) (kindly provided by Dr. J.P. Luzio, University of Cambridge). The cancer cells and their transfected variants were propagated in RPMI 1640 (Invitrogen) supplemented with 6% heat-inactivated fetal calf serum (FCS; Biological Industries) and penicillin-streptomycin. The medium of MCF-7-DNErbB2 and MCF-7-pTRE was further supplemented with 5 mg/ml tetracycline. To induce the DN-ErbB2 expression, tetracycline (5 mg/ml) was removed and the cells were washed 5 times in PBS before plating. All cells were kept at 37uC in a humidified air atmosphere at 5% CO 2. Analysis of lysosome-associated proteins by mass spectrometry using stable isotope labeling with amino acids in cell culture (SILAC) MCF-7-DNErbB2 and MCF-7-pTRE were grown in customsynthesized RPMI 1640 medium with either normal lysine 12C614N2 (Lys0) or isotope labeled L-lysine 13C615N2 (Lys8) (Sigma-Isotec, St. Louis, MO) supplemented with 10% dialyzed foetal calf serum (Invitrogen) for at least 5 cell divisions to fully incorporate the labeled amino acids. Lysosomes were purified by Iron-Dextran (FeDex) fractionation according to a protocol published previously. Briefly, cells (80-90610 6 in total) preincubated with FeDex (8 h) were lysed mechanically in a dounce homogenizer and the light membrane fraction was loaded on a MiniMachs column attached to a magnet (MACS Separator system, Miltenyi Biotec). Lysosomes trapped on the column were eluted in sucrose extraction buffer (250 mM sucrose, 20 mM Hepes, 10 mM KCl, 1.5 mM MgCl 2, 1 mM EDTA, 1 mM EGTA, and 1 mM pefabloc, pH 7.5) by removing the column from the magnet and flush out lysosomes by a plunger. Lysosomes were dissolved and proteins separated by electrophoresis on NuPAGE Bis-Tris 4-12% gradient gels (Invitrogen) and stained with Comassie Blue. Gel slices were cut into small pieces and incubated with 12.5 ng/ml trypsin at 37uC overnight. The resulting peptides were analyzed by liquid chromatography (Agilent HP1100) combined with tandem mass spectrometry (LC MS/MS) using a linear ion-trap Fourier-transform ion-cyclotron resonance mass spectrometer (LTQ-FT-ICR, Thermo-Finnigan). Peak list were extracted using an in-house developed scripts (DTA-supercharge), combined for each gel slides, and used for protein database searches. Stringent criteria were required for protein identification in the International Protein Index database using the Mascot program (Matrix Science): at least two matching peptides per protein, a mass accuracy within 3 p.p.m., a Mascot score for individual peptides of better than 20, and a delta score of better than 5. MS-Quant (http://msquant.sourceforge.net/), an in-house developed software program was used to calculate peptide abundance ratio and to evaluate the certainty in peptide identification. Measurement of cell viability and microscopic analysis Viable cells were measured by their ability to reduce the tetrazolium salt 3-(4,5-dimethylthiazole-2-y)-2,5-diphenyltetrasodiumbromide (MTT; Sigma) to a formazan dye detectable by spectrophotometric analysis in a VersaMax microplate reader (Molecular Devices Ltd., Wokingham, United Kingdom) as described previously. Phase contrast pictures of cell lines were taken with an inverted Olympus IX-70 microscope connected to an Olympus DP70 digital camera. Time lapse microscopy was performed with a Carl Zeiss Axiovert 200M fluorescence microscope using MetaMorph software. Analysis of GFP-LC3 translocation Autophagy was induced by incubating MCF-7-LC3-eGFP cells with 2.5 mM rapamycin (Sigma-Aldrich, St. Louis, MO, USA) for 24 h. The percentage of cells with eGFP-LC3 translocation into dots (a minimum of 100 cells/sample) was counted in eGFP-LC3 expressing cells fixed in 3.7% formaldehyde and 0.19% picric acid (vol/vol) applying Zeiss Axiovert 100 M Confocal Laser Scanning Microscope. Cells with $5 green cytosolic vesicles were considered positive. Measurement of enzyme activities Caspase-3-like (DEVD-AFC, Enxzyme System Products), cysteine cathepsin (zFR-AFC, Enzyme System Products), acid phosphatase and -N-acetyl-glucosaminidase (NAG) activities were determined essentially as previously described. Briefly, the cytoplasmic fraction was extracted with 20-35 mg/ml digitonin and the total cellular fraction with 200 mg/ml digitonin and the rate of the appropriate substrate hydrolysis V max was measured over 20 min at 30uC on a Spectramax Gemini fluorometer (Molecular Devices, Sunnyvale, CA, USA). Lactate dehydrogenase (LDH) activity of the cytosol determined by a cytotoxicity detection kit (Roche) was used as an internal standard. RNA extraction, cDNA synthesis and reverse transcription-PCR (RT-PCR) The RNA was harvested from cell culture with RNeasy columns (QIAGEN) and cDNA synthesis was made with the TaqMan RT Kit (Roche) using oligo-(dT) 16 KIF5C-rev:ACCTCACCCAAACACTCCAG. PBGD-forw: CATGTCTGGTAACGGCAATG; PBGD-rev:AGGGCATG-TTCAAGCTCCTT. Porphobilinogen deaminase (PBGD; PubMed entry BC000520) was used as internal control together with the gene of interest. PCR products were size-separated on a 1,5%-agarose gel containing ethidium bromide, visualized under UV light, photographed using Polaroid film. Subcellular fractionation For density gradient fractionation cells were pooled in ice-cold homogenization buffer (250 mM sucrose, 20 mM Hepes and 1 mM EDTA, pH 7.4) and lysed in a dounce homogenizer on ice. Homogenates were centrifuged and the supernatant spun down at 3000 g for 10 min at 4uC and the pellet was discarded. The supernatant was centrifuged at 17000 g for 20 min at 4uC. Iodixanol gradients were formed by sequential addition of 4, 10, 16 and 24% solutions in homogenization buffer at 25uC for 1 hour, resulting in the formation of a continuous gradient. The final pellet was resuspended in homogenization buffer and loaded onto a continuous 4-24% iodixanol gradient and centrifuged at 20000 g in a SW41Ti rotor (Beckman) for 17 h at 4uC. Gradients were separated into a total of twenty 500 ml fractions, collected from the bottom. The density of each fraction was determined by measuring OD at 244 nm. Cathepsins B/L, N-acetylglucosaminidase (NAG) and acidic phosphatase activities were measured for each fraction after addition of digitonin. Analysis of exocytosis activity upon plasma membrane wounding Membrane wounding by electroporation was performed as described previously. Briefly, cells were suspended in hanks balanced salt solution (HBSS) (Gibco, Invitrogen), subjected to electroporation at 200 V with variable levels of capacitance in a 0.2-cm electrode gene pulser cuvette (Bio-Rad), and incubated for 1 min at 37uC. Cells were then incubated with anti-LAMP-1 (sc-20011, Santa Cruz Biotechnology) antibody on ice for 30 min, washed, fixed, and stained with Alexa Fluor 488 secondary antibodies (Molecular Probes). Flow cytometry on 10000 cells per sample was performed with a FACS (Becton Dickinson) and data were analyzed by CELLQUEST software (Becton Dickinson). To measure ionomycin induced exocytosis activity cells were incubated in HBSS containing 10 mM ionomycin (Sigma). A cathepsin B specific probe, zfr-AMC (VWR International) was added to each well at a final concentration of 100 mM at time 0 and 10 min. The rate of substrate hydrolysis, as measured by the liberation of AMC (excitation wavelength, 400 nm; emission wavelength, 489 nm) at 30uC on a Spectramax Gemini fluorometer (Molecular Devices, Sunnyvale, CA, USA). Results DNErbB2 increases the level of KIF5B in lysosomes MCF-7 breast carcinoma cells expressing amino-terminally truncated constitutively active form of ErbB2 receptor tyrosine kinase (DNErbB2) display a highly motile phenotype characterized by extensive membrane ruffling, plasma membrane projections and scattering of the cells (Fig. 1A). Furthermore, DNErbB2 expression induces the localization of lysosomes to the filopodia (Fig. 1A) and a 3-4-fold up-regulation of lysosomal cysteine cathepsin activity suggesting that DN-ErbB2 changes the lysosomal trafficking and content. In order to identify motor proteins involved in lysosomal trafficking in cancer cells, we compared the proteomes of lysosomes isolated from control MCF-7 and MCF-7-DNErbB2 cells by stable-isotope labeling by amino acids (Lys0/Lys8) in cell culture (SILAC) followed by mass spectrometry analysis. Six myosin motors and one microtubule specific kinesin motor could be detected by this approach as lysosome-associated motor proteins (Table 1 and Dataset S1). The lysosomal association of three of the identified motor proteins (Myosin Ib, Myosin Ic and kinesin heavy chain KIF5B) was upregulated by more than 25% upon ectopic DN-ErbB2 expression in MCF-7 cells. To characterize the functional significance of these three motors with regard to growth, survival and lysosomal distribution we depleted them in MCF-7 and HeLa cervix carcinoma cells by RNA interference. Only the siRNAs specific for KIF5B affected these parameters ( Fig. 2 and data not shown). Thus, we chose to study the role of KIF5B on lysosomal function in more detail. KIF5B is highly expressed in various cancer cells First, we examined the mRNA expression levels of KIF5B in three cancer cell lines (MCF-7, HeLa and U2OS osteosarcoma) and found it to be highly expressed in all three cell lines when compared to other N-kinesins including KIF5A/KIF5C (Kinesin 1), KIF3A (Kinesin 2) and KIF1A (Kinesin 3) (Fig. 1B). We detected only low levels of KIF3A whereas both KIF5A and KIF5C mRNAs were non-detectable in all three cell lines, in agreement with earlier findings suggesting that their expression is limited to neurons (Fig. 1B). In order to challenge the lysosomal localization of KIF5B detected by proteomic analysis of MCF-7 cells, we subjected the light membrane fraction of HeLa cells to a density gradient fractionation and analyzed the different fractions by immunoblotting and lysosomal enzymatic activity measurements. As shown in Figure 1C, KIF5B was exclusively present in fractions containing high levels of lysosomal LAMP-2 protein and lysosomal enzymatic activity markers (cysteine cathepsins, acidic phosphatase and -N-acetyl-glucosaminidase ). It should, however, be noted that fractions containing mitochondrial GRP75 marker protein were partly overlapping with lysosomal fractions and KIF5B. Depletion of KIF5B induces lysosomal leakage and cell death in HeLa cells To elucidate the role of KIF5B in cell growth and survival, HeLa and MCF-7 cells were depleted for KIF5B by RNA interference (Fig. 2A). Interestingly, KIF5B-depleted HeLa cells acquired an elongated cellular phenotype followed by significant growth inhibition and cell death (Fig. 2B-D). KIF5B depletion induced similar but clearly weaker cytostatic/cytotoxic effects in MCF-7 cells and slightly more in MCF-7-DN-ErbB2 cells as analyzed by light microscopy (Fig. 2B). In spite of the induction of significant cell death, only very low caspase-3 like activity was detected in KIF5B-depleted HeLa cells (Fig. 2E). Instead, the KIF5B depleted cells displayed a significant increase in cytosolic cysteine cathepsin activity indicative of lysosomal membrane permeabilization (Fig. 2F). Depletion of KIF5B in HeLa cells induces peripheral aggregations of lysosomes in HeLa cells Since KIF5B deficient mouse extraembryonic cells display perinuclear clustering of mitochondria and decreased acidtriggered trafficking of lysosomes towards the cell periphery, we next studied the distribution of these organelles after KIF5B depletion. In order to investigate the lysosomal distribution, we took advantage of HeLa cells expressing an eGFP-LIMP1 as a lysosomal marker. In KIF5B-depleted HeLa-eGFP-LIMP1 cells the distribution of eGFP-LIMP1-postive lysosomes was dramatically altered from a diffuse perinuclear pattern to large peripheral aggregates (Fig. 3A). These lysosomes appeared in clusters and were actively transported to and from the aggregates as observed by video time-lapse microscopy (Video S1 and S2). Contrary to the lysosomes, the mitochondrial distribution was not affected by KIF5B depletion in HeLa cells (Fig. 3A). Since KIF5B functions as a plus-end motor, i.e. a motor that transports cargo from the centrosome to the cell periphery, the accumulation of lysosomes to the cell periphery in KIF5Bdepleted cells could be due to a peripheral localization of the centrosome or a failure of the lysosomes to fuse with the plasma membrane. In order to test the first possibility, we stained HeLa-eGFP-LIMP1 cells with an antibody against c-tubulin to mark the centrosomes. However, the lysosomal clusters did not accumulate around the centrosomes (Fig. 3A). To examine if KIF5B is essential for lysosomal exocytosis, we applied three different methods (mechanical scratching, electroporation and ionomycin) to induce plasma membrane lesions that trigger Ca 2+ influx and induction of the resealing response that involves exocytosis of lysosomes. A scalpel was used to scratch on a semi-confluent layer of HeLa cells to mechanically induce plasma membrane damage, and lysosomal exocytosis was immediately assayed using an antibody detecting a luminal epitope of lysosomal LAMP-1 on the cell surface. The surface fluorescence of LAMP-1 was significantly increased at the damage site indicative of lysosomal membrane resealing, and an additional co-localization and accumulation of KIF5B at the damage site was observed suggesting that KIF5B is involved in this response (Fig. 3B). Since this method is not suitable for quantitative studies, we used electroporation to induce small hydrophilic pores in the plasma membrane to investigate if KIF5B was essential in the transport process of lysosomes to the damage sites. The method is widely used to introduce proteins and DNA into cells and depends on the cells ability to reseal their plasma membrane after electroporation. HeLa cells were electroporated with increasing capacitance and immediately after they were stained for surface LAMP-1 (Fig. 3D). Quantification of LAMP-1 exposed on the plasma membrane by flow cytometry revealed a detectable level of LAMP-1 on 3,3% of untreated cells. In contrast, when cells were electroporated at 125 and 250 mF, LAMP-1 was detected on the surface in 11,4 and 21% of the cells respectively. Cells depleted for KIF5B and exposed to 125 mF did not display any significant change in surface LAMP-1 as compared to control treated cells exposed to 125 mF (Fig. 3D). Similarly, the ionomycin-induced lysosomal exocytosis of luminal proteases was unaffected by KIF5B depletion (Fig 3C). These data demonstrate that KIF5B is not crucial for the lysosomal exocytosis and plasma membrane resealing. Furthermore, the uptake of Alexa Flour 488-Dextran (10 kDa) was not affected by KIF5B depletion indicating that KIF5B is not required for fluid phase endocytosis (data not shown). Depletion of KIF5B induces peri-nuclear accumulation of autophagosomes Next, we examined if KIF5B plays a role in autophagy, the major lysosomal degradation pathway. For this purpose, we treated MCF-7 cells stably expressing the autophagosomeassociated LC3 protein fused to the enhanced green fluorescence protein (MCF-7-LC3-eGFP) with either rapamycin that induces autophagy by inactivating the mammalian target of rapamycin complex 1 (mTORC1) or concanamycin A that inhibits the vacuolar V-ATPase activity in lysosomes resulting in reduced turnover of autophagosomes as well as induction of autophagosome formation via inhibition of mTORC1. Interestingly, the depletion of KIF5B by three non-overlapping siRNAs significantly decreased the ability of rapamycin to trigger the formation of LC3postive autophagic vesicles (Fig. 4A). This effect was not brought about by changes in the ability of rapamycin to inhibit mTORC1 since the depletion of KIF5B did not have any influence on mTORC1 activity as analyzed by the phosphorylation status of p70 S6 kinase 1 (p70 S6K1 ) (Fig. 4B). To explore the phenomenon further, we followed the autophagosome formation in MCF-7-LC3-eGFP cell treated with rapamycin (not shown) or concanamycin A by time lapse video microscopy for 45 min. Surprisingly, the distribution of autophagosomes was dramatically altered by KIF5B depletion. In KIF5B depleted cells the autophagosomes appeared and accumulated mainly around the nucleus (Fig. 4C-E; Video S3) whereas in cells treated with control siRNA they were distributed diffusely throughout the cytoplasm (Fig. 4C; Video S4). The perinuclear autophagosomes in KIF5B depleted cells were located in close proximity to the golgi apparatus as visualized by staining with an antibody against a trans-Golgi network membrane protein Golgin-97 ( Fig. 4F and Video S5). Since the distribution of Golgi was not affected by KIF5B depletion, these data suggest that KIF5B may transport a component(s) involved in the formation and/or localization of autophagosomes in the cytoplasm. Discussion The data presented here suggest that KIF5B is implicated in several pathways involving lysosomes and that its highly expressed in all cancer cell lines tested, as compared to other N-Kinesins including KIF5A/KIF5C (Kinesin 1 family), KIF3A (Kinesin 2 family) and KIF1A (Kinesin 3 family). The three KIF5 subfamily members (KIF5A, KIF5B and KIF5C) display high similarity in the amino acid sequence and probably share functional redundancy and similar properties. However, at present it is unknown how the individual KIF5 members might contribute or co-operate in various transport mechanisms and why they differ in their expression patterns in neurons and non-neuronal cells. Density fractionation methodology from HeLa cells revealed, that KIF5B is mainly represented in light-membrane fractions containing lysosomes and to some extent in mitochondria. MCF-7 and MCF-7-DN-ErbB2 cells depleted for KIF5B were inhibited in their growth but displayed a less pronounced death phenotype as observed in HeLa cells. Moreover, we were unable to detect any changes in lysosomal or mitochondrial distribution in MCF-7 cells, whereas HeLa cells displayed a distinct change in lysosomal distribution followed by significant death. This discrepancy between the two cell lines may be ascribed to differences in motor protein expression levels and HeLa cells are probably more dependent on KIF5B for plus-end directed motor protein activity. The aggregation of lysosomes in the pericellular area preceding death in HeLa cells suggested that KIF5B might play a role in trafficking lysosomes proximal to the plasma membrane. However, we were unable to detect any reduction in exocytosis activity triggered either by ionomycin, or electroporation induced membrane damage, following KIF5B depletion. Nevertheless, a significant KIF5B translocation was observed, when HeLa cells were exposed to mechanical induced plasma membrane lesions to induce the resealing response and facilitate lysosomal exocytosis. Here, we observed a considerable recruitment of LAMP-1 positive lysosomes to reseal the damage site and significant colocalization with KIF5B. These data signify that KIF5B could play a role in transporting lysosomes to the plasma membrane destined for exocytosis, but functional redundancy probably exists that implicates other motor proteins as well. The aggregation of lysosomes following KIF5B depletion can also be explained by an alternative possibility: that KIF5B besides its role as a transporter plays a role in positioning lysosomes at distinct sites in the cytoplasm (predominantly perinuclear). Accordingly, depletion of KIF5B would liberate lysosomes allowing their transport by other motor proteins including N-kinesins towards the cortical areas of the cell resulting in lysosomal aggregations. This implies that recruitment of lysosomes by KIF5B to microtubules may localize or queue them and not necessarily facilitate long distance travel. The death pathway induced after KIF5B depletion in HeLa cells triggered the aggregation of lysosomes followed by lysosomal destabilization and subsequent release of lysosomal cathepsins to the cytosol. We observed only limited caspase-3 like activation suggesting that the classical caspase mediated death pathway plays a minor role in the death mode observed. Lysosomal cathepsins function as effective mediators of programmed cell death but the pathways leading to LMP are, however, poorly understood. We have recently shown that vincristine, a compound that destabilizes microtubules and is frequently used in cancer therapy, induce dramatic aggregations of lysosomes and induces LMP and death in HeLa cells. The two treatments might have similar consequences for lysosomal distribution resulting in lysosomes that are brought together in an uncontrolled manner inducing aggregations and subsequent destabilization of lysosomes resulting in death. The more pronounced death phenotype observed in HeLa cells when compared to MCF-7/MCF-7-DN-ErbB2 cells could be explained by a higher dependency on proper KIF5B motor protein activity basically to deal with a high metabolic activity and growth rate. Alternatively, MCF-7 cells might encompass more functional redundancy through expression of various motor proteins than in HeLa cells. Most of the knowledge about KIF5 mediated transport is based on studies in neurons where KIF5 family members can transport various cargoes and its activity seems to be dominant over other motor activities, however the data in non-neuronal cells are still limited. Depletion of KIF5B by RNA interference in MCF-7 cells had a surprising impact on autophagosome formation/localization as determined by the translocation of LC3-eGFP to autophagosomes. We estimated autophagic activity by scoring LC3-eGFP positive autophagosomes after stimulation with rapamycin and observed a significant reduction upon KIF5B depletion. This finding prompted us to follow autophagosome formation in real time by time lapse microscopy using higher concentrations of either rapamycin or concanamycin A enabling us to follow the process in a shorter time frame. The remarkable accumulations of autophagosomes around the nucleus in cells depleted for KIF5B suggested that KIF5B could be involved directly in the transport of autophagosomes along microtubule tracks to the cytoplasm. However, our recent quantitative mass-spectometry analysis on membrane associated proteins purified from autophagosomes (from MCF-7 cells treated with rapamycin or concanamycin A) revealed that KIF5B is not directly associated with autophagosomes (MHH, JA, JO, unpublished data). Alternatively, KIF5B might be involved in the transport of critical factor(s) important for initiation of the autophagic process and proper distribution of autophagosomes in the cytoplasm. Accordingly, removal of KIF5B might prevent the transport of initiation factor(s) and result in formation of autophagosomes accumulating close to the microtubule organizing center and nucleus. Since the autophagosomes appeared mainly at one site of the nucleus (Fig. 4C and Video S3) and close to the trans-golgi network ( Fig. 4F and Video S5) suggests, that the putative initiation factor(s) transported by KIF5B could be golgi derived and affect the distribution of autophagosomes after KIF5B depletion. In addition, KIF5B depletion did not have any influence on the activity of mTOR (as determined by the phosphorylation status of p70 S6 kinase) indicating that the motor protein is acting down-stream of mTOR. Our data demonstrate that KIF5B is highly expressed in cancer cells and plays a significant role in growth and survival of HeLa cells. Additionally, we show that KIF5B is very abundant at the site of plasma membrane damage co localizing with lysosomes destined for exocytosis although it's not essential suggesting that functional redundancy probable exist. Moreover, we provide data indicating that KIF5B is involved in the initial formation/ localization of autophagosomes and might transport component(s) important for the autophagic process. Supporting Information Dataset S1 Motor proteins associated with lysosomes in MCF-7-deltaN-ErbB2 compared to control MCF-7 cells as quantified by SILAC mass spectrometry. |
/**
* A decorator for a {@link SubstitutionMap} that records which values it maps.
*
* @author bolinfest@google.com (Michael Bolin)
*/
public class RecordingSubstitutionMap implements SubstitutionMap.Initializable {
private final SubstitutionMap delegate;
private final Predicate<? super String> shouldRecordMappingForCodeGeneration;
// Use a LinkedHashMap so getMappings() is deterministic.
private final Map<String, String> mappings = Maps.newLinkedHashMap();
private RecordingSubstitutionMap(
SubstitutionMap map, Predicate<? super String> shouldRecordMappingForCodeGeneration) {
this.delegate = map;
this.shouldRecordMappingForCodeGeneration = shouldRecordMappingForCodeGeneration;
}
/**
* {@inheritDoc}
* @throws NullPointerException if key is null.
*/
@Override
public String get(String key) {
Preconditions.checkNotNull(key);
if (!shouldRecordMappingForCodeGeneration.apply(key)) {
return key;
}
if (delegate instanceof MultipleMappingSubstitutionMap) {
// The final value only bears a loose relationship to the mappings.
// For example, PrefixingSubstitutionMap applied to a MinimalSubstitutionMap
// minimizes all components but only prefixes the first.
// We can't memoize the value here, so don't look up in mappings first.
ValueWithMappings valueWithMappings =
((MultipleMappingSubstitutionMap) delegate).getValueWithMappings(key);
mappings.putAll(valueWithMappings.mappings);
return valueWithMappings.value;
} else {
String value = mappings.get(key);
if (value == null) {
value = delegate.get(key);
mappings.put(key, value);
}
return value;
}
}
/**
* @return The recorded mappings in the order they were created. This output may be used with
* {@link OutputRenamingMapFormat#writeRenamingMap}
*/
public Map<String, String> getMappings() {
return ImmutableMap.copyOf(mappings);
}
@Override
public void initializeWithMappings(Map<? extends String, ? extends String> newMappings) {
Preconditions.checkState(mappings.isEmpty());
if (!newMappings.isEmpty()) {
mappings.putAll(newMappings);
((SubstitutionMap.Initializable) delegate).initializeWithMappings(newMappings);
}
}
/** A-la-carte builder. */
public static final class Builder {
private SubstitutionMap delegate = new IdentitySubstitutionMap();
private Predicate<? super String> shouldRecordMappingForCodeGeneration =
Predicates.alwaysTrue();
private Map<String, String> mappings = Maps.newLinkedHashMap();
/** Specifies the underlying map. Multiple calls clobber. */
public Builder withSubstitutionMap(SubstitutionMap d) {
this.delegate = Preconditions.checkNotNull(d);
return this;
}
/**
* True keys that should be treated mapped to themselves instead of passing through Multiple
* calls AND.
*/
public Builder shouldRecordMappingForCodeGeneration(Predicate<? super String> p) {
shouldRecordMappingForCodeGeneration =
Predicates.and(shouldRecordMappingForCodeGeneration, p);
return this;
}
/**
* Specifies mappings to {@linkplain Initializable initialize} the delegate with. Multiple calls
* putAll. This can be used to reconstitute a map that was written out by {@link
* OutputRenamingMapFormat#writeRenamingMap} from the output of {@link
* OutputRenamingMapFormat#readRenamingMap}.
*/
public Builder withMappings(Map<? extends String, ? extends String> m) {
this.mappings.putAll(m);
return this;
}
/** Builds the substitution map based on previous operations on this builder. */
public RecordingSubstitutionMap build() {
// TODO(msamuel): if delegate instanceof MultipleMappingSubstitutionMap
// should this return a RecordingSubstitutionMap that is itself
// a MultipleMappingSubstitutionMap.
RecordingSubstitutionMap built =
new RecordingSubstitutionMap(delegate, shouldRecordMappingForCodeGeneration);
built.initializeWithMappings(mappings);
return built;
}
}
} |
def migrate_repo(self, auth, clone_addr,
uid, repo_name, auth_username=None, auth_password=None,
mirror=False, private=False, description=None):
data = {
"clone_addr": clone_addr,
"uid": uid,
"repo_name": repo_name,
"mirror": mirror,
"private": private,
"description": description,
}
data = {k: v for (k, v) in data.items() if v is not None}
url = "/repos/migrate"
response = self.post(url, auth=auth, data=data)
return GiteaRepo.from_json(response.json()) |
Mike Jacobs, a facilities employee with the Palo Alto Unified School District, talks about how the emergency response system he invented, Safeguard Notifier, works by emitting a loud alarm and displaying one of three LED lights that that are used to communicate the safety status of classrooms to first responders. Photo by Veronica Weber.
In the event that a shooting were to take place at one of Palo Alto Unified School District's 17 campuses, teachers would follow a long-held safety procedure: Rush to tape a color-coded piece of paper to the outside of their classroom doors to indicate the safety or lack thereof inside the room (green means "all are safe and accounted for," red means "immediate help is needed"); close the doors (they lock from the outside); turn off the lights and gather quietly with their students to wait for law enforcement to arrive. They might think to call 911; they might not.
To Mike Jacobs, a Palo Alto parent and longtime Palo Alto Unified maintenance employee, this process is archaic, inefficient and unsafe.
"In a day and age like today where we have so much technology at our fingertips, it's not practical," Jacobs said. "We can make these (procedures) so much better."
And so, this spring, Jacobs founded Safeguard School Systems, hoping to bring school safety into the 21st century with the invention of a device that streamlines communication in the event of an on-campus emergency.
The Safeguard Notifier, a discreet, small black box that Jacobs hopes will be mounted outside classrooms above their doors, is equipped with a small LED light to replace the color-coded card system. It similarly flashes green if all inside are safe; blue if not everyone is accounted for (for example, a student may have gone to the restroom before the campus lockdown began); red if assistance is needed. The red light is accompanied by a loud Piezo horn that can immediately alert law enforcement where they need to go first.
The second the device is activated teachers do this with the push of a single button on a key fob that they can hang from a lanyard or key chain multiple things happen. A call is automatically placed to 911. The door is locked with a 1,200-pound electromagnetic door locking system (that weight is the minimum). Schools can opt to have cameras installed in classrooms; if so, they are also instantaneously activated to provide first responders real-time monitoring.
"To think that we can close the door, turn off our lights, lock everything up and act like no one's home and think that that's sufficient enough, it's not," Jacobs said. "What we're doing is we're leaving our law enforcement professionals in the dark. When they arrive at the scene, they don't know what to do. There's no communication. They don't know if the shooter is in a room; they don't know if the folks in the classrooms are OK."
Jacobs, a father of three young children, said in the wake of the December 2012 Sandy Hook Elementary School shooting, he felt compelled to do something to improve school safety.
"I just thought to myself, 'Something's got to change,'" he said.
So Jacobs, a maintenance/operations foreman with expertise in the electrical field, invented the simple emergency-response device. He has spent time talking to local law enforcement, teachers, administrators and students about what they would like to see in such a system to ensure it meets all of their needs.
"What are the challenges that they're facing? What are some of the things they don't like about the procedures that are in place? One of the common problems that teachers have with current procedures is that they do have to run to the door and lock the door themselves," Jacobs said.
Jacobs is continuing these discussions and said he's meeting with a recently retired Bay Area police chief next month to discuss a potential partnership.
Much of Safeguard's work is also informed by a 2004 school safety report issued by the U.S. Secret Service and the U.S. Department of Education on preventing school violence. A major takeaway from the report is that schools should no longer approach school safety passively, said Safeguard partner and advisory board member Robert Gonzalez. Gonzalez, a corporate recruiter and father with a military background, is a longtime friend of Jacobs. He said knowing Jacobs as someone who works passionately with and for youth, from participating in youth outreach programs to working with troubled teenagers, made him immediately want to join the Safeguard team.
"Eventually, when a teacher communicates the condition of their room, that specific condition notification will reach the 911 dispatch level," Jacobs said. "Dispatch can then communicate those conditions to officers without even having to speak to teachers directly. Those officers will also be able to log into our software directly from their squad cars and make risk-informed decisions with real-time visuals into each area. The more confusion that we're able to eliminate, the less time an attacker has to hurt someone."
The company has yet to officially approach Palo Alto Unified, or any district, to roll out the Safeguard Notifier. The company is still "in the seed stage of the seed stage," Gonzalez joked, and struggling to find funding.
"We want to take this and we want to be able to productize it," Gonzalez said. "The major hurdle that we're having right now is investors."
The company launched an ambitious crowdfunding campaign on Indiegogo early last month, hoping to raise $150,000 by Dec. 5. The funds would help the company expand its hardware and software capabilities and pay for a pilot installation at a Bay Area school, a test they want to do free of cost but which would involve costs for manufacturing, installation, operating, permits and more.
Jacobs and Gonzalez see themselves as pioneers in the still emerging field of school-safety technology. Across the nation, some companies and schools have looked at implementing automated communication systems. Just this week, 20 public schools in New York tested out a system that links school radios, phones and mobile devices to emergency-dispatch systems and allows school officials to instantly share video, audio or other data with law enforcement. Other organizations are checking out smartphone applications that target emergency response. But Jacobs and Gonzalez see Safeguard Notifier as a more comprehensive, all-encompassing system, and they said they are ready to roll it out so that kids are as safe as possible.
"We want to take this (to schools)," Gonzalez said. "It's real. ... We want to bring it out to the market and make it a reality for schools."
We still don't know what worked in our emergency preparedness when the Challenger School had an incident a few weeks ago. Surely even a mistake can be a useful tool in assessing the readiness of our police and local schools?
The biggest problem we know about is that there was no way to control regular traffic from just getting in the way. Have any changes in procedures been made by PAPD as a result?
Or are we still not allowed to know anything about this event?
91 School Shooting since Sandy Hook. Schools have to prepare because it's not if, but when. Our system of government has failed to keep our schools safe.
This system, as described above, is incredibly complicated, dependent on grid power, or a significant battery for power, and would require constant testing to insure that it would be functional all of the time.
With schools now virtually all using the wireline/wireless Internet, and in-class telephones/cell phones pretty ubiquitous--got to wonder if there isn't a simpler system that could be devised?
Those who attack schools, know the schedule and have studied their plan of attack. Our district current plan, to turn lights off and pretend that no one is home, will not work.
With today's advanced technology we need more, we need to safeguard our schools, a green paper and a peace of tape will not do. It is our duty and schools duty to help along Safeguard Notifier to make schools safer for our children. Let us make it happen.
I will do my part, I have 4 grandchildren in the district.
Technology is a double edged sword when it comes to security.
For example, people might think it would be a good idea to have cameras on campus so the main office could easily monitor any security issues.
However, if the intruder makes it into the office and uses the system- they can now survey the entire school and know when, where and how the first responders are closing in.
I suspect it is merely a matter of time before we have a school shooter on one of the PAUSD campuses. Especially in light of all the unresolved bullying that goes on in the middle schools and high schools. As in the Columbine shootings, for example, in which two long-bullied kids vowed to " get even" for years of peer abuse.
Safeguard has a unique approach. As a parent of school aged children, I think it's nice to hear about someone "thinking outside the box" on this issue.
With all the options the system seems to provide, it would make sense for all schools to at least automate the color coded card procedure. It's simply taking an old procedure and making it better.
ANY use of technology can be a double edged sword, but its also important to note that no safety plan is full proof. Even the most guarded facilities in America have proven to be vulnerable.
The world of technology isn't slowing down and it's expanding into every field imaginable. Why not school safety?
I think there is a very poor safety plan at our schools, although I could be wrong.
The reason I am saying this is because whatever plans are there are a big secret. There are obviously some good reasons for keeping a lot of this secret, but there are some things that parents should be aware of and this just isn't the case.
They have various drills in school, but these are all done in class time. What would happen if there was an emergency 15 minutes before or after class, or in a passing period, or brunch, or lunch? Do the schools know what would happen then? Do the students?
There is no way of preventing students from turning up at school with a 15 minute warning. Text alerts to all staff, parents and students are not available. The schools answer is to check emails or Infinite Campus is not practical in an emergency. We need to have text alerts and these need to be emergency only and not for non emergencies because once that starts happening they will be ignored.
Our schools cannot be put on emergency lockdown efficiently as there are too many entrances and there are no gates that can be shut. Likewise, there is no way of knowing exactly who is on campus at any one time due to the open campus policy and although we don't want to necessarily end this policy, we do need to have some technological method of accounting for all students at any one time. I should think that some type of technology could be used for immediate check in for all students in an emergency could be invoked.
Particularly now with the construction still ongoing, it is not easy to spot a suspicious person on campus. Even bikes can be stolen during class time, so getting some security cameras at strategic points around the campus, particularly at street entrances to the campus, would make a lot of sense and would probably help the bike thefts problem too.
It is about time that the security of our schools was discussed by the community as a whole. It doesn't just have to be an unthinkable tragedy, but something as simple as a fire or similar that should be planned for.
Mike Jacobs shouldn't rely on a crowdfunding campaign on IndieGoGo to fund this effort.
He wants to raise $150,000 by asking people on the Internet for money. So far, he's raised $356 (as of Thanksgiving Day) and has nine more days to go to raise the full amount.
The way to go to build this device is to raise capital from savvy investors, whether they be angel investors or venture capitalists.
They will ask hard questions. They'll force Jacobs and his people to put a lot more thought into both their product and into the type of environment in which they intend it to operate.
From the description that Elena Kadavny provides in this story, I can immediately see at least ten major issues that would plague the development and especially the operation of this device.
Savvy investors would immediately raise these questions, and many more, before choosing to invest.
Crowdfunding this type of product development doesn't make sense. I'd advise Jacobs to raise money in a more traditional fashion. He is after all in the Bay Area, and real investors grow on trees around here.
I would advise the inventor to keep doing what he's doing. He should try every option available to raise funds. Eventually, someone who knows the ropes and who genuinely cares about school safety will partner with this young man to make it happen.
As far as the US school security integration market is concerned, it's projected to be 4.9 billion dollar market by 2017. Maybe this young man has done some thinking already...who are we to say? I'm sure he's more concerned about school safety though.
It is important to note that no one has died in a school fire for more than 56 years. That's something widely known in school safety today. That record is something to work with--fire safety is a combination of technology(complex fire alarm systems), practice and education. It's a successful framework that has cost our district (tax payers) millions of dollars to implement, but a framework that could potentially save the lives of our children.
I'm sure that our "lockdown" drills work beautiful in practice, but no one anticipates how quickly an emergency unfolds. Why don't we have systems like Safeguard in place already? Our firemen can respond to a school fire and know exactly where to go because of the systems we have in place. It seems that this inventor is trying to accomplish the same for the OTHER half of our first responders. I for one am all for it. I sincerely hope that this company manages to get off the ground and that the inventor has his intellectual property rights in place. Finally, someone gets it!
It's about time that someone has finally came up with a product that can improve the safety of our kids. It's been far to long that the current system has been out dated and not up to standards in protecting our kids. We have fire protection throughout all our schools and this equipment with fire drills protects all the people within that school. It would be truly sad if we just sit and do nothing when we could of had this equipment installed to save our kids lives. Nothing is full proff in life but with this system we stand a better chance in protecting our kids. Our kids have a right to feel safe in schools and not have to worry about being hurt or killed on campus. As for the office monitoring and spying on each classroom, from what I read about this product, the teachers in each classroom has the control to turn the system on in an emergency, as well as staff that might be in a hall, cafitetia, front office but it's only in the area where the problem is. This system is amazing. The teacher activates one of the three buttons on the key fab which automatically locks the door of her classroom, protecting the kids and the teacher, with that same press of the button it notifies the police, the police can see in the classroom by their laptops mobile phone before even arriving at the school, causing less confusion and wasted time trying to figure out where the problem is in the school.
The system also lights up above each classroom door, if it lights up green that means everything in the classroom is safe, if it lights up blue, there are students that are out of the classroom, if it lights up red the the police can see the classroom right away and know that's where the problem is. This product seems to be making progress with law enforcement officials taking out the guess work, saving valuable time, and the response time would be greatly improved. Some have sugested this system needs a lot of maintenance and very costly, but from what I have read it's very cost affective, way cheaper then installing a fire alarm system in a school, and it only needs once a year maintains. Even a fire alarm system requires maintains to keep it up and running. Fire marshals check systems regularly weather it's schools, office buildings all safety required buildings.
This system will protect lives, the lives of this countries future. Glad someone has finally taken the time to make this product to protect our kids. |
/**
* Reset status fields to UNCHANGED.
*
* @param newRecord -- the new record created.
*/
public static void resetStatus(BibliographicRecord newRecord) {
newRecord.getFields().forEach(field -> {
if (Global.MANDATORY_FIELDS.contains(field.getCode()))
field.setMandatory(true);
field.setFieldStatus(Field.FieldStatus.UNCHANGED);
});
} |
Old Wine in New Bottles they are in danger again, from this same insidiousspirit, ofhigh-mindedly persecuting "for the name" ofworking up a general and undiscriminating hostility to groups whose very designations become symbols of all unrighteousness. It should not be assumed that the World Council acts in this matter for those historic Churches which, while unequivocal in their rejection of racialism, are guided in determining their policies by the established principles of proportion and moral discrimination. These principles are exemplified, in fact, in a pamphlet just come to hand, Violence and Oppression: A Quaker Response; its practical wisdom derives from the reality of Quaker experience.' |
import sys
if sys.version_info >= (3,10):#pragma: no cover
from importlib.metadata import entry_points, version, PackageNotFoundError
elif sys.version_info >= (3,8):# pragma: no cover
from importlib.metadata import version, PackageNotFoundError
def entry_points(*,group):
import importlib.metadata
return importlib.metadata.entry_points()[group]
else:
from importlib_metadata import entry_points, version, PackageNotFoundError |
For playwright Sina Skates it’s all about turning work into play.
“I get paid to play all the time and that’s kind of awesome,” she said.
And, playing is exactly what Skates does with the new production "Little Miss Muffet and the Lost Sheep," which is currently showing at the Birmingham Children's Theatre. The play, aimed at a pre-K audience, will hold its first family day performance Saturday, Feb. at 10 a.m. and noon.
It will also run Feb 10., April 12 and April 26. The play will run on school days until Feb. 21 at 9:30 and 11 a.m. Tickets are $9 for children and $11 for adults.
She said the production teaches children how to face fear through the journey of Little Miss Muffet who runs away once she encounters the famous spider from the original nursery rhyme. Throughout her journey, she runs into Little Bo Peep looking for her sheep and Little Boy Blue who wants to be a jazz musician. All three characters learn important lessons concerning staying true to oneself.
Skates said she presented the idea to the Theatre a year ago and spent time working in conjunction with BTC and composer Jay Tumminello to flesh out the tale based on the original nursery rhymes.
Skates has been writing since she was a child. In addition to writing children’s productions, she teaches dance and also writes plays for adults.
She said her work with BCT gives her a chance to reach young children and teach them lessons that she shares with her own three kids.
Children aren’t the only one who will get enjoyment from the production, according to her.
Currently, Skates is working on another play that addresses bullying.
“I think that for me, those are the kind of messages and themes that I want to tackle,” she said. |
Measles in infants during the first year of life Aim of the research to study epidemiological data, clinical symptoms and the course of measles among hospitalized infants during the first year of life in Infectious Diseases Communal Clinical Hospital, Lviv city. Material and methods. In 2018, 235 infants with measles aged from one month to one year were treated in an inpatient department of Infectious Diseases Communal Clinical Hospital. Diagnosis of measles was based on typical clinical manifestations of the disease, data of epidemiological anamnesis, common laboratory (hemogram, urinalysis), and immunological investigations (determination of the level of specific IgM class antibodies in blood serum). Results. Among hospitalized infants in the first year of life with the diagnosis measles to Infectious Diseases Communal Clinical Hospital, city residents prevailed 74.9%. Most frequently, the disease cases were recorded in 7-9-month-old (33.6%) and 10-12-month-old infants (29.4%). In 87.7% of patients, the sources of infection were school age children and ill parents. Incorrect diagnoses at pre-hospital stage were established in 22.1% cases. In 71.1%, infants were admitted to Infectious Diseases Communal Clinical Hospital before the rash originated, in 28.9% during rash period. A clinical picture of measles was characterized by typical course with minimal catarrhal manifestations in early period of the disease. Signs of conjunctivitis and blepharitis intensified in 34.9% of children from the onset of rash. Kopliks spots, as a rule, were atypical and observed only in 79.6% of patients. Moderate forms of the disease prevailed (70.2%). Complications developed only in patients with severe forms of the disease and burdened premorbid background. Among complications, pneumonia (5.5%) and obstructive bronchitis (2.1%) dominated. Conclusions. Measles in infants during the first year of life demonstrates all typical features and is characterized by the prevalence of moderate forms. In mild forms of the disease, clinical symptoms are not evident, which enables to compare it to mitigated measles. Severe forms with the development of complications on the part of the respiratory tract were observed in 7.7% of children, usually with premorbid background. Rapid increase in the number of infants with measles in the first year of life is a predictable consequence of the situation, which occurred due to incomplete conduction of prophylactic measures aimed at prevention of measles, primarily, non-compliance of vaccination schedule for children and mandatory conduction of anti-epidemic measures in the focus of infection. |
#include <stdio.h>
#include <math.h>
int main()
{
int t;
scanf("%d",&t);
for (int iCase=0;iCase<t;iCase++)
{
int n,m;
scanf("%d",&n);
m=ceil(log((double)n)/log(2.0));
printf("%d\n",m);
}
return 0;
} |
VSEC-LDA: Boosting Topic Modeling with Embedded Vocabulary Selection Topic modeling has found wide application in many problems where latent structures of the data are crucial for typical inference tasks. When applying a topic model, a relatively standard pre-processing step is to first build a vocabulary of frequent words. Such a general pre-processing step is often independent of the topic modeling stage, and thus there is no guarantee that the pre-generated vocabulary can support the inference of some optimal (or even meaningful) topic models appropriate for a given task, especially for computer vision applications involving"visual words". In this paper, we propose a new approach to topic modeling, termed Vocabulary-Selection-Embedded Correspondence-LDA (VSEC-LDA), which learns the latent model while simultaneously selecting most relevant words. The selection of words is driven by an entropy-based metric that measures the relative contribution of the words to the underlying model, and is done dynamically while the model is learned. We present three variants of VSEC-LDA and evaluate the proposed approach with experiments on both synthetic and real databases from different applications. The results demonstrate the effectiveness of built-in vocabulary selection and its importance in improving the performance of topic modeling. Introduction Topic modeling was originally developed for organizing, interpreting and summarizing large collections of unstructured textual information in document analysis. Typical formulations include Latent Semantic Indexing (LSI), Probabilistic Latent Semantic Indexing (PLSI), Latent Dirichlet Allocation (LDA) and some related approaches. Over the years, it has been generalized to various applications with other data types, with different motivations and modified modeling techniques. For example, it can be used for analyzing cross-collection textual data, paired text and image data, and biological or medical data. Nevertheless, the basic idea of topic modeling remains the same, which is to find the hidden structure of the given data by discovering the latent topics. The latent topics are defined as collections/distributions of co-occurring (general-sense) words. In general, a topic model is a generative hierarchical probabilistic model, and exact inference is intractable for raw data from real applications except for trivial cases. Therefore, in real applications, raw data are typically prepossessed to create conditioned data with some simplifying properties: real-valued feature vectors may be discretized and represented by indices from a finite set; and a finite vocabulary needs to be determined where each word should have sufficient occurrences in the data to support distribution-based modeling and analysis. For textual documents, typically a vocabulary is formed by counting the number of occurrences of each word and removing a standard list of stop-words and other words that rarely appear in the corpus. For visual documents (images), feature representations such as color histograms, SIFT descriptors, etc., are extracted first, then followed by a proper discretization process like k-means clustering, to form a raw vocabulary with a reasonable size. The clusters with very few samples may be removed to yield the final vocabulary. As such, the visual words of each image consist of indices corresponding to the final set of k-means centers. Similar steps are followed for processing biological documents like neuroimaging data, where functional regions are parsed and then represented by a list of terms from which the vocabulary is built. However, such general techniques for pre-processing the raw data are independent of subsequent topic modeling, which raises the question of whether the discovered topic model is heavily biased by the predefined vocabulary, not reflecting the true latent semantic structures we set out to find. In practice, there are other unanswered questions, like, what is the reasonable size for a vocabulary? what is the optimal set of words to use for a given vocabulary size? etc. Ultimately, it appears that a better solution would be to address the topic modeling and the vocabulary-forming tasks jointly in order to obtain optimal results. To this end, in this paper we propose a new topic-modeling approach termed Vocabulary-Selection-Embedded Correspondence-LDA (VSEC-LDA), which simultaneously addresses the task of latent structure discovery and the task of selecting a suitable vocabulary for supporting model-learning. Both synthetic and real data are used to evaluate the performance of VSEC-LDA. The improved performance of our approach in modeling and processing image-text documents and related tasks not only indicates that a better latent structure can be learned by our model but also shows potential of our model for diverse applications. Related Work In recent years, a lot of efforts have been made to generalize original topic models to various applications. Many of them focus on modifying the model structure to better fit the specific data arising from the target applications. Generally, the modification may happen in three different forms: introduction of proper priors or topic-word distributions, utilization of multi-modal topic models, and a combination of the above two. For the first form, develops the correlated topic model, which replaces the Dirichlet prior with a logistic normal distribution. As an extension, replaces the Dirichlet prior with a non-parametric setting. Additionally, proposes the nested Chinese restaurant process (nCRP) model by introducing an nCRP prior to a tree. Furthermore, introduces an additional prior to the topic-word distribution to discover interpretable topics, and proposes a words embedding space and uses the cosine similarity between the word and topic vector as the criteria to pick the word which yields better topic cohesion. Some other works introduce constraints to the topicword distribution so that the model can capture the sharing aspect among the topics and/or the words. The above works have showed their valuable usage in modeling crosscollection textual data. For the second form, some attempts have been made for better fitting data from various sources while capturing the correlation among different modalities. Corr-LDA models the conditional correspondence between visual and textual words through discrete indexing that links topics of textual words to visual words. In, the model learns a regression from the topics in one modality to those in the other, thus the topic correspondence between different modalities is based on the sets of topics describing each modality rather than one-to-one relationship. The idea has been extended to the scenario where different modalities only have loose correspondence by introducing Markov random field to topic model. In a separate line of work, such dual-modal LDA models have been generalized to multi-modal LDA to allow learning from multiple views. While the previous efforts are comprehensive and have provided valuable insights, many approaches also have their limitations. For example, regardless of different model structures, they typically do not consider the construction of vocabulary and modeling at the same time. Instead, for continuous data, they first extract a set of descriptors, then a codebook is created by clustering all the local descriptors using k-means, and each descriptor is quantified into a word accordingly. In a separate stage, these discrete words with finite size are fed into the model for learning. Indeed, such a general vocabulary construction pipeline can achieve dimensional reduction, which will alleviate the model workload by quantizing continuous data into discrete format and removing low-frequency descriptors. However, ideally, a model should be able to dynamically select the vocabulary according to its objective. In contrast to previous works, our model incorporates 1) the ability to associate vocabulary construction with model learning, and 2) flexibility in determining the words for each modality. Thus, as we will show, our model not only improves the quality of the estimated topics, but also further reduces computational complexity. Vocabulary-Selection-Embedded Corr-LDA We build our core model VSEC-LDA based on the established Corr-LDA approach, through embedding a dynamic word selection component in latent topic learning. We choose Corr-LDA as the basic building block due to the following two considerations: 1) Corr-LDA has demonstrated performance in many applications with dual modalities and such a dual-modality model can be generalized to multi-modality easily; and 2) Our primary goal is to design an approach embedding vocabulary selection in a typical topic model, not focusing on various topic modeling structures. Due to the page limit, we show the details of two different variants in the supplemental material to further demonstrate the generalization of our method. Our work is based on the assumption that some words in the vocabulary may contribute to the latent model differently than some other words, and possibly some are even detrimental to learning a meaningful topic model, especially if the learning starts from a general-purpose large vocabulary (which is often the case for applications involving visual words built on top of visual features). In this section, we first introduce the architecture of the core VSEC-LDA model, and then describe the word selection and learning process, and at last discuss the tunable threshold in the model. A summary of the notations used throughout the paper is provided in Table 1. Figure 6 shows the graphical representation of our proposed VSEC-LDA model (a higher-resolution version is Relevance in this context refers to whether the words contribute to the underlying latent topics to be learned. Textual and visual words are correlated in the latent topic space: Each topic of textual words has a corresponding topic of visual words. The generative process of relevant words in our model is similar to that in Corr-LDA, except that the visual words are drawn from a multinomal distribution parametered by instead of a multivariate Gaussian distribution. As for irrelevant words, we assume they are drawn from an uniform distribution (this is for simplicity of discussion; the only real constraint is that they do not contribute to the latent topics in any significant way). We summarize the gen- erative process of each document d in Model I as follows: The VSEC-LDA Model words, and maximize the likelihood of training data regardless of the presentation of the irrelevant words. The joint distribution of the visual / textual words and the latent topics is: p(y n |v)p(w n |y n, )). Learning the Model Since exact inference for the model is intractable, one can rely on one of the usual approximate inference methods, such as variational inference, expectation propagation, or Gibbs sampling. We use Gibbs sampling for its benefit in avoiding local optima while yielding relatively simple algorithms. There are three parameters to be inferred, two topic-word distributions and, and one topic proportion for each document. These parameters do not need to be directly updated, because they can be integrated out while sampling the topic indicator variables z m and y n. It is worth emphasizing that the model do not know which words are irrelevant or relevant. Thus, all words are considered in the inference process until after vocabulary selection is conducted. To initialize the model, first we randomly assign visual topic indicator z m and textual topic indicator y n with values drawn from Uniform(1,..., K) for every document d. Then, M d,k and N d,k are initialized via counting the number of times that topic k appears in the visual modality and the textual modality, respectively. After that, M d,c and N d,t are filled with the frequency that word c and word t has in current document d. We also initialize all the entries in the count matrix used during model inference that have a value zero, by setting them to some small number to avoid numeric stability issues. After initialization, the model iterates through the following steps until convergence: 1. Updating hidden topic z and y (a) For each visual word v m in document d, update the corresponding visual topic indicator z m conditioned on 1) the current estimation of topic-visual word distribution, 2) the current assignment of topics z d, 3) the current estimation of multinomial distribution over topics d, and 4) the current estimation of textual assignment of topics y d. (b) For each textual word w n in document d, update the corresponding topic y n, conditioned on 1) the current estimation of topic-textual word distribution, 2) the current assignment of visual topics z d, and 3) the current assignment of visual topics y d. (c) Estimate current model parameters by the following equations, followed by computing negative log likelihood using Eq. and Eq.. Note that a lower score indicates better model fitting. Repeat step (a)-(c) until the negative log-likelihood converges or is lower than the predefined threshold. 2. Filtering non-contributing words (a) Compute the entropy of each visual word v m across all the topics. The probability of a certain visual word being picked by a given topic is derived from the current estimation of topic-visual word distribution, which can be approximated by the normalized counting matrix M ct k (Eq. ). Then, words with greater than predefined threshold entropy (the choice of the threshold will be discussed in the next subsection) are removed from the current vocabulary, and M ct k, M d,c are updated by deducting the count that is associated with the removed visual words. (b) Compute the entropy of each textual word v m across all the topics. The probability of a textual word being picked by a given topic is derived from the current estimation of topic-textual word distribution, which can approximate by the normalized counting matrix N ct k (Eq. ). Then, words with greater than predefined threshold entropy are removed from current vocabulary, and N ct k N d,t are updated by deducting the count that is associated with the removed visual words. 3. Re-estimate hidden topic z and y Repeat step 1 and 2 until change of the likelihood is smaller than and no more word can be filtered. Next, redo step 1 to obtain a better estimation of the hidden topic. This step is not mandatory but was found to be beneficial, since estimating hidden topics and filtering non-contributing words simultaneously might lead to a sub-optimal estimation of the model parameter. Finally, we estimate model parameter using Eq.. Threshold for Vocabulary Selection Intuitively, one can rank the entropy of all the words and set the threshold to the point where there is a sharp change (the cut-off point) in the ranked entropy. However, this may have potential issues if the cut-off point happens to be at the tail of the ranked entropy, which indicates the majority of the words will be removed. Practically speaking, for any general feature selection and clustering problem, it is reasonable to assume that noise cannot dominate over useful data (or a data-driven learning approach may focus on only the noise). Thus we assume that irrelevant words are the minority and accordingly use this to guide the selection of the cut-off point (e.g., it should not remove more than 30% of the words). Further, in our experiments, it was observed that the cut-off point does not change too much from one iteration to another. Therefore, the threshold is determined from the ranked entropy at the first iteration and it will remain the same throughout the inference process. Another practical issue is, if we keep removing words according to the threshold, the model may get trapped into the situation where only few words left and thus the partition of words (according to topics) may more likely become disjoint, which is not a desired property for topic models. In our implementation, each time the vocabulary selection is employed, we allow the model to learn without vocabulary selection till some local optimum is reached, and then invoke an iteration of vocabulary selection. Furthermore, in additional to the vocabulary selection threshold, we utilize the threshold to terminate the training if the model has converged. These steps help to avoid having too few words left. Note that the partition of words primarily depends on the predefined hyper-parameter of Dirichlet distribution. The parameter controls the the topic distribution (i.e, how words are related to the topics). The smaller is, the more disjoint the partition of words may be. Therefore, avoiding using an extremely small plus the above strategies together will alleviate the concern of having too few words with disjointed topic distributions. Variants of VSEC-LDA Model I can be extended to different variants, and we illustrated two here: Model II and Model III, targeting different applications/purposes. Model II shown in blue dash line in Fig. 6 considers the situation when more than one visual modality is available, and is equivalent to the core model if one of the visual modality is removed. The generative process for each visual modality is the same as that in the core model, and the visual modality is independent of each other. For the textual modality, a relevance variable r is introduced to monitor the relevance between the textual modality and two visual modalities. Model III (in black dash line in Fig. 6 is for a neuroscience application on analyzing the correspondence between neural activities and brain functions. This model uses GC-LDA as the building block, which employs R = 2 components for each topic distribution. This allows the model to learn topics where a single cognitive function is associated with spatially-discontiguous patterns of activation. One can refer to for more details. With the vocabulary embedded, the text words associated with one topic are more correlated than GC-LDA. Due to the page limitation, we will elaborate the two variants in the supplemental material. Evaluation We evaluate our model using both synthetic and real data. Experiments with synthetic data allow us to explicitly check if the model can identify the ideal vocabulary. With real data, we compare Model I to Corr-LDA with different configurations on Corel-5K and SUN databases; Model II is compared with VELDA, a multi-modalities topic model on Wikipedia POTD database; and Model III is compared with GC-LDA on the Neurosynth database. Experiments with synthetic database We use simulations to test if the proposed method can remove non-contributing words in the vocabulary as well as learn the model. The synthetic database used in the simulations is formed by following the generative process of the core model mentioned in the previous section. To be specific, the number of topic K is 20, and the relevant textual and visual words have a size of 80 and 800 respectively. The vocabulary size of irrelevant words in the modalities varies from 20-140 and 200-1400, respectively. We generated 8000 documents for each corpus. For each document d, we first draw topic proportion d from Dirichlet distribution, then we draw each relevant visual word from multinomial distribution. For textual word, first we derive topic proportion given the corresponding visual word, then relevant textual word is drawn from multinomial distribution. After relevant part is generated, irrelevant words are drawn from two uniform distributions corresponding to each modality. At last, we randomly select 90% of the corpus as the training set and the remaining 10% as the testing set. Additionally, we fix the model hyper-parameter to be: = 0.2, = 0.1 and = 0.1. We set the parameter for Dirichlet distribution to be relatively small, as the experiment with synthetic data only serves as a proof-of-concept. Thus we expect that each document links to some topics, and each topic only picks part of the relevant words with high probability. We ran the simulation 50 times for each synthetic database. In each experiment, 20 iterations of inference are conducted to yield an initial estimation for all parameters before applying vocabulary selection. Vocabulary selection is performed when the model reaches a local plateau. The experiment automatically terminates when no more words is filtered out and the likelihood stabilizes. Fig. 2 shows the number of filtered irrelevant words with fixed and dynamic filtering threshold for one synthetic database. It illustrates that 1) irrelevant words in both visual and textual modalities can be removed by the algorithm and 2) fixed and dynamic thresholds produce similar results. Thus, in the following experiments, we use the fixed threshold to alleviate computational complexity. We also test the limitation of vocabulary selection by experiments with synthetic databases of various number of ir- relevant words. We set the maximum training iteration to be 150 and use the fixed filtering threshold for all experiments for fair comparison. Table 2 shows part of the results given by 50 rounds of experiments for each synthetic database (full results are shown in supplementary material). With the constraint of maximum training iteration, the filtering component starts to see difficulty when irrelevant words become dominant compared to relevant word in the corpus, which is intuitive. Moreover, The trend of average negative log-likelihood, which is used to measure the quality of modeling, also verify the efficiency of our model. The details are presented in the supplementary material. Experiments with Real databases Since our model is a multi-modality model, we use Corr-LDA, VELDA, and GC-LDA as baselines, which are all multi-modality models. We first conduct experiments comparing our Model I with Corr-LDA on two real databases with various configurations. Then, we compare the two variants with another two baselines: VELDA and GC-LDA respectively to show the generalization ability of our model. Real databases and Experimental Settings We use four widely used benchmarks in this paper and the statistic are detailed in the following: 1). Corel 5K contains 4500 and 499 images in the training set and testing set,respectively. Each image is manually annotated with 1 to 5 tags. The word vocabulary contains 260 words. 2). SUN is a large scene understanding database. We build a balanced subset by selecting images from four categories with 4,749 images in total, and around 1,000 images in each category. We have 647 words in the textual vocabulary. 3). Wikipedia POTD is a set of pictures with descriptions from Wikipedia. We use the database provided by, which has 2,274 documents for training and 250 for testing. There are 1,000 and 997 words for two visual modalities respectively. The textual modality has 3224 words. 4). Neurosynth is a publicly available database consisting of data automatically extracted from a large collection of functional magnetic resonance imaging (fMRI) publications. The database employed here has 11,362 total publications, which has on average 35 peak activation tokens (viewed as a visual modality), and 46 word tokens (viewed as a textual modality) after preprocessing. We follow to restrict visual vocabulary to SIFT descriptors. We build three visual vocabularies of size 1,200, 1,500 and 1,700 for Corel-5K and 1,000, 1,500 and 2,000 for SUN, respectively. We use the Wikipedia POTD and Neurosynth database provided by the author without further preprocessing. For fair comparison, we fix the hyper-parameters to typical settings: = 1, = 0.1 and = 0.1 and tune the number of topics K for Model I and Corr-LDA. We test both methods with different settings: 1)filtering both modalities/ one modality, 2) doing/not doing re-estimation of hyper-parameters. In the comparison experiments with VELDA and GC-LDA, we simply following their settings for fairness. Results and Analysis The textual vocabulary is built with tags/objects, which are typically from a pre-processing step for Corel-5K and SUN databases. Hence we keep them all to avoid over-filtering. Therefore, in the following experimental analysis, if not specified otherwise, only the visual modality gets filtered. Test set likelihood. It is used for evaluating the generalization performance of a topic model. If a model fit the data better, it should assign a higher likelihood to the test set. We compute the per-image average negative log-likelihood of the test set with various hidden topic number K to check 1) the capability of our model to fit the data and 2) the robustness of our model to various K. Fig. 3 illustrates the results conducted on Corel-5K. As expected, VSEC-LDA provides a better fit than Corr-LDA on the test data regardless of the initial vocabulary size. Also, the larger the vocabulary size is, the larger the negative log likelihood is for Corr-LDA. This implies that a dense vocabulary is beneficial for learning the model. On the other hand, our approach achieves similar optimal likelihood across different settings of the initial size, and it is lower than that of Corr-LDA. This phenomenon implies our method can achieve a reasonable vocabulary size robustly. For Corel 5K, when the visual vocabulary size is 1200, the optimal average negative log-likelihood given by both method are similar. However, Corr-LDA fails to fit the data well for the other two vocabulary size settings. In this sense, 1200 is the optimal visual vocabulary size for Corel-5K. Image annotation performance. After a model is trained, given an image without its captions, one can use the model to compute a distribution over textual word conditioned on the visual word, given by p(w|v). This distribution reflects a prediction of the missing tag for the image. We show some examples of image annotation in the supplementary material. But if a predicted caption is accurate may be subjective, we employ "perplexity" as an objective measure. Perplexity is introduced from conventional lan-guage modeling to evaluate the uncertainty of the predicted captions. It is equivalent algebraically to the inverse of of the geometric mean per-word likelihood. Similar to negative log likelihood measurement, a lower perplexity score indicates a better predictive performance. Fig. 4 shows the predictive perplexity of Corel-5K (similar trend for SUN) with various topics k, where we notice that our model always achieves a lower value than Corr-LDA. It is worth mentioning that, with the initial visual vocabulary size increasing (from left to right in Fig. 4), both methods gradually reach lower perplexity value, and the difference between the methods decreases. This is because when the number of topics k increases, words close to each other (in terms of having similar probability given the topic-word distribution under the same topic) will have a higher chance of being assigned to different topics. Text-based image retrieval. If a model is learned well, the relationship between visual and textual words should be tight, meaning given one modality, it can refer to the other modality. Thus, we consider text-based image retrieval task: Given the textual word of an image, attempt to retrieve its accompanying image from the image set. To be specific, given a textual query Q = 1, 2,..., q, we compute a relevance score for image i by the following formula: wherep(q| i ) is the probability of the q-th query word under the distribution p(q|C)p(C|). Having the score of each image at hand, we return a list of images ranked in descending order by conditional likelihood. Since we know the corresponding image for each textual query list, the position of the ground-truth image in the ranked list is considered as a measurement: An image is correctly retrieved if it appears in the top t percent of the image test set, and the error rate is computed by Eq.. wherewhere rank i is the rank of the retrieved image i in the ranking list. Here we only present the results of Model II and VELDA on Wikipedia POTD as an example to show both the performance improvement and demonstrate the generalization ability of our model. More experiment results and analysis of Model I and Corr-LDA with various configurations can be found in the supplemental material. Since the vocabulary size of all the three modalities is around or more than 1000, we consider removing irrelevant words from all modalities. Results of the error rate of POTD are shown in Fig. 5. The figure depicts the retrieval errors averaged over all testing queries. A better performance is equivalent to lower error rateCurves are closer to the bottom left corner. From the curve, it is obvious that our method achieve a much lower error rate than that of VELDA approach. We also compute the recall on the top 10% level. each topic, we may identify the function of the activated area in the brain. In the table, the second and third rows show the words from topic28 by VSEC-LDA and GC-LDA respectively. It is obvious the words given by our model are all related to sentence understanding, while GC-LDA gives words like 'slowly' and 'input', which are not related to the given topic. Conclusion We proposed VSEC-LDA, a new topic model with integrated vocabulary selection. Existing topic models generally require a data pre-processing step before learning the models. This step is typically independent of the model learning step and thus does not necessarily support finding an optimal topic model for a given problem. The proposed approach uses entropy-based metrics to account for the relevant importance of the words to the model and dynamically determine the relevant words while learning the model. Experiments with synthetic databases show that the proposed model can select the relevant words effectively. Real-data-based experiments also demonstrate improved performance of the proposed model in different applications. More experiment results In this section, we will further demonstrate the improved performance of our model compared with baselines by more experiment results with both synthetic database and real database. Experiments with synthetic database The full results of testing the limitation of vocabulary selection is shown in Table 5. The maximum training iteration is 150 and fixed filtering threshold is applied for all experiments regarding fair comparison. With the constraint of maximum training iteration, the filtering component starts to see difficulty when irrelevant words become dominant compared to relevant word in the corpus, which is intuitive. Moreover, Figure 8 illustrates the trend of average negative log-likelihood, which is used for measuring the quality of modeling (the lower the value is, the better the quality is), along with the inference iterations on one training set. The black curve represents the Corr-LDA performs on corpus of both relevant and irrelevant words, while the blue curve represents VSEC-LDA (without re-estimation step) performs on the same corpus. It is obvious that our method gave better results than Corr-LDA. Moreover, if we conduct re-estimation step, VSEC-LDA (green curve) performs even better than Corr-LDA (red curve) on corpus without irrelevant words. We also tried to randomly select words from the given corpus to form a new corpus, and applied Corr-LDA to the new corpus. The result (pink curve) shows it achieves better fitting than Corr-LDA with original corpus, but by simply randomly selecting words, it cannot get optimal result. Experiment with real database We show some examples of image annotation from SUN in Table 4. Both Corr-LDA and our model can capture the majority objects (captions) in the given image. But the retrieved textual words are more correlated given by our model. We also conduct experiment to test the text-based image retrieval performance of Model I and Corr-LDA on SUN. We consider filtering words on both modalities or only visual modality for SUN, of which textual vocabulary size is 695. Results of the error rate of SUN are shown in Fig.. Each figure depicts the retrieval errors averaged over all testing queries. For all figures, better performance is equivalent to lower error rateCurves are closer to the bottom left corner. In general, our method with re-estimation step (red and blue curve) performs the best, followed by the baseline (black curve). Our method without re-estimation step (red curve) performs slightly worse than the baseline. This phenomenon demonstrates performing filter and model parameter estimation simultaneously sometimes leads to a Groudtruth: house building mountain pole vineyard window trees VSEC-LDA: house tree mountain grass window Corr-LDA: building trees window Groudtruth: sky mountain ground grass river water tree VSEC-LDA: sky cloud mountains grass water Corr-LDA: sky mountain trees Groudtruth: trees grass bottles bag tree truck tent bicycle VSEC-LDA: bicycle grass tent Corr-LDA: tree tent bicycle sub-optimal parameter estimation for the model. Therefore, re-estimating the model parameter with filtering component turned off is an intuitive way to achieve optimal estimation. It is worth mentioning that, with filtering component performing on both the modalities or one of them, our method performed differently, and the best performance achieved when the filtering component only conducted the visual modality. This is not surprising since the words we used to build our vocabulary are not the raw tags/objects, and have been pre-processed. Moreover, we found regardless the difference initial conditions, we got similar relevant textual words when the model converged, only around 30 words being filtered. Thus if excrescent filtering is applied, the performance will depreciate. Table 6 shows the top 10 words according to the topic distributions learned by Model III and GC-LDA respectively for sample topics (topic28 and topic 57). Here there is no groundtruth for direct evaluation as before, and thus we rely on examining how relevant the words are. By looking at how the textual words are related to each topic, we may identify the function of the activated area in the brain. In the table, the second and third rows show the words from # of irrelevant words 40/400 60/600 80/800 100/1000 120/1200 140/1400 # of filtered textual words 40 ± 2 59 ± 2 72 ± 5 86± 7 105 ± 7 121 ± 7 # of filtered visual words 392 ± 9 587 ± 11 775 ± 8 966 ± 15 1089 ± 12 1121 ± 15 topic28 by VSEC-LDA and GC-LDA respectively. It is obvious the words given by our model are all related to sentence understanding, while GC-LDA gives words like 'slowly' and 'input', which are not related to the given topic. Similar to topic 28, in topic 57, the topic-word distribution learned by GC-LDA produce word 'eight', which is not related to any of other words. |
Are observed decadal changes in intermediate water masses a signature of anthropogenic climate change? Recent observations have shown relatively large changes in the temperature and salinity of intermediate water masses in the ocean on decadal timescales. We compare the observed changes with simulations of the coupled climate model HadCM3. In simulations driven by anthropogenic forcing, we see significant changes in intermediate waters in the Southern Ocean which are similar in both pattern and magnitude to the observations. Our results suggest that the observed changes are most likely to be a signal of anthropogenic climate change. The strong signal in the Southern Ocean, which is detectable in the model from the 1980s, is in marked contrast with the intermediate waters of the Northern hemisphere oceans, where internal climate variability is large and a signal of anthropogenic climate change is not detectable in the model until 2020 at the earliest. Our results suggest that intermediate waters, particularly those of the Southern hemisphere, are a potentially sensitive indicator of anthropogenic climate change, and could be an important part of a climate monitoring network. |
/**
* Servlet implementation class Summary
*/
@WebServlet(description = "Portfolio summary servlet", urlPatterns = { "/summary" })
@ServletSecurity(@HttpConstraint(rolesAllowed = { "StockTrader", "StockViewer" } ))
@RequestScoped
public class Summary extends HttpServlet {
private static final long serialVersionUID = 4815162342L;
private static final String EDITOR = "StockTrader";
private static final String LOGOUT = "Log Out";
private static final String CREATE = "create";
private static final String RETRIEVE = "retrieve";
private static final String UPDATE = "update";
private static final String DELETE = "delete";
private static final String BASIC = "basic";
private static final String BRONZE = "bronze";
private static final String SILVER = "silver";
private static final String GOLD = "gold";
private static final String PLATINUM = "platinum";
private static final String UNKNOWN = "unknown";
private static final String DOLLARS = "USD";
private static Logger logger = Logger.getLogger(Summary.class.getName());
private static HashMap<String, Double> totals = new HashMap<String, Double>();
private static HashMap<String, org.eclipse.microprofile.metrics.Gauge> gauges = new HashMap<String, org.eclipse.microprofile.metrics.Gauge>();
private NumberFormat currency = null;
private int basic=0, bronze=0, silver=0, gold=0, platinum=0, unknown=0; //loyalty level counts
private @Inject @ConfigProperty(name = "TEST_MODE", defaultValue = "false") boolean testMode;
private @Inject @RestClient PortfolioClient portfolioClient;
private @Inject JsonWebToken jwt;
private @Inject MetricRegistry metricRegistry;
//used in the liveness probe
public static boolean error = false;
public static String message = null;
// Override Portfolio Client URL if config map is configured to provide URL
static {
String mpUrlPropName = PortfolioClient.class.getName() + "/mp-rest/url";
String portfolioURL = System.getenv("PORTFOLIO_URL");
if ((portfolioURL != null) && !portfolioURL.isEmpty()) {
logger.info("Using Portfolio URL from config map: " + portfolioURL);
System.setProperty(mpUrlPropName, portfolioURL);
} else {
logger.info("Portfolio URL not found from env var from config map, so defaulting to value in jvm.options: " + System.getProperty(mpUrlPropName));
}
}
/**
* @see HttpServlet#HttpServlet()
*/
public Summary() {
super();
currency = NumberFormat.getNumberInstance();
currency.setMinimumFractionDigits(2);
currency.setMaximumFractionDigits(2);
currency.setRoundingMode(RoundingMode.HALF_UP);
}
/**
* @see HttpServlet#doGet(HttpServletRequest request, HttpServletResponse response)
*/
protected void doGet(HttpServletRequest request, HttpServletResponse response) throws ServletException, IOException {
String rows = null;
try {
rows = getTableRows(request);
} catch (Throwable t) {
logException(t);
message = t.getMessage();
error = true;
}
boolean editor = request.isUserInRole(EDITOR);
Writer writer = response.getWriter();
writer.append("<!DOCTYPE html>");
writer.append("<html>");
writer.append(" <head>");
writer.append(" <title>Stock Trader</title>");
writer.append(" <meta http-equiv=\"Content-Type\" content=\"text/html; charset=UTF-8\">");
writer.append(" </head>");
writer.append(" <body>");
writer.append(" <img src=\"header.jpg\" width=\"534\" height=\"200\"/>");
writer.append(" <br/>");
writer.append(" <br/>");
if (error) {
writer.append(" Error communicating with the Portfolio microservice: \""+message+"\"");
writer.append(" <p/>");
writer.append(" Please consult the <i>trader</i> and <i>portfolio</i> pod logs for more details, or ask your administator for help.");
writer.append(" <p/>");
} else {
writer.append(" <form method=\"post\"/>");
if (editor) {
writer.append(" <input type=\"radio\" name=\"action\" value=\""+CREATE+"\"> Create a new portfolio<br>");
}
writer.append(" <input type=\"radio\" name=\"action\" value=\""+RETRIEVE+"\" checked> Retrieve selected portfolio<br>");
if (editor) {
writer.append(" <input type=\"radio\" name=\"action\" value=\""+UPDATE+"\"> Update selected portfolio (add stock)<br>");
writer.append(" <input type=\"radio\" name=\"action\" value=\""+DELETE+"\"> Delete selected portfolio<br>");
}
writer.append(" <br/>");
writer.append(" <table border=\"1\" cellpadding=\"5\">");
writer.append(" <tr>");
writer.append(" <th></th>");
writer.append(" <th>Owner</th>");
writer.append(" <th>Total</th>");
writer.append(" <th>Loyalty Level</th>");
writer.append(" </tr>");
writer.append(rows);
writer.append(" </table>");
writer.append(" <br/>");
writer.append(" <input type=\"submit\" name=\"submit\" value=\"Submit\" style=\"font-family: sans-serif; font-size: 16px;\"/>");
writer.append(" <input type=\"submit\" name=\"submit\" value=\"Log Out\" style=\"font-family: sans-serif; font-size: 16px;\"/>");
writer.append(" </form>");
}
writer.append(" <br/>");
writer.append(" <a href=\"https://github.com/slombardo2\">");
writer.append(" <img src=\"footer.jpg\"/>");
writer.append(" </a>");
writer.append(" </body>");
writer.append("</html>");
}
/**
* @see HttpServlet#doPost(HttpServletRequest request, HttpServletResponse response)
*/
protected void doPost(HttpServletRequest request, HttpServletResponse response) throws ServletException, IOException {
String submit = request.getParameter("submit");
if (submit != null) {
if (submit.equals(LOGOUT)) {
request.logout();
HttpSession session = request.getSession();
if (session != null) session.invalidate();
response.sendRedirect("login");
} else {
String action = request.getParameter("action");
String owner = request.getParameter("owner");
if (action != null) {
if (action.equals(CREATE)) {
response.sendRedirect("addPortfolio"); //send control to the AddPortfolio servlet
} else if (action.equals(RETRIEVE)) {
response.sendRedirect("viewPortfolio?owner="+owner); //send control to the ViewPortfolio servlet
} else if (action.equals(UPDATE)) {
response.sendRedirect("addStock?owner="+owner); //send control to the AddStock servlet
} else if (action.equals(DELETE)) {
// PortfolioServices.deletePortfolio(request, owner);
portfolioClient.deletePortfolio("Bearer "+getJWT(), owner);
doGet(request, response); //refresh the Summary servlet
} else {
doGet(request, response); //something went wrong - just refresh the Summary servlet
}
} else {
doGet(request, response); //something went wrong - just refresh the Summary servlet
}
}
} else {
doGet(request, response); //something went wrong - just refresh the Summary servlet
}
}
private String getTableRows(HttpServletRequest request) {
StringBuffer rows = new StringBuffer();
if (portfolioClient==null) {
throw new NullPointerException("Injection of PortfolioClient failed!");
}
if (jwt==null) {
throw new NullPointerException("Injection of JWT failed!");
}
// JsonArray portfolios = PortfolioServices.getPortfolios(request);
Portfolio[] portfolios = testMode ? getHardcodedPortfolios() : portfolioClient.getPortfolios("Bearer "+getJWT());
basic=0; bronze=0; silver=0; gold=0; platinum=0; unknown=0; //reset loyalty level counts
metricRegistry.remove("portfolio_value");
for (int index=0; index<portfolios.length; index++) {
Portfolio portfolio = portfolios[index];
String owner = portfolio.getOwner();
double total = portfolio.getTotal();
String loyaltyLevel = portfolio.getLoyalty();
setPortfolioMetric(owner, total);
if (loyaltyLevel!=null) {
if (loyaltyLevel.equalsIgnoreCase(BASIC)) basic++;
else if (loyaltyLevel.equalsIgnoreCase(BRONZE)) bronze++;
else if (loyaltyLevel.equalsIgnoreCase(SILVER)) silver++;
else if (loyaltyLevel.equalsIgnoreCase(GOLD)) gold++;
else if (loyaltyLevel.equalsIgnoreCase(PLATINUM)) platinum++;
else unknown++;
}
rows.append(" <tr>");
rows.append(" <td><input type=\"radio\" name=\"owner\" value=\""+owner+"\"");
if (index == 0) {
rows.append(" checked");
}
rows.append("></td>");
rows.append(" <td>"+owner+"</td>");
rows.append(" <td>$"+currency.format(total)+"</td>");
rows.append(" <td>"+loyaltyLevel+"</td>");
rows.append(" </tr>");
}
return rows.toString();
}
Portfolio[] getHardcodedPortfolios() {
Portfolio john = new Portfolio("John");
john.setTotal(1234.56);
john.setLoyalty("Basic");
Portfolio karri = new Portfolio("Karri");
karri.setTotal(12345.67);
karri.setLoyalty("Bronze");
Portfolio ryan = new Portfolio("Ryan");
ryan.setTotal(23456.78);
ryan.setLoyalty("Bronze");
Portfolio greg = new Portfolio("Greg");
greg.setTotal(98765.43);
greg.setLoyalty("Silver");
Portfolio eric = new Portfolio("Eric");
eric.setTotal(123456.78);
eric.setLoyalty("Gold");
Portfolio kyle = new Portfolio("Kyle");
kyle.setLoyalty("Basic");
Portfolio[] portfolios = { john, karri, ryan, greg, eric, kyle };
return portfolios;
}
void setPortfolioMetric(String owner, double total) {
totals.put(owner, total);
if (gauges.get(owner)==null) try { //gauge not yet registered for this portfolio
org.eclipse.microprofile.metrics.Gauge<Double> gauge = () -> { return totals.get(owner); };
Metadata metadata = Metadata.builder().withName("portfolio_value").withType(MetricType.GAUGE).withUnit(DOLLARS).build();
metricRegistry.register(metadata, gauge, new Tag("owner", owner)); //registry injected via CDI
gauges.put(owner, gauge);
} catch (Throwable t) {
logger.warning(t.getMessage());
}
}
@Gauge(name="portfolio_loyalty", tags="level=basic", displayName="Basic", unit=MetricUnits.NONE)
public int getBasic() {
return basic;
}
@Gauge(name="portfolio_loyalty", tags="level=bronze", displayName="Bronze", unit=MetricUnits.NONE)
public int getBronze() {
return bronze;
}
@Gauge(name="portfolio_loyalty", tags="level=silver", displayName="Silver", unit=MetricUnits.NONE)
public int getSilver() {
return silver;
}
@Gauge(name="portfolio_loyalty", tags="level=gold", displayName="Gold", unit=MetricUnits.NONE)
public int getGold() {
return gold;
}
@Gauge(name="portfolio_loyalty", tags="level=platinum", displayName="Platinum", unit=MetricUnits.NONE)
public int getPlatinum() {
return platinum;
}
@Gauge(name="portfolio_loyalty", tags="level=unknown", displayName="Unknown", unit=MetricUnits.NONE)
public int getUnknown() {
return unknown;
}
private String getJWT() {
String token;
if("bearer".equals(PropagationHelper.getAccessTokenType())) {
token = PropagationHelper.getIdToken().getAccessToken();
logger.fine("Retrieved JWT provided through oidcClientConnect feature");
} else {
token = jwt.getRawToken();
logger.fine("Retrieved JWT provided through CDI injected JsonWebToken");
}
return token;
}
static void logException(Throwable t) {
logger.warning(t.getClass().getName()+": "+t.getMessage());
//only log the stack trace if the level has been set to at least INFO
if (logger.isLoggable(Level.INFO)) {
StringWriter writer = new StringWriter();
t.printStackTrace(new PrintWriter(writer));
logger.info(writer.toString());
}
}
} |
Preliminary Phytochemical, Antioxidant and Hypolipidemic Potential of Aqueous Extract of Ferula asafoetida-An in vitro Study Aim: To analyse the preliminary phytochemical, antioxidant, anti-cholesterol potential of aqueous extract of Ferula asafoetida. Background: Hyperlipidemia is considered as one of the leading causes behind the occurrence of deadly disorders like diabetes, cardiovascular diseases, atherosclerosis etc. It is characterised by elevated levels of plasma lipids, mainly total cholesterol. Antioxidants are compounds which can inhibit oxidative damage. is the herbaceous plant belonging to the family Umbelliferae. It is used as spice in food and also used as digestive aid. It is used in the treatment of asthma and whooping cough and it also helps to reduce blood pressure. Methods: Aqueous extract of Ferula asafoetida was prepared by hot percolation method. The screening of phytochemical constituents, assessment of in vitro antioxidant activity and anticholesterol activity were done using standard procedures and the data were analysed statistically using one-way analysis of variance (ONE-WAY ANOVA) and the significance was considered at the levels of p<0.05. Results: Ferula asafoetida extract was rich in phytochemicals and possessed potent in vitro antioxidant activity. Anti-cholesterol activity of Ferula asafoetida extract was examined and it was Original Research Article Asmidha et al.; JPRI, 34(6A): 10-18, 2022; Article no.JPRI.80930 11 observed that the plant extract exhibited significant anti cholesterol potential in a dose dependent manner with an IC 50 value of 400 g/ml. Conclusion: The study established the in vitro antioxidant and hypolipidemic potential of aqueous extract of Ferula asafoetida. It is concluded that the extract of Ferula asafoetida possesses potent antioxidant and anticholesterol activity. INTRODUCTION Ferula asafoetida is a herbaceous plant belonging to the Umbelliferae family. It has been used as a spice in food for more than thousands of years. It is also used as a digestive aid. It can change the dull food into an appetizing meal by their colour, flavor and pungency. The resin gum obtained from Ferula asafoetida is used to treat whooping cough and ulcer. It is used in the treatment of cancer,women's allergies, blood pressure, and chemoprotive therapy. It is used in modern treatments in whooping cough, asthma, and also it reduces cholesterol. The oil of Ferula asafoetida is reported to have antifungal activity. It has cardioprotective effects in low doses and cardiotoxic effects in higher doses. It is also used in organic farming to kill several insect pests. It helps to treat gastrointestinal, neurological and respiratory disorders. Phytochemicals are the compounds produced in the plants. Secondary metabolites are present only in plants possessing various biological activities. They help in the normal processing and are capable of prevention or treatment of various disorders. Phytochemicals generally are regarded as an important compound of research interest because of their possible health effects. Antioxidant is the material which terminates the chain reaction by removing the free radical. Antioxidants reduce the risk of cancer, cardiovascular diseases, diabetes and other diseases which are caused by oxidative stress. Hyperlipidemia is a secondary metabolic disorder. It is characterised by elevated levels of serum triglycerides, cholesterol and low density lipoproteins and decreased serum levels of high density lipoproteins. Hyperlipidemia is regarded as a major risk factor for the premature development of cardiovascular disease like atherosclerosis, hypertension, coronary heart disease etc.. It is also associated with other metabolic disorders like obesity, diabetes mellitus etc. Statin drugs are widely used for the treatment of hypercholesterolemia as it is an inhibitor of the key enzyme HMG CoA reductase, which is involved in the cholesterol biosynthesis. Our team has extensive knowledge and research experience that has translate into high quality publications. Hence the study aimed to assess the anticholesterol and antioxidant activity of Ferula asafoetida extract by in vitro methods. Preparation of Aqueous Extract of Ferula asafoetida Ferula asafoetida was purchased from a herbal health care centre. Crushed and made into powder. 80% aqueous extract was obtained. The extract was prepared by a hot percolation method. Phytochemical analysis, assessment of in vitro antioxidant and anti cholesterol potential was evaluated. Test for phlobatannin 1 ml of the extract was treated with 1ml of 1% HCl and boiled for 10 mins. The formation of red color precipitate indicates the presence of phlobatannin. Test for Carbohydrates Three to five drops of Molisch reagent was added with 1 mL of the extract and then 1 mL of concentrated sulphuric acid was added carefully through the side of the test tube. The mixture was then allowed to stand for two minutes and diluted with 5 mL of distilled water. The development of a red or dull violet ring at the junction of the liquids showed the presence of carbohydrates. Test for Flavonoids Few drops of 1% liquid ammonia were taken in a test tube and along with it 1ml of the extract was added resulting in the formation of yellow color thereby indicating the presence of flavonoids. Test for Alkaloids 2ml of sample was mixed with 2ml of HCl. Then 6 drops of HCN was added and further 2 drops of picric acid was added that resulted in a creamish pale yellow ppt indicating the presence of alkaloids. Test for Terpenoids 2 ml of sample along with 2ml of chloroform and 3 ml of con.H2SO4 was added. Red color ppt obtained indicates the presence of terpenoids. Test for proteins One milliliter of ninhydrin was dissolved in 1 mL of acetone and then a small amount of extract was added with ninhydrin. The formation of purple colour revealed the presence of protein. Detection of saponins Foam test: A fraction of the extract was vigorously shaken with water and observed for persistent foam. Test for steroids One ml of chloroform was mixed with 1 mL of extract and then ten drops of acetic anhydride and five drops of concentrated sulphuric acid were added and mixed. The formation of dark red colour or dark pink colour indicates the presence of steroids. In vitro Antioxidant Activity (DPPH Free Radical Scavenging Activity) Scavenging of 2, 2-Diphenyl-1-picrylhydrazyl (DPPH) radicals was assessed by the method of Hatano et al,. DPPH solution (1.0 ml) was added to 1.0 ml of extract at different concentrations (0.1 to 0. 5mg/ml). The mixture was kept at room temperature for 50 minutes and the activity was measured at 517 nm. Ascorbic acid at the same concentrations was used as standard. The capability to scavenge the DPPH radical was calculated and expressed in percentage (%) using following formula: DPPH radical scavenging (%) = Control OD -Sample OD/ control OD X 100 In vitro Anti-cholesterol Activity The anti-cholesterol assay was carried out as described as per the kit method (Spinreact, S.A.U-Ctra Santa Coloma, Girona, Spain). Cholesterol was dissolved in chloroform at a concentration of 2.5 mg mL/ml. Ten microliter of the extract was pipetted into a microtiter plate followed by the addition of 2000 L of R1 reagent and 10 L of cholesterol as sample. Twenty microliter of distilled water and 2000 L of R1 reagent were used as blank. Negative control consisted of 20 L cholesterol and 2ml R1; standard consisted of 20 L simvastatin and 2mL R1 reagent. The contents were incubated between 0-30 min at room temperature and the absorbance was read at 500 nm in a UV-Vis spectrophotometer against reagent blank. Anticholesterol assay of the extract was calculated using the following equation: Inhibition (%) = Negative control-Sample. 100/Negative OD. Statistical Analysis The data were subjected to statistical analysis using one-way analysis of variance (ANOVA) and Duncan's multiple range test to assess the significance of individual variations between the groups. In Duncan's test, significance was considered at the level of p<0.05. Phytochemical Screening of Ferula asafetida The results of phytochemical analysis showed the presence of proteins, flavonoids, alkaloids, terpenoids, saponins and steroids in Ferula asafoetida.The carbohydrates and the amino acids are absent in ferula asafoetida ( Table 1). The presence of phytochemicals in the extract might be the underlying reason for the therapeutic properties of Ferula asafoetida. Antioxidant Activity of Aqueous Extract of Ferula asafetida Antioxidant activity of aqueous seed extracts of Ferula asafoetida was determined by performing DPPH free radical scavenging assay (Fig. 1). The extract showed a dose-dependent increase in the antioxidant activity although its activity is less compared to the standard vitamin C. The P value is p<0.05. Anti Cholesterol Activity of Aqueous Extract of Ferula asafetida anticholesterol activity of Ferula asafoetida aqueous extract revealed a dose dependent increase in the percentage of inhibitory activity. The extract showed a potent anti cholesterol activity with a IC 50 value of 450 g/ml (Fig. 2). The activity of the extract is less compared to the standard drug simvastatin in all the tested concentrations. Simvastatin is a standard anticholesterol drug which acts as HMG CoA reductase inhibitor. It showed that the extract is showing less activity compared to simvastatin. The p value is p<0.05. DISCUSSION Plants are an excellent source of biologically active compounds known as phytochemicals. It is well established that free radicals are the major cause of various chronic and degenerative diseases, like coronary heart disease, inflammation, stroke, diabetes mellitus and cancer. The medicinal value of plants is related to the content of phytochemicals including phenolic compounds, flavonoids, alkaloids, tannins etc. The results of phytochemical analysis showed the presence of proteins, flavonoids, alkaloids, terpenoids, saponins and steroids in Ferula asafoetida ( Table 1). The presence of phytochemicals in the extract might be the underlying reason for the therapeutic properties of Ferula asafoetida. Antioxidant activity of aqueous seed extracts of Ferula asafoetida was determined by performing DPPH free radical scavenging assay (Fig. 1). The extract showed a dose-dependent increase in the antioxidant activity although its activity is less compared to the standard vitamin C. Free radicals are molecules possessing unpaired electrons. Antioxidants are the most effective ingredients to eliminate free radicals which can create oxidative stress and are possible protective tools that protect the cells from reactive oxygen species. They retard the progress of many diseases as well as lipid peroxidation. Additionally, they also possess antiinflammatory, anti-viral and anti-cancer properties. Ferula asafoetida extracts which are rich in phytochemicals which have great importance in free radical scavenging. The effect of antioxidants on the DPPH free radical scavenging was considered due to their hydrogen donating ability. The extract of Ferula asafoetida exhibited a significant antioxidant potential with standard IC 50 value of 380 g/ml Further studies may be needed to find out the potential health benefits of the extracts in prevention and scavenging of free radicals. Investigation of anticholesterol activity of Ferula asafoetida aqueous extract revealed a dose dependent increase in the percentage of inhibitory activity. The extract showed a potent anti cholesterol activity with a IC 50 value of 450 g/ml (Fig. 2). The activity of the extract is less compared to the standard drug simvastatin in all the tested concentrations. Simvastatin is a standard anti-cholesterol drug which acts as HMG CoA reductase inhibitor. Although statin groups of drugs do well in many people, the use of it is associated with potential adverse effects which include cognitive loss, neuropathy, pancreatic and hepatic dysfunction, and sexual dysfunction. Hence our study showed that even though the extract is showing less activity compared to simvastatin, since the extract is natural in origin it can avoid the adverse side effects created by the synthetic drugs. Each bar represents Mean ± SEM of 3 independent observations. Significance at p < 0.05 CONCLUSION The present study can be concluded that the extract of Ferula asafoetida possesses potent antioxidant and anticholesterol activity. Further detailed studies on in vitro cell lines and in vivo experimental animal models need to be done on the plant extract for the formulation of the anticholesterol drug towards the clinical utility. CONSENT It is not applicable. ETHICAL APPROVAL It is not applicable. SOURCE OF FUNDING The present study was supported by the following agencies |
Sodium in the Atmospheres of Thick-Disk Red Giants The parameters and elemental abundances of atmospheres for ten thick-disk red giants was determined from high-resolution spectra by the method of model stellar atmospheres. The results of a comparative analysis of the abundances in the atmospheres of the investigated stars and thin disk red giants are presented. Sodium in the atmospheres of thick-disk red giants is shown to have no overabundances typical of thin-disk red giants. INTRODUCTION The abundances of chemical elements in the atmospheres of stars are known to change in the course of their evolution. When investigating the chemical composition of giants, the first to be detected were the CNO abundance anomalies compared to the chemical composition of dwarf stars (see, e.g., ). The observed sodium overabundance in the atmosphere of the red giant V ir was first mentioned in 1963. Subsequently, sodium overabundances were detected in red giants of the Hyades open cluster. In another paper, the same authors found several giants with a overabundance and may have been the first to point out a systematic difference in sodium abundance between red giants and dwarfs. They hypothesized that sodium might be synthesized in some red giants. Slightly later, in 1970 noted that the abundance increases with temperature parameter =5040/T eff. In other words, the abundance in cooler stars turned out to be, on average, higher, while giants are in the majority among these stars, i.e., the sodium abundance difference between dwarfs and giants is also noticeable, although the authors did not reach this conclusion. were the first to describe the dependence of the overabundances in the atmospheres of normal red giants on surface gravity. This dependence turned out to be similar to that discovered previously for supergiants. It was explained by the hypothesis that sodium could be produced in the 22 Ne(p,) 23 Na reaction entering the neonsodium cycle of hydrogen burning in the cores of main sequence stars and could then be brought from deep layers into the stellar atmosphere through developing convection as the star evolves from the main sequence to the red giant branch. The calculations performed by confirmed the validity of this hypothesis.Further studies revealed overabundances in the atmospheres of various classes of red giants: moderate and classical barium stars, super-metal rich stars. A database of characteristics for red giants was created from the accumulated material at the Institute of Astronomy, the Russian Academy of Sciences. Our studies using the database revealed several stars with a underabundance relative to the observed dependence on surface gravity. A comparative analysis showed that these stars are distinguished by slightly higher space velocities. To confirm this conclusion, we investigated the chemical composition of red giants with high Galactic velocities and pointed out that 12 of the 14 investigated stars also have a underabundance. There exist quite a few works devoted to the abundance analysis of Galactic thin-and thick-disk stars. These works are based on spectroscopic observations of dwarfs, because their atmospheric elemental abundances undergo no changes as they evolve on the main sequence. Besides, in comparison with giants, dwarfs form a larger group and their continuum spectrum exhibits a smaller number of spectral lines, which increases the accuracy of abundance determinations. Therefore, there are comparatively few works devoted to thick-disk giants. There are even fewer studies of sodium in thick-disk giants (see, e.g., ). In these papers, the dependences of the abundance on mass and metallicity are discussed, but such an important factor as the evolutionary status is overlooked. Since the sodium abundances in stellar atmospheres do not appear instantaneously but gradually increase since the first deep mixing, the evolutionary stage of the star should be taken into account. In this paper, we perform a comparative analysis of the abundance in the atmospheres of thin and thick-disk red giants as a function of their metallicity and surface gravity, which changes during the lifetime of a star. Thus, investigating the changes in the atmospheric sodium abundance of red giants is necessary for understanding the stellar evolution, while the detected sodium underabundance in thick disk giants requires a further study. Selection of Stars We selected the objects for our observations from the Hipparcos catalogue by analyzing the Galactic velocities (U V W ) calculated using reduced Hippar- cos parallaxes and CORAVEL radial velocities. The selection criteria were the (B-V) color indices and calculated surface gravities corresponding to red giants; the calculated Galactic velocities exceeding those typical of thin-disk stars (34.5, 22.5, 18.0) km s −1, with W < 50 km s −1. The list of program stars is presented in Table I, where their ordinal numbers, HD numbers, coordinates, V magnitudes, and spectral types are given. The last two columns in Table I provide the membership probabilities of the program stars in the Galactic thin and thick disks, whose kinematic characteristics were taken from. The probabilities were calculated using formulas from. It can be seen from the table that all stars can be assigned to thick-disk objects with a high probability. Table II presents kinematic characteristics of the investigated stars. These include the Galactic velocity vector (U, V, W ) relative to the Sun and Galactic orbital elements: the perigalactic distance R min, the apogalactic distance R max, the maximum orbital distance from the Galactic plane Z max, the eccentricity e, and the inclination i. The distance to the Galactic center was assumed to be 8.5 kpc, while the necessary correction of the veloc-ities for the solar motion, (+10.2, +14.9, +7.8) km s −1, was taken from. We calculated the orbital elements through numerical integration of the stellar motion by Everhart's 15th-order method using a three-component model Galactic potential. The integration accuracy was controlled by the conservation of the necessary integrals of motion. For example, in ten orbital revolutions, the typical relative error was ∆h/h<10 −13 in angular momentum and ∆E/E<10 −8 in total energy. The errors in the space velocities (U, V, W ) were calculated from the errors in the stellar proper motions, radial velocities, parallaxes and the errors in the solar velocity components relative to the local standard of rest. We calculated the errors in the Galactic orbital elements based on the model Galactic gravitational potential using the probable errors in the stellar space velocities. It can be seen from Table 2 that all stars have a maximum orbital distance from the Galactic plane Z max > 1000 pc, which exceeds considerably the characteristic scale height for thin-disk objects, 90-325 pc (,, ). More than half of the investigated stars have orbits with eccentricities larger than 0.2. About half of the stars recede to a distance of more than 10 kpc from the Galactic center when moving in their orbits. Spectroscopic Observations The spectroscopic observations of the selected stars were performed in 2010 with a two-band echelle spectrograph attached to a 2.16-m telescope at the Xinglong station of the National Astronomical Observatories of China (NAOC). The spectrograph operated in the redband mode. The detector was a 2048x2048 CCD array on which 45 spectral orders were recorded. The in the range from 5500 to 9830spectrograph resolution was R = 40 000; the signal-to-noise ratio in the spectra was S/N > 100. The echelle package of the MIDAS software system was used for the preliminary spectroscopic data reduction, the search for and extraction of the spectral orders, the wavelength calibration using the spectrum of a thorium-argon lamp, and the spectrum normalization. The equivalent widths of the selected spectral lines were measured in the EW code that I developed. The code is a set of modules written in Perl and C, the access to which is organized via a graphical user interface. The automatic equivalent width determination module is the main one in the code. It uses the nonlinear Levenberg-Marquardt algorithm from the PDL package. For each measured spectral line with wavelength 0, the part of the spectrum 0 ± 10 0 /R in which the spectral lines are searched for is cut out. An iterative deconvolution method is applied for a better line detection; it allows the blends to be separated. More or less significant lines can be identified by varying the number of iterations. Each spectral line is fitted by a Gaussian, which is a normal approximation for most of the lines with equivalent widths < 100 m. The investigated part of the spectrum is fitted by the sum of Gaussians F () = k i=0 h i * e −(i−0) 2 / 2 i, where h i and i are the and width of the i spectral line. We take the depth of the observed line as the initial approximation for h and = 0 /R for the width. The module operation result is a set of parameters of the Gaussians that best fit the observed spectrum. The parameters of the investigated spectral line are written in a file. In addition, information about the quality of the fit and the degree of line blending is written. During its operation, the code displays a theoretical spectrum of atomic and molecular lines that provides an additional possibility for the selection of lines. DETERMINATION OF STELLAR ATMOSPHERE PARAMETERS We determined the stellar atmosphere parameters using a technique based on Kurucz's model atmospheres and analysis of the relative abundances of iron-peak elements. The technique is described in detail in ) and allows the stellar atmosphere parameters for G-K giants to be determined with an accuracy of about 70-100 K for T eff, 0.10-0.15 for lg g, and 0.10-0.15 km s −1 for V t. For late-K giants, the accuracy can be lower due to the greater influence of blending by atomic and molecular lines. In this paper, when analyzing the relative abundances of ironpeak elements when determining the stellar atmosphere parameters, we disregarded titanium, because it is well known that the abundance can be enhanced at low metallicities and for thick-disk stars. Using the derived parameters (T eff, lg g, V t ), we computed the corresponding model stellar atmospheres with the ATLAS9 code. Table III gives our estimates of the stellar atmosphere parameters (effective temperature T eff, surface gravity lg g, microturbulence V t, and metallicity ), masses, and interstellar extinctions A V. We determined the masses based on evolutionary tracks from by taking into account the stellar metallicity. The interstellar extinctions were estimated from the color excess E(B-V); the dereddened colors were calculated from calibrations based on Kurucz's model stellar atmospheres. Based on the measured equivalent widths of the selected unblended spectral lines, we estimated the elemental abundances with the WIDTH9 code. These are presented in Table IV and Fig. 1, where the open circles and asterisks denote the abundances determined from the spectral lines of neutral and ionized atoms, respectively. The list of selected lines with their characteristics and equivalent widths is available in electronic form. The cobalt abundance was determined by taking into account the hyperfine splitting effect, which can be strong in the case of cool giants. The abundance errors given in Table IV and marked by the bars in Fig. 1 were determined as the dispersion of the individual abundances calculated from individual spectral lines. As an example, the possible abundance errors associated with the determination of stellar atmosphere parameters are listed in Table V for two stars. Table V gives the number of lines used (N ) and the changes in the abundance of each element when changing individual model parameters (∆T eff =+100K, ∆lg g=+0.10, ∆V t =+0.10 km s −1 ) and the total change in abundance ∆. The sodium abundance was determined from the 6154 and 6160lines without any correction for non- LTE processes. According to, this doublet is formed deeper than other sodium lines, and the non-LTE processes do not introduce significant deviations in the abundance determination (<0.1 dex). Figure 2 are located in the same regions, i.e., the abundances of these elements do not change over the elapsed lifetime of the star. In the case of sodium, a compact arrangement is observed only for dwarfs. The trends for thin-and thick-disk stars coincide, within the error limits. In contrast, red giants are chaotically scattered and no dependence on metallicity can be distinguished. Such a behavior suggests that the abundance is determined not only by the chemical evolution of the Galaxy but also by the evolution of the star itself. The amount of sodium that was synthesized in the stellar core at the main-sequence stage and that was brought by convective flows into the stellar atmosphere is added to the initial abundance. The thick disk red giants in Fig. 2, within the error limits, are located in the region of dwarfs and exhibit no detectable sodium overabundances. However, this is most likely the selection effect, because some of the thick-disk red giants from and have significant overabundances. HD80966 with the lowest metallicity exhibits the lowest (and atypical of the remaining stars) abundance =-0.21 dex with an error of 0.02 dex calculated as the mean of the abundances from the two lines. Analysis of the possible errors by taking into account the uncertainty in the iron abundance and model parameters leads to a total error of about 0.12 dex. If we explain the low abundance by the error in determining the temperature, then it turns out that we underestimate the temperature approximately by 200 K. Such a change will lead to a significant discrepancy between the elemental abundances determined from lines with different lower level excitation potentials, while the position of the star on other plots in Fig. 2 Fig. 3, which shows the dependence on surface gravity lg g. Both these quantities are determined directly from observations. In the figure, the thin-disk stars (filled circles) show a rise as the surface gravity decreases from lg g≈3-3.5. The surface gravity is related to other fundamental stellar parameters by the relation lg g= -10.607+lg M /M ⊙ + 4lgT eff -lgL, where M /M ⊙ is the stellar mass in solar masses and lgL is the stellar luminosity. At a constant stellar mass, the decrease in lg g can be associated with a decrease in the effective temperature and an increase in the stellar luminosity. Such a behavior is typical of a star as it evolves along the red giant branch, and the appearance of sodium overabundances and the increase in sodium abundance due to the rise of sodium-enriched matter through convective flows are associated precisely with this evolutionary stage. At this stage of stellar evolution, an enrichment of matter by sodium is possible only in the regions of nuclear reactions at the main-sequence phase. The hydrogen burning reaction in the neon-sodium cycle is the main source of sodium. In the interiors of low-mass (1-3 M ⊙ ) stars, this cycle is not closed and turns into the chain of reactions 20 N e(p, ) 21 N a( + ) 21 N e(p, ) 22 N a( + ) 22 N e(p, ) 23 N a The first quantitative estimate of these reactions for supergiants is given in : depending on the temperature, from 1/3 to 2/3 of the 23 N a nuclei pass into 20 N e. At the same time, the sodium abundance in the atmospheres of F-K supergiants increases by a factor of 5 (0.7 dex), in agreement with the observational data and Fig. 3. Various conditions for the nuclear reactions of the neon-sodium cycle were also considered by and. The thick-disk red giants are marked in Fig. 3 by the large circles (from this paper, the ordinal numbers correspond to Table I) and squares (previously investigated ones). It can be seen from the figure that they are located systematically below the thin disk red giants, while for the region lg g<2 the separation between these two groups close in the number of stars exceeds the abundance errors. Thus, the difference in abundance in the atmospheres of thin-and thickdisk red giants is significant. Among the stars whose parameters were entered into the database, there are several pairs of thin-and thick-disk stars with similar parameters (T eff, lg g, ) (see ). Such stars also have similar masses and ages but differ only by the abundance and membership in different kinematic groups of stars in the Galaxy. Such a difference is not observed for the thinand thick-disk dwarfs (see Fig. 2). Consequently, the initial sodium abundance must be also the same for those stars that have presently become red giants, while the nature of the difference is in the production of sodium in the NeNa cycle. A neon underabundance may exist in thick-disk objects. This is hard to check, because the neon lines are difficult to observe. However, the main 20 Ne isotope is an -process element, and its behavior with metallicity must follow the trends of other -elements, i.e., an increase in abundance with decreasing metallicity. In this case, we should have seen higher overabundances of sodium formed from neon in thick-disk giants, but the reverse is true. The 22 N e underabundance is a possible cause. Indeed, point out a signifficant reprocessing of 22 N e into 23 N a, while the initial 22 N e/ 23 N a abundance ratio in dwarfs is about 4.1. In any case, the difference between the atmospheric abundances of thin-and thick disk red giants requires an explanation and invoking theoretical calculations. CONCLUSIONS Thus, based on spectroscopic observations, we determined the stellar atmosphere parameters for ten thick-disk red giants. We confirmed the difference in atmospheric abundance between thin-and thick-disk giants. Its nature may be related to the neon isotope abundance difference and this hypothesis requires verification. |
Plant toxicity testing to derive ecological soil screening levels for cobalt and nickel Phytotoxicity tests were performed to set ecological soil screening levels for cobalt (Co) and nickel (Ni) following the American Society for Testing and Materials international E196398 Standard Guide for Conducting Terrestrial Plant Toxicity Tests. Two soils (a modified artificial soil mixed with 5% organic matter, pH 5.01, and a native riverine sandy soil with 0.1% organic matter, pH 6.3) were treated with cobalt(II) chloride or nickel chloride and allowed to age for four weeks before initiating tests. Alfalfa, barley, radish, perennial rye, and brassica were used to determine the appropriate range of concentrations and to select the most sensitive plant species for definitive tests. The tests were designed to have one to three test concentrations below the 20% effects concentration (EC20), and five to six test concentrations above the EC20. Definitive tests for each chemical used two soil matrices, three plant species, and replicates at 10 nominal concentrations, including negative control. Soil chemical concentrations were determined before planting and on completion of the phytotoxicity tests. Threshold responses interpreted as the EC20 for each species endpoint were calculated from regression analyses. The geometric mean of the EC20 values (excluding emergence, mortality, and nodule numbers) for each species resulted in values of 30.6 mg/kg for Co and 27.9 mg/kg for Ni. |
<gh_stars>10-100
package com.cathive.fx.guice.example;
import java.util.ResourceBundle;
import javafx.event.ActionEvent;
import javafx.fxml.FXML;
import javafx.scene.control.Button;
import javafx.scene.control.Label;
import javax.inject.Inject;
import javax.inject.Named;
import com.cathive.fx.guice.FXMLController;
import com.google.inject.Injector;
import com.google.inject.Key;
import com.google.inject.name.Names;
/**
* Example FXML controller class.
* <p>This class will be automatically bound be Guice and can be used in
* you FXML's "fx:controller" definition.</p>
*
* @author <NAME>
*/
@FXMLController
public final class FxmlExampleAppController {
@Inject
private Injector injector;
@Inject
@Named("i18n-resources")
private ResourceBundle resources;
@FXML
private Button redButton;
@FXML
private Button greenButton;
@FXML
private Label statusLabel;
@FXML
void redButtonPressed(final ActionEvent e) {
statusLabel.setText(resources.getString("ON_RED_BUTTON_PRESSED"));
}
@FXML
void greenButtonPressed(final ActionEvent e) {
statusLabel.setText(resources.getString("ON_GREEN_BUTTON_PRESSED"));
}
/**
* This method will automatically be called by the FXML loader.
* See Oracle's JavaFX/FXML documentation for further details about
* that mechanism. (Note: it was necessary to implement {@ link Initializable})
* int the past, but that mechanism has been deprecated.
*/
public void initialize() {
// Use the guice injector to fetch our color code definitions.
final String red = injector.getInstance(Key.get(String.class, Names.named("red-button-color-string")));
final String green = injector.getInstance(Key.get(String.class, Names.named("green-button-color-string")));
// We use the Guice-injected value for our desired color to update
// the style of our FXML-injected button.
redButton.setStyle(String.format("-fx-base: %s;", red));
greenButton.setStyle(String.format("-fx-base: %s;", green));
}
}
|
<filename>server/src/test/java/org/elasticsearch/plugins/spi/NamedXContentProviderTests.java
/*
* Licensed to Elasticsearch under one or more contributor
* license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright
* ownership. Elasticsearch licenses this file to you under
* the Apache License, Version 2.0 (the "License"); you may
* not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing,
* software distributed under the License is distributed on an
* "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
* KIND, either express or implied. See the License for the
* specific language governing permissions and limitations
* under the License.
*/
package org.elasticsearch.plugins.spi;
import org.elasticsearch.common.ParseField;
import org.elasticsearch.common.io.Streams;
import org.elasticsearch.common.xcontent.NamedXContentRegistry;
import org.elasticsearch.search.aggregations.Aggregation;
import org.elasticsearch.search.aggregations.pipeline.ParsedSimpleValue;
import org.elasticsearch.search.suggest.Suggest;
import org.elasticsearch.search.suggest.term.TermSuggestion;
import org.elasticsearch.test.ESTestCase;
import java.io.IOException;
import java.io.InputStream;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.List;
import java.util.ServiceLoader;
import java.util.function.Predicate;
public class NamedXContentProviderTests extends ESTestCase {
public void testSpiFileExists() throws IOException {
String serviceFile = "/META-INF/services/" + NamedXContentProvider.class.getName();
List<String> implementations = new ArrayList<>();
try (InputStream input = NamedXContentProviderTests.class.getResourceAsStream(serviceFile)) {
Streams.readAllLines(input, implementations::add);
}
assertEquals(1, implementations.size());
assertEquals(TestNamedXContentProvider.class.getName(), implementations.get(0));
}
public void testNamedXContents() {
final List<NamedXContentRegistry.Entry> namedXContents = new ArrayList<>();
for (NamedXContentProvider service : ServiceLoader.load(NamedXContentProvider.class)) {
namedXContents.addAll(service.getNamedXContentParsers());
}
assertEquals(2, namedXContents.size());
List<Predicate<NamedXContentRegistry.Entry>> predicates = new ArrayList<>(2);
predicates.add(e -> Aggregation.class.equals(e.categoryClass) && "test_aggregation".equals(e.name.getPreferredName()));
predicates.add(e -> Suggest.Suggestion.class.equals(e.categoryClass) && "test_suggestion".equals(e.name.getPreferredName()));
predicates.forEach(predicate -> assertEquals(1, namedXContents.stream().filter(predicate).count()));
}
public static class TestNamedXContentProvider implements NamedXContentProvider {
public TestNamedXContentProvider() {
}
@Override
public List<NamedXContentRegistry.Entry> getNamedXContentParsers() {
return Arrays.asList(
new NamedXContentRegistry.Entry(Aggregation.class, new ParseField("test_aggregation"),
(parser, context) -> ParsedSimpleValue.fromXContent(parser, (String) context)),
new NamedXContentRegistry.Entry(Suggest.Suggestion.class, new ParseField("test_suggestion"),
(parser, context) -> TermSuggestion.fromXContent(parser, (String) context))
);
}
}
}
|
<filename>Tests/QtAutogen/RerunMocOnMissingDependency/MocOnMissingDependency/inc2/foo.h
#include <qobject.h>
|
Graphene - a rather ordinary nonlinear optical material An analytical expression for the nonlinear refractive index of graphene has been derived and used to obtain the performance metrics of third order nonlinear devices using graphene as a nonlinear medium. None of the metrics is found to be superior to the existing nonlinear optical materials. Nonlinear optics, tracing its origins to the invention of laser in 1960's has been an exciting field in the last 50 plus years: significant advances have been made in understanding nonlinear phenomena, and a number of practical applications have emerged, such as in frequency conversion and optical parametric generation for the second order nonlinear phenomena, and mode locking, comb generation, and a variety of nonlinear spectroscopic tools for the third order nonlinear effects. Yet the most enticing promise of nonlinear optics -that of controlling light by light and hence all optical switching and computing --remains unfulfilled because the nonlinear susceptibilities of the host of materials that have been explored to date are simply too small to support efficient all-optical switching. Efficient switching requires one to achieve either a nonlinear phase shift commensurate with 180 degrees (or absorption change by 90% or so) over a distance that is not much longer than one absorption length, and, when it comes to fast (say sub-picosecond) nonlinearities, the best results have been obtained with optical fibers, but only due to long propagation length, which makes these schemes impractical. Although all-optical switching has been demonstrated in a variety of waveguides, such as silicon or chalcogenide, the relatively large length and high power requirements prevented these materials from becoming practical. The nonlinear index of the best nonlinear materials in the optical/near IR range is on the order of 13 2 2 10 / n cm W − ≤ which follows from very simple considerations: the electrons are confined within the bonds by a potential that is parabolic near the equilibrium and becomes nonparabolic only when the externally applied field approaches the intrinsic field whose magnitude is on the scale of 8 10 / i E V cm which immediately provides the right scale for When it comes to free electrons in either free space or condensed matter the situation is analogous. Low energy electrons are all subject to the parabolic dispersion relation between the energy and momentum, this relation becomes non-parabolic (and nonlinearity ensues) only when the energy of electron approaches some "threshold value", e.g the Fermi energy in metals, bandgap energy in doped semiconductors or relativistic energy mc 2 for free electrons. As a result, the nonlinear index even in narrow bandgap materials, such as InAs is insufficient to achieve switching in less than the absorption length. It is hardly surprising then, that each time a new class of materials becomes available, there is immediately a healthy impulse to explore the possibility that in these materials optical nonlinearities, often referred to as "giant", will exist and the old curse of "insufficient for optical switching ratio of n 2 to the absorption coefficient" will be finally lifted. This had been the case with polymers in 1970's, semiconductor quantum wells and superlattices in 1980's, quantum dots in 1990's, and, more recently, with a multitude of metal-incorporated plasmonic and meta-materials. Each time though, the initial enthusiasm has waned when faced with indisputable realities of nonlinearity being controlled by a very few parameters (essentially only the bandgap, frequency, and optical transition matrix elements), and, in the end, not being sufficient for all optical switching due to the small total phase shift, excessive absorption loss, optical damage, or, most often, the combination of all three. To this day, the list of practical materials for all optical switching remain relatively short, and none of them works sufficiently well to develop low power all-optical switches with a footprint small enough for applications in integrated optics. Thus, it was expected that once graphene, a new form of carbon, appeared on the stage in early 2000's, its nonlinear optical properties would be explored and claims of "giant" nonlinearity would be made. Indeed, one did not have wait long before a large number of reports of extraordinary large nonlinearity in graphene appeared. Huge values of nonlinear susceptibility were estimated and even measured, and predictions of various switching schemes, including single photon optical switching were made, but when it comes to practical optical switching, progress has been less obvious, and, other than passive mode locking of lasers, which does not require large change in index or absorption, there is no report of working graphene devices with performance exceeding that of more conventional materials. It is the goal of this paper to establish whether graphene is indeed a better material when it comes to ultrafast all-optical switching. It is important to stress that, for all optical switching, a large value of nonlinear index of refraction is definitely not enough. The one and only figure of merit is the maximum amount of phase shift that can be attained over a distance that is less than one absorption length. Two processes can limit this figure: first, and most obvious, the optical absorption itself, and second, saturation of the nonlinearity. And both of these processes play important roles in graphene. Optical transitions in graphene can be both interband and (in doped graphene) intraband and both types engender nonlinearities. The first thing that can be said about the interband nonlinearity is that it has absolutely nothing to do with Dirac-like dispersion of electrons and is no different from the optical nonlinearity of any 2D electron systems excited far above the bandgap, as in, for instance SESAM saturable absorption. In fact in both and the nonlinearity scales linearly with the number graphene monolayers, even though Dirac dispersion no longer exists after two monolayers. The similarity between the interband transitions in graphene and any narrow gap semiconductor is evident if one simply takes a look at the dispersion in the conduction band of, say InAs away from the bandgap. It is almost linear with the slope, i.e.the electron velocity approaches 1.110 8 cm/s comparable to (in fact larger than) graphene, as shown in Fig1.a. Clearly, the behavior of the electron with a high enough energy, say 1eV in the band cannot be affected by the presence (or absence) of the bandgap. The nonlinear index for the interband transitions can become very large due to the resonant enhancement, but so does the absorption coefficient. In the equivalent 3D nonlinear refractive index for the graphene was found to be as high as absorption and nonlinear susceptibility will be equally diluted. This result is easily predictable since according to Eq in the ratio between linear interband sheet conductivity of and the third order sheet conductivity is roughly ( is the same "intrinsic field" that is characteristic of any nonlinear process. Since nonlinear phase shift is proportional to graphene may find a niche as a fast saturable absorber for mode-locking but due only to its ability to operate over wide range of frequencies, particularly long wavelengths, and not because of better performance than say SESAM. Let us now turn our attention to the interband optical nonlinearity in doped graphene, shown schematically in Fig.1b ). The sheet current density J also saturates as shown in Fig.2d for two values of current density. With higher carrier density current is larger and it saturates at larger values of k 0, i.e. at higher field. This can be also seen from Fig2e where the sheet conductivity normalized to the frequency To demonstrate, we first introduce the linear 2D susceptibility of the graphene sheet as Then the entire field dependent susceptibility can be written as Two caveats should be mentioned: first of all, to avoid excessive interband absorption the condition 2 F E < applies (even though in reality, especially at room temperature, optical phonon assisted mid-gap absorption will limit the range of frequencies for well below 2 F E ). Also, at low frequencies the free carrier scattering will set in, effectively placing a lower boundary on the frequency range, and, in all our expression one should use 2 2 − + in place of 2, where ~100 fs is the scattering time. The results for the nonlinear refractive index are shown in Fig. 3a, for the combinations of frequencies and doping densities (from 10 10 to 10 13 cm -2 ) that allow relatively low loss propagation. As one can see the numbers are quite impressive, particularly in the mid-IR range of frequencies from 10 to 30THz where the nonlinear index can be as high as 10 -4 cm 2 /W at lower doping densities. Of course, a significant part of this "giant" nonlinearity is simply due to division by the small (.33nm) thickness of the graphene monolayer. If one considers waveguide geometry, then the effective nonlinearity is somewhat less I -beyond that the change of index essentially saturates, just as in any nonlinear material. The phase shift achieved in N l layers of graphene is then Essentially, all the electrons residing further away from the Dirac point hardly contribute to nonlinearity at all, but they are needed to keep the Fermi level high enough to mitigate the loss due to band-to-band absorption. Clearly, absence of a real bandgap (rather than one induced by blocking the absorption) is a handicap that seems to have no remedy. It is also important to mention here that we have considered only the low temperature case. It is easy to see that at higher temperatures nonlinearity is reduced as carriers are promoted farther away from the Dirac point while the residual absorption increases due to phonon-assisted processes. In the end, the nonlinear optical properties of any material depend on a very limited set of parameters -transition dipole matrix elements, density of states, and scattering (dephasing) rates, and none of these parameters is strikingly different in graphene from other materials, hence, in a hindsight, one should not be surprised that when it comes to practical nonlinear optical devices graphene exhibits no particular advantage over more conventional materials, such as SiN, Si, or chalcogenides. That does not preclude graphene from being used in a few specific niches, such as in a saturable absorber for mode-locked lasers, but that is due mostly to its easy availability for a particular wavelength range rather than to superior nonlinear figure of merit. In conclusion we have developed an analytical expression for the nonlinear index of graphene and using it have estimated the doping densities, optical powers and number of layers (normal geometry) or interaction length (waveguide geometry) required to achieve a high efficiency of third order nonlinear effects at different wavelengths. The results do not show any advantages held by graphene over other nonlinear optical materials when it comes to the practical figures of merit, but one also cannot say that graphene is dramatically inferior to the other materials. Whether or not graphene becomes a material of choice will probably depend on considerations of mechanical and thermal robustness, manufacturability and cost. The Author acknowledges support of NSF DMR-1207245 |
Syndrome of nonverbal learning disabilities: Age differences in neuropsychological, academic, and socioemotional functioning Abstract Previous research has suggested that changes in the manifestations of the nonverbal learning disabilities syndrome (NLD) occur over the lifespan and that they do so in a manner that is consistent with the tenets of the NLD model. Although the model would predict that age-related changes would also be evident within the childhood years, no study has yet examined this possibility. Based on the tenets of the model, specific predictions were formulated regarding developmental changes in the features of the NLD syndrome that would be expected to occur across the middle childhood and early adolescent years. The pattern of neurocognitive and socioemotional changes observed within the context of the cross-sectional data provided strong support for the predictions. Due to methodological limitations, no firm conclusions regarding the developmental manifestations of the NLD syndrome could be derived from the results of the longitudinal study. At most, these results suggested that some improvements in areas of neurocognitive deficiency may occur with the implementation of an appropriate remedial intervention program. |
Staphylococcus aureus colonization in Qazvin University hospitals healthcare workers Abstract Background: Staphylococcus aureus (SA) colonization of hospital personnel is a source of hospital acquired infections. Objective: The aim of this study was to determine the prevalence of nasal carriage rate of SA and methicillin resistant Staphylococcus aureus (MRSA) among Health Care Workers (HCWs) at Qazvin university hospitals. Methods: A cross sectional study was conducted among 396 employees of five teaching hospitals from October 2016 to April 2017. After obtaining informed consent and completion of the questionnaire, a sample was taken from the anterior nasal cavity for microbiology. The isolation of SA and their antimicrobial sensitivity were carried out by standard bacteriological procedures (disk diffusion and E-Test method). MRSA were confirmed by cefoxitin disk diffusion test. Chi square and independent t test were used to analyze the collected data. Findings: From the 198 HCWs, 32 people (16.1%) carried SA that the most carriers were workers of intensive care units (20.3%). 3% of all HCWs were identified as MRSA carriers. Colonization with SA is significantly lower among nursing and higher education. All SA isolates were sensitive to vancomycin and rifampin. Conclusion: The rate of nasal SA (especially MRSA) carriage among HCWs of Qazvin university hospitals is low. Also, staff teaching appears to be a promising approach for reducing nasal carrier. Rifampin and mupirocin, for eradication of Staphylococcus colonization in health workers (even MRSA) are acceptable. |
Short, Medium and Long Range Spatial Correlations in Simple Glasses Local stresses and pressures always exist in glasses. In this letter we consider their effects on the structure and structural correlations in simple glasses. We find that extreme values of local pressures are related to well defined local structures. The correlations related to these extreme stresses extend to full system size and decay as a power law with the distance. This result is especially striking, since at large scales, the total density fluctuation exhibits exponentially damped decay similar to the decay in simple liquids. Thus at medium and large distances, the atoms with extreme values of local pressures exhibit higher degree of correlation than the rest of the system. These results were found for glasses with very different short range structure, indicating their general nature. The nature of correlations in amorphous solids and super cooled liquids poses a fascinating question that has received a lot of attention (see e.g. ). While in crystals there is a clear cut definition of structural correlations that uses the period of the crystalline lattice, how to define structural correlation in glasses and liquids is an open question that may have various answers. A related question is which physical quantity should be used in looking for correlations. There are various possibilities. The simplest one is to analyze the average number of atoms in a distance r from a central atom, using the radial distribution function (RDF). For simple liquids with a known atomic pair potential, the RDF can be calculated from integral equations supplemented by a closure relation and requirements for thermodynamic consistency. The same methods cannot be applied for the glassy phase, because glasses are non ergodic systems. The RDFs describe average radial correlations and thus fail to touch the finer details of the structure for glasses and liquids. They tend to average out variances in the local structure. A second option is to find some strongly varying local property of the system, like the local stress, distribution of energy wells, Voronoi volume or another local structural parameter, and use it in the calculation of conditional correlation functions. An obvious question in such case is whether these parameters are manifestations of random statistical noise or do they exhibit a more basic nature of the glass. Local stresses and pressures always exist in glasses. They result from variations in the local environments, which persist even in a zero-temperature glass, where all the local forces on the atoms vanish. The average stress and pressure might be zero, but the local stresses have a wide distribution. The glass can be described as a discrete atomic network in which the stresses balance each other throughout the network. To understand such a network one should study the correlations between the stressed configurations. The role of the most compressed/stretched sites is an important issue. We show that they are manifestation of the basic structure of glasses. Thus they are relevant for classification of the local structural units and the degree of randomness in a glass. The characterization of stressed networks is naturally complicated since they involve tensorial description and statistical noise. However, even without knowledge of the full details of the network, one can use the local stresses to obtain an elastic and structural description of a glass. An appropriate scalar, such as the trace of the site stress tensor or the local pressure, can be used as a parameter for the study of correlations in the glass. In this letter we present our study of the effects of extreme values of local pressures on structural correlations in simple glasses. For mono-atomic systems where atoms interact through a pair potential U (r ij ), where r ij is the distance between atoms i and j, we define the term as "local effective pressures" (LEP). J's so defined are related to an effective pressure p ef f i on a sphere of radius r around atom i by p ef f i = J i /(4r 2 ). Since the local pressures depend on the choice of the cell, using J's as the conditional parameter enables us to investigate directly the effect of simple elastic features on structural correlations. Unlike the forces f i = j ∇U (r ij ) that vanish at zero temperature, LEP as well as local pressures do not vanish in amorphous solids, even when the total (bulk) pressure is zero. The probability distribution of J is generally quite wide, although most of the atoms feel relatively small pressures (see examples in ). The conditional radial pair distribution function n J (r) is defined as the average number of neighbors at distance r for atoms with a given J. The extra structural correlations, induced by stresses, can be calculated from the function where n(r) is the the average number of neighbors at distance r. This provides a measure of the difference between the stressed sets and the total set. In this letter we study two different simple mono atomic glasses: a Lennard-Jones glass (LJ) and a glass defined by an IC potential, which is a modified LJ potential, designed to favor icosahedral local order. The IC glass has more pronounced structural ordering than the LJ glass. Furthermore, while the LJ glass does not have a well defined glass transition, it is well defined for the IC glass. These differences leads one to expect differences in the organization of the stress network. We prepared systems of 21952 atoms using molecular dynamics simulations for both glasses. Periodic boundary conditions were assumed. Both potentials were studied extensively by numerical simulations. Details about systems' parameters and preparation are given in for LJ. The parameters of IC potential are given in with density as in. The initial LJ liquids were prepared and equilibrated at T = 1. LJ glass was prepared by steepest decent quench to zero temperature glass. The initial IC liquids were prepared and equilibrated at T = 1.6. Glasses at various temperatures were prepared by coupling to a heat bath and then equilibrating the systems till the systematic decrease in energy stops. To subtract the effects of thermal noise, all the glasses where quenched to zero temperature by steepest descent procedure. We used 4 independent systems of IC glass and 3 independent systems of LJ glass. Results reported below are similar for all systems. All the quantities are expressed in the LJ reduced units and the distances are given in atomic distances. First, we briefly discuss the density correlation function (r) = n(r)/4r 2. It is well known that in simple liquids, long range decay of (r) is (e.g. ) where 0 is the bulk value, 0 is decay length and 1 is the wavelength of the decay. For simple liquids, 0 depends on the temperature. We calculated numerically the variations of (r) for both potentials in the liquid phase and in zero temperature glass. In liquid phase, after the first coordination shell, (r) decays in accord to eq. 3. Fig. 1 shows the variations in log((r) − 0 ) for both potentials in the zero temperature glass. After a few coordination shells, the glassy correlations exhibit decays of the type of eq. 3 with 1 ≈ 1.12 2, 0 ≈ 0.56 for IC glass and 1 ≈ 1.14 2, 0 ≈ 0.62 for LJ glass at T = 0. A comparison with the liquids at T = 1 shows that the differences in the wavelength 1 are very minor, and the variations in the decay length 0 are about 20%. Thus it is apparent that density fluctuations at longer scales in glasses are similar to those in liquids. This result is in accord with the view that 'inherent' glasses and liquids have a very similar structure but the temperature disguises it. The main qualitative effect of the temperature on (r) is to smooth structural details in the 2-4 coordination shells. For similar results on different systems see in. As stated before, the probability distribution of J is quite wide. Its width and shape are similar for the LJ and the IC glasses. To evaluate the effects of the stresses we calculated n J (r) for the extreme parts of the J distributions (4 percent of the particles). As shown in Fig. 2, the first 3 peaks of the conditional distributions are considerably narrower than the peaks of n(r). The peaks are shifted relative to each other by about 0.1 atomic distance. This shift in the position is a result of the different compression of the environments. The shift toward smaller r for small J's indicates compressed local environments whereas the shift toward larger r's for large J's indicates stretched local environments. The radiuses of those environments are consistent with effective potential estimates. The shifts in r in the first coordination shell are analog to variations in volume in a Voronoi scheme. The apparent differences between LJ and IC glasses ( Fig. 2a and 2b) are a manifestation of the different finer structural details. For the LJ glass, the n J (r) look similar but shifted regardless of the values of J, indicating the absence of a well defined local structure. However, in the IC glass case, the width of the first peak is much narrower for small Js than for the larger ones (a factor of about 2). The structural origin of this becomes even more apparent when the ratios between the positions of the 1 st and 2 nd peaks in n J (r) are calculated. In the LJ glass the ratio is ≈ 1.7 for both high/small values of J. In the IC glass, the ratio is ≈ 1.7 for small Js and ≈ 1.63 for large J's. The ratio of 1.7 does not suggest any specific local structure, although it is very pronounced in IC glass, where atoms barely appear in the interval between the peaks. The 1.63 ratio corresponds to the distances between atoms on the faces of an Icosahedron! Thus the non-compressed environments are consistent with atoms that sit on the faces of empty icosahedra. Those icosahedra are the voids with radius of one atomic distance observed in ! On the other hand, the compressed environments have an almost full icosahedral nature, i.e. there are exactly 12 neighbors to an atom, and almost all atoms in the first neighbor shell have five nearby neighbors. The neighbors in the second peak of the compressed environments, are very sharply defined (the ratio between the stressed and compressed 2 nd peaks is about 3.) They are held by the more diffuse, stretched empty icosahedral packing. The stretched and compressed environments are spread throughout the system. Since the average number of neighbors in the glass is about 12, the two sets and their first neighbors span the whole glass. Thus the two types of extremely stressed environments define all of the glass. Note that in both glasses, the average nearest neighbor distances change as a function of the distance from these sets of atoms. Those changes happen up to second/third peak position in order to balance compressed environments with stretched ones and vice versa. This is a result of radial gradients of the stress around these atoms. The tensorial nature of the glasses is manifested in the adjustment of the local distances. The next obvious question is whether these special environments are correlated at a longer range. To answer this question, we use the functions dn J (r), defined in eq. 2 for extreme values of J, to identify variations from the 'average structure'. In Figure 3 we present dn J (r) for the compressed sites of the LJ and the IC glasses. This difference indicates the existence of excess correlations between the subset J and the total lattice. Similar results have been obtained for the stretched subsets except for a phase shift in the position of the peaks. We checked that for randomly chosen sets of atoms one observes no correlations at all in dn J, as is indeed expected. The intensity of the difference in short distances is larger by a factor of two in the IC glass. The correlation lengths can be seen from the integrals of dn J. In the IC glass they are constant up to half of the system size (14 atomic units). On the the LJ case the correlations decay from a scale of 2-3 until they are randomized at a distance of 10. We have already noted that while in the IC glass there is a distinct midrange structure, it is absent in the LJ glass. This is the probable cause for the cutoffs in the correlations in the LJ glass. The limited range of excess correlations in the LJ glass is also consistent with the lack of glass transition in the LJ glass. The additional structure and additional correlations can also be observed in the Fourier transform which also reduce the statistical noise. We performed Fourier transforms on the sets using dn J (q) = j exp(iqr j )dn J (r j ). In Figure 4 we present |dn J (q)| 2 for lowest values of J for LJ and IC glasses. Very similar figures were obtained for |n J (q)| 2 for highest values of J. A sharp pronounced peak at q = q peak exists is both glasses indicating that the period of the oscillations is about one atomic distance. In the IC glass, |n J (q)| 2 exhibits additional structure at q > q peak which is absent in the LJ glass. The intensity of the peaks in the IC glass is much stronger. Inverse transforms confirmed the previous ranges of the correlations. The normalized conditional densities dn J (r)/(4r 2 ) decay with the radius as r −2. The interpretation of these results is that though the total correlations in the system are randomized there are long range correlations within the stressed networks. A possible explanation for such correlations is the existence of correlated chains. In this letter we demonstrated how additional structural features, which are not apparent in n(r) due to its wide peaks, emerge when one considers n J (r). For two simple glasses with very different short range structure, we have shown that whereas the averaged structural correlations have exponentially dumped decay analog to dense liquids, stronger and extended structural correlations are revealed throughout the system when one considers finer measures using LEP. We have demonstrated the existence of two critical subsets of atoms in the structure. These subsets are not random structures but a result of interrelations between local structure (and packing) and local elastic features (see also ref. ). We have shown that those sets induce long range structural correlations which decay much slowly than the total radial distribution function. These finding suggest a fascinating picture of simple glasses. There are natural extensions of this study. First, using LEP one can proceed to study more complicated glasses, where the need to identify structural correlations is even more acute. One might expect that for stronger type of local organization there might be stronger and longer range correlations. It is thus interesting to study different kind of potentials to see what are the universality classes of the decay of the conditional correlations. It is natural to assume that atomic diffusion near the glass transition will be dominated by movements in the stretched environments. This is obviously related to the existence of free volume due to the icosahedral voids. The existence of correlations between these states suggest that atomic diffusion, in temperatures slightly above the glass transition, will occur as a correlated set of jumps in the stressy network. Below the glass transition such correlated movements are inhibited. Since one expect LEPs to vary smoothly at the glass transition, this parameter is especially convenient for study and comparison of structural correlations in glasses in a wide range of temperatures as well as supper cooled liquids near glass transition. |
The invention relates to a brake pad for a disc brake of a vehicle, in particular a commercial vehicle, having a lining carrier plate which is configured as a cast part and having a friction lining which is fastened in the lining carrier plate.
In addition to a brake disc, disc brakes for vehicles, in particular for commercial vehicles, also include two brake pads, which can be pressed against the friction faces of a brake disc when required, that is to say in the case of a braking operation.
Here, each brake pad includes a lining carrier plate and a friction lining which is fastened thereto and bears frictionally against the associated friction face of the brake disc in the use position.
Special significance is given to the connection of the friction lining to the lining carrier plate, since high loads occur resulting from jolts caused by operation and from frictional heat produced during braking.
In order to produce a correspondingly fixed connection which, in addition to the above-mentioned jolt and thermal loading, can absorb bending and shearing loads that occur, it is known, for example, from DE 195 32 019 C1, to integrally form positive connection mechanisms in the form of round or angular, lug-shaped elevations on a basic body of the lining carrier plate. In order to anchor the friction lining, in particular in order to absorb shearing forces, the positively connected parts are provided with undercuts, which can be produced, however, only in a very complicated process. Here, the friction lining is held positively only on the positively connected parts, with the result that there is the risk that, if the friction lining tears, parts are released from it, which not only limits the functional capability of the disc brake overall to a great extent, but possibly leads to its failure, since the released parts can jam in the brake caliper and impede the displacement movement of the brake pads and/or the brake disc.
Moreover, the production of the known brake pad is very expensive, above all also for the reason that the mold bodies which are used in the mold and include an elastomer for the formation of the positive connection mechanisms of the lining carrier plate are subjected to high wear and, therefore, have to be replaced frequently. Special significance is given to this circumstance precisely because brake pads of this type are used in large quantities.
The invention is therefore based on the object of further developing a brake pad such that it can be produced less expensively while improving its functional reliability.
According to the invention, a brake pad for a disc brake of a vehicle, in particular a commercial vehicle, has a lining carrier plate which is configured as a cast part, and has a friction lining, which is fastened in the lining carrier plate. The lining carrier plate has a flat-shaped basic body and elevated or raised positively connected parts, which are integrally cast thereon and are enclosed by the friction lining. At least one part of the positively connected parts is surrounded, at least in regions, by an adjoining depression of the basic body.
As a result of this structural embodiment of the brake pad, both less expensive production and an improvement in the adhesion of the friction lining material on the basic body are possible.
Since undercuts which are to be introduced in a targeted manner are dispensed with, the lining carrier plate can be cast in a substantially simpler and less expensive manner. This relates to both the casting itself and preparatory work, such as the production of a sand mold and the like, and also to the unproblematic production of the casting mold which is then possible and in which the use of correspondingly elastic mold means can be dispensed with, so that a considerably higher service life of the corresponding model results. Overall, the invention leads to a substantial cost reduction of the brake pad, which cost reduction has a particular significance in view of the series production of such brake pads.
Although the introduction of predefined undercuts is dispensed with, the production-related roughness of the cast surface on its own leads to interlocking of the friction lining with the elevated, positively connected, parts, just as otherwise with the surface of the basic body.
The embedding of the friction lining into the depressions of the basic body which surround at least a part of the positively connected parts at least in regions ensures additional security for holding the friction lining on the lining carrier plate, which depressions make possible improved absorption of shearing forces that occur and additionally guarantee improved securing of friction lining parts if the friction lining is destroyed, for example, by the formation of cracks. That is to say, the friction lining parts continue to adhere to the carrier plate even when the brake is not in engagement, that is to say, is released.
The shape, dimensions and arrangement of the positively connected parts can be of different designs. Here, the configuration of the positively connected parts is substantially dependent on the force absorption which occurs during braking, the temperature which is produced, and the properties of the friction materials used (which can be different as a result of various production processes and compositions).
For instance, the positively connected parts can be integrally molded in the form of round or rectangular pins or pegs, which are arranged offset with respect to one another, in a row or extending obliquely in one or more directions. It is also contemplated to configure the positively connected parts as meandering webs, which are integrally cast such that they extend parallel to one another, for example.
Further advantageous embodiments of the invention are described and claimed herein.
Other objects, advantages and novel features of the present invention will become apparent from the following detailed description of one or more preferred embodiments when considered in conjunction with the accompanying drawings. |
Gene-Gene Interaction Analysis for the Survival Phenotype Based on the Kaplan-Meier Median Estimate In this study, we propose a simple and computationally efficient method based on the multifactor dimensional reduction algorithm to identify gene-gene interactions associated with the survival phenotype. The proposed method, referred to as KM-MDR, uses the Kaplan-Meier median survival time as a classifier. The KM-MDR method classifies multilocus genotypes into a binary attribute for high- or low-risk groups using median survival time and replaces balanced accuracy with log-rank test statistics as a score to determine the best model. Through intensive simulation studies, we compared the power of KM-MDR with that of Surv-MDR, Cox-MDR, and AFT-MDR. It was found that KM-MDR has a similar power to that of Surv-MDR, with less computing time, and has comparable power to that of Cox-MDR and AFT-MDR, even when there is a covariate effect. Furthermore, we apply KM-MDR to a real dataset of ovarian cancer patients from The Cancer Genome Atlas (TCGA). Introduction In this era of precision medicine, one of the main goals of human genetics is to understand the biological relationship between diseases and their treatment so that each patient receives the best treatment based on his or her genetic and/or environmental exposures. To achieve this goal, it is necessary to identify the genetic or environmental factors associated with various diseases. With the recent development of high-throughput technologies, information has been made available on a large number of genetic variants, such as single-nucleotide polymorphisms (SNPs), and genome-wide association studies (GWAS) have successfully discovered many susceptibility genes associated with various diseases. As of June 20, 2019, the GWAS catalog contains 4,038 publications and 118,709 associations (http://http://www.ebi.ac.uk/gwas). However, the SNPs identified in GWAS have limitations in explaining missing heritability. One possible way to overcome this problem is to identify the effects of gene-gene interactions and/or gene-environmental interactions on complex diseases. The multifactor dimensionality reduction (MDR) method has been widely applied to identify gene-gene interactions (GGIs). The main idea of MDR is to reduce the dimensionality of multilocus information into one-dimensional binary attributes by pooling multilocus genotypes into either a high-risk group or a low-risk group. Then, cross-validation is used to evaluate the ability of the generated binary variables to classify and predict outcomes. Since its first introduction for binary traits, numerous studies have modified and extended the original MDR method. Log-linear modelbased MDR (LM-MDR) improved factor combinations using log-linear models and reclassifications of risk, whereas odds-ratio-based MDR (OR-MDR) used odds ratios instead of naive classifiers. MDR methods for imbalanced data and incomplete data have been also developed. For continuous traits, the generalized MDR (GMDR) method was proposed to extend the MDR algorithm to be applicable to both dichotomous and continuous phenotypes by using the residual score of the generalized linear model as a new classifier. The model-based MDR (MB-MDR) method was suggested as a flexible way to consider covariates or continuous traits based on a regression model. In addition, quantitative MDR (QMDR) was proposed as a simple and efficient way to treat continuous traits. QMDR classifies multilocus genotypes into either a high-or low-risk group by comparing the mean value of each multilocus genotype to the overall mean and then uses the t test statistic as a score to determine the best model. As multivariate extensions of QMDR, multi-QMDR and multi-CMDR based on Hotelling's T 2 statistic were also proposed. However, there have been relatively few attempts to develop statistical methods to identify GGIs in the context of survival analysis. In cancer research, survival time has served as an important phenotype in association studies that have investigated genetic factors as well as environmental and clinical variables. Therefore, it is more informative to consider the censored survival phenotype than simply to treat the phenotype as a binary variable of death or survival. The Surv-MDR method was first proposed to handle censored time-to-event data, in which log-rank test statistics are calculated to compare the survival time between samples with and without the specific genotype combination for each multilocus genotype combination. If the log-rank statistic is positive, the corresponding genotype is labeled as highrisk, and if not, it is classified as low-risk. Once all genotypes are classified as high-or low-risk, a new binary attribute is defined by pooling the high-risk genotype combinations into one group and the low-risk genotype combinations into another group. Next, the log-rank test is applied to compare these two survival curves, and the square of this log-rank statistic is used as the score to choose the best model, while the remaining cross-validation procedure is the same as done in the traditional MDR method. Although Surv-MDR uses a computationally simple log-rank test, 3 q log-rank test statistics should be calculated for each SNP combination for a q th -order interaction, which requires intensive computing time. Additionally, as extended versions of GMDR for the survival phenotype, both Cox-MDR and accelerated failure time MDR (AFT-MDR) were proposed as methods in which the residual score of the generalized linear model is replaced with the appropriate measures corresponding to the survival models. In other words, Cox-MDR uses the martingale residual of a Cox model, and AFT-MDR uses the standardized residual from an accelerated failure time model to classify multilocus genotypes into high-and low-risk groups. Both Cox-MDR and AFT-MDR have the major advantage of adjusting for the covariate effect because these methods are based on regression models such as the Cox model and the AFT model. The cross-validation procedure is the same as the traditional MDR and uses balanced accuracy. In this study, we propose a simple and computationally efficient method using the Kaplan-Meier median survival time, referred to as KM-MDR. This method conceptually extends the key idea of QMDR to the survival phenotype by replacing the mean value and the t test statistic with the median survival time and a log-rank test statistic, respectively. Since survival time is commonly censored and has a skewed distribution, the median survival time is more useful and popular statistic to make statistical inferences in survival analysis. KM-MDR uses the Kaplan-Meier median survival time to classify multilocus genotypes into the binary attribute of high-and low-risk groups. Once all multigenotypes are classified, we pool the high-risk genotype combinations into one group and the low-risk combinations into another group. Next, the log-rank test is applied to these two groups to choose the best model among all possible SNP combinations. Since the log-rank test is model-free, KM-MDR is nonparametric, as is Surv-MDR. However, KM-MDR is more efficient in the sense that the computing time for the median survival time is shorter than that for the log-rank test. Furthermore, Surv-MDR uses all data repeatedly by comparing the survival time between samples with and without a specific genotype combination, while KM-MDR uses all data once by comparing the median survival time of a specific genotype combination with the overall median survival time. In Section 2 we detail the algorithm of the proposed KM-MDR method and present simulation results to compare the performance of KM-MDR with that of Surv-MDR, Cox-MDR, and AFT-MDR in Section 3 We apply the KM-MDR to a real dataset of ovarian cancer patients from The Cancer Genome Atlas (TCGA) in Section 4 and a short discussion is presented in Section 5. Methods As mentioned in the previous section, we propose a simple and computationally efficient method, KM-MDR, as an extension of QMDR to handle survival phenotypes. Instead of using the mean value of the quantitative trait as done in QMDR, we compare the Kaplan-Meier median survival time of each multilocus genotype combination to the overall median survival time. In addition, a log-rank test statistic is used instead of the t test statistic as a criterion for finding the best model. To determine the q-loci that provides the best model overall, we use 10-fold cross-validation as done in QMDR as follows. (i) Suppose that we have p SNPs in the dataset and select q SNPs from p SNPs to consider a q-way interaction. For 10-fold cross-validations, the samples are randomly split into 10 subgroups of equal size. Then, 9/10 sets of samples are taken as the training dataset and the remaining sample is used as a testing dataset to evaluate the model. Step 2. Classify multilocus genotypes into high-and low-risk groups by the Kaplan-Meier median survival time. (i) For a q-way interaction, we construct multilocus genotype combinations defined by the q SNPs. (iii) We compare t mj to t m and classify each cell into either the high-risk group if t mj < t m, or the lowrisk group, otherwise. (iv) If the median survival time is not available due to heavy censoring or the sparsity of the sample for a specific genotype combination, we make use of a complementary sample corresponding to this specific cell to estimate the median survival time. Here, this complementary sample is a pooled sample excluding the sample with the specific genotype combination. Since the complementary sample is made up of 3 q − 1 cells, it may be large enough to have the median survival time. If the median survival time for the complementary sample is larger than the overall median survival time, then this pooled sample is considered as the low-risk group. Thus, the sample with the specific genotype has smaller median survival time, which leads to be classified as the high-risk group. Step 3. Calculate the log-rank test statistic to find the q-way best model. (i) Once all genotype combinations are classified into high-and low-risk groups, we pool the high-risk genotype combinations into one group, "H," and the low-risk combinations into another group, "L." For the training set, the log-rank test statistic is calculated to test the equivalence of the two survival curves for the H and L groups as follows: Let d li and n li be the observed number of events and the number of individuals at risk in the l th group l = 1, 2 at time t i and let d i = d 1i + d 2i and n i = n 1i + n 2i be the number of events and the number of at risk in the combined sample at t i. Then, the log-rank statistic is defined as (ii) We take the square of the log-rank test statistic as the score to characterize the relationship between survival time and gene-gene interaction. We use the training score from the log-rank test to select the best SNP pairs with the maximum training score among all possible SNP pairs. Next, we predict highand low-risk status in the testing set corresponding to the training set. (iii) We repeated the above procedure 10 times so that each partition is included in the testing dataset once. Then, we calculate a testing score by using the log-rank test for all possible 10 testing sets. In addition, the number of times is counted that each q-way model chosen from the training datasets is identified as the best model; this is called cross-validation consistency (CVC). Then we select the best q-way interaction model with the maximum CVC. Step 4. Find the overall best model. (i) Repeat Step 3 above for all possible q-way models of interest q = 1, 2,⋯,p. Then we select the best model that has the maximum testing score and the highest CVC. The latter is used as a tie-break. If both statistics are tied, then the more parsimonious model is selected as the overall best model as done in. As described in steps 1-4 above, any higher-order interaction model can be easily considered by the algorithm of MDR through cross-validation procedure. If possible, this cross-validation can be repeated to avoid the fluctuations due to chance divisions of the data. Furthermore, we determine the statistical significance of the selected model by comparing the testing score from observed data to the distribution of the testing scores under the null hypothesis of no association derived empirically from 1000 permutations. The null hypothesis is rejected if this empirical p-value is less than the significance level. Here, we display the algorithm of KM-MDR for 2-way interaction model in Figure 1. Simulation Study. In this section, we present the results of simulation studies comparing the performance of the proposed KM-MDR method with that of Surv-MDR, Cox-MDR, and AFT-MDR. As discussed in the previous section, both Surv-MDR and KM-MDR are nonparametric methods that cannot adjust for covariate effects, while both Cox-MDR and AFT-MDR are based on a regression model and can adjust for covariate effects. We consider two different scenarios for the power comparison: the first one is a model without a covariate effect, and the second is a model with a covariate effect. For the simulation study, we considered 10 SNPs that satisfied the assumption of Hardy-Weinberg equilibrium and linkage equilibrium. Among them, we let only two SNPs (SNP1 and SNP2) have causal SNP interactions with the survival time. We generated datasets based on different 3 BioMed Research International penetrance functions that define a probabilistic relationship between the outcome and SNPs in which the outcome was dependent on genotypes from two loci in the absence of any marginal effects. These models were distributed across seven different heritability values (0.01, 0.025, 0.05, 0.1, 0.2, 0.3, and 0.4) and two different minor allele frequencies (0.2 and 0.4). A total of five models for each of the 14 heritability minor allele frequency combinations were generated for a total of 70 epistatic models, which are given in detail in. Let f ij be an element from the i th row and the j th column of a penetrance function for the two causal SNPs (SNP1 and SNP2), defined as. We sampled 400 patients from each of the 70 penetrance models to generate one simulated dataset and repeated this process 100 times to generate 100 datasets for each model. Then we simulated the survival times using both a Cox model and an AFT model. Let x be an indicator variable with a value of 1 for highrisk patients and 0 for low-risk patients, and let z be an adjusting covariate generated from N. For a Cox model, given as t | x, z = 0 t exp x + z, we set = 1:2 and = 0 or 1. The baseline hazard function, 0 t, was assumed to follow a Weibull distribution with a shape parameter of 5 and a scale parameter of 2, and the censoring time was generated from a uniform distribution U as in. For an AFT model given as log T = + x + z +, we set = 0, = −1:0, and = 0 or 1: For the error distribution, we assumed that followed the normal distribution and that = 1:0. We also considered four different censoring fractions, (C): 0.0, 0.1, 0.3, and 0.5. In all, 70 penetrance model 2 survival model 2 covariate effect 4 censoring fraction = 1,120 different simulated datasets were repeated 100 times. The power was estimated as the percentage of times that each method correctly chose the causal SNP pairs (SNP1 and SNP2) as the best model among all possible two-way interaction models out of each set of 100 datasets. Parallel to the definition of this power, the Type I error of MDR has estimated the percentage of times that each method chose a certain SNP pair as the best model among all possible two-way interaction models under the null hypothesis. We generated 1000 null models with 8 non-causal SNPs for the Type I error. Then the selection rate of each SNP pair under the null model is Table 1, the Type 1 error was well controlled (<0.03577) and seemed to be rather conservative. For the power comparison, Figure 2 displays the power obtained from a Cox model and Figure 3 shows the power Step 1. Step 2. Step 3. First of all, a common property of all methods is that the power tended to increase as heritability increased, but decreased as the censoring fraction increased. In addition, the power seemed to be greater with MAF = 0:2 than with MAF = 0:4. It is also noted that both KM-MDR and Surv-MDR have almost the same power across all cases. SNPs The power of KM-MDR was greater than that of both Cox-MDR and AFT-MDR when there was no covariate effect, while KM-MDR had a somewhat lower power than either Cox-MDR or AFT-MDR when there was a covariate effect. However, the power of KM-MDR was comparable with that of Cox-MDR when the censoring fraction was greater than 0.3, although there was a covariate effect. The power of KM-MDR also seemed to be robust to the censoring fraction, similarly to Cox-MDR. However, the power of AFT-MDR was very sensitive to the censoring fraction because it was very low under a Cox model and had the lowest power even under an AFT model when the censoring fraction was greater than 0.3. In summary, KM-MDR had a power almost identical to that of Surv-MDR across all cases and outperformed Cox-MDR under a model without a covariate effect. In addition, KM-MDR had reasonable power even under heavy censoring and showed comparable power to that of Cox-MDR even when there was a covariate effect. Real Data Analysis. We illustrate the proposed method by analyzing ovarian cancer patient data from The Cancer Genome Atlas (TCGA) at https://gdc.xenahubs.net. This dataset consists of 433 ovarian cancer patients with 565 SNPs, in which 207 patients died from ovarian cancer, whereas 226 patients were censored. We first analyzed this data with all 565 SNPs using both KM-MDR and Surv-MDR and compared these two sets of results. In addition, 20 of the 565 SNPs showed a significant main effect in a single SNP analysis using a Cox regression model. In order to distinguish a true multiplicative model from an additive effect model, we also reanalyzed this dataset with 545 SNPs after removing the 20 SNPs that had significant main effects. We present the 20 SNPs with a significant main effect when we fit a Cox regression model with only each single SNP in Table 2. We applied 10-fold cross-validation by keeping the censoring fraction the same for alandomly split samples. For simplicity of comparison, we display the top three two-way interaction models identified by KM-MDR and Surv-MDR with all 565 SNPs and with 545 SNPs, respectively, in Table 3. Similarly, we display the top three twoway interaction models identified by Surv-MDR in Table 4. We implemented 10-fold cross-validation and selected the best pair of SNPs by comparing both the CVC and testing scores. As displayed in Figure 1, the CVC was used for selecting the best pair, and the testing score was used as a tiebreaker. However, the maximum value of CVC only identified one of the pairs in 10-fold cross-validation, and thus the testing score was used to select the top three pairs. In Tables 3 and 4, the training score is the average of 10 training scores across 10-fold cross-validation, whereas the testing score is a log-rank test statistic calculated from the whole dataset by combining all 10 disjoint testing sets. We also conducted 1000 permutations to obtain a p value to check the statistical significance of the selected model. Each table provides the selected model, training score, testing score, CVC, and permutation p value. Comparing the results shown in Tables 3 and 4, four SNP pairs overlapped between KM-MDR and Surv-MDR, and their testing scores were very similar. This result is consistent with the finding from the simulation studies that the power of both KM-MDR and Surv-MDR is almost the same. When all 565 SNPs were included in the analysis, only one pair (rs143372586 and rs61937629) included the rs143372586 SNP, which had a significant main effect, as shown in Table 2. This pair was identified as one of the top three pairs by both KM-MDR and Surv-MDR but did not appear in the analysis with 545 SNPs because the rs143372586 SNP was excluded. It can also be noted that all the permutation p values were significant for both KM-MDR and Surv-MDR. Furthermore, we plotted the survival curves for high-risk versus low-risk groups defined by the KM-MDR and Surv-MDR attribute for SNP pairs listed in Tables 3 and 4. For a given pair of SNPs, we implemented KM-MDR (or Surv-MDR) to classify high-and low-risk groups for the training set (9/10 sets) and applied this classification rule to the corresponding testing set (1/10 sets). By repeating this procedure 10 times, all patients could be assigned into high-or lowrisk groups by KM-MDR (or Surv-MDR), and the Kaplan-Meier survival curves for these two groups are plotted in Figure 4. Since the same procedure is implemented in calculating the testing score of the corresponding SNP pairs, the log-rank test statistic is the same as the testing score in Tables 3 and 4. As shown in Figures 4(a) and 4(b), all six of the survival curves obtained by both KM-MDR and Surv-MDR for the attribute of SNP pairs are substantially separated, with a significant p value of the log-rank test. Since the 10-fold crossvalidation process randomly divides the sample into 10 even groups, the results shown in Figure 4 could be a randomly chosen set of results. Nonetheless, the groups representing Discussion In this paper, we propose a simple and computationally efficient method, KM-MDR, to identify GGIs associated with the survival phenotype. The KM-MDR method can be considered as an extension of QMDR to the survival phenotype by replacing the mean value of the quantitative trait with the Kaplan-Meier median survival time to classify multilocus genotypes into high-and low-risk groups. Since the survival time is commonly censored and often has a skewed distribution, the median survival time is a more useful and robust statistic for the central measure than the mean survival time in survival analysis. Therefore, it is natural to consider the median survival time as a new classifier to reduce highdimensional genotypes into a binary attribute for the survival phenotype. In addition, the log-rank test statistic is most BioMed Research International popularly used to compare the survival times between two groups, analogously to the t test in QMDR. When comparing the process of KM-MDR with that of Surv-MDR, either the median survival time or a log-rank test statistic should be calculated nine times for two-way interaction model, because there are nine different genotypes available for two SNP pairs. As shown in the simulation results for the power comparison, the power of KM-MDR is almost the same as that of Surv-MDR for all cases. However, KM-MDR has the advantage of faster computation. Comparing the run time for one loop of the two-way interaction model, KM-MDR took 31.55 seconds and Surv-MDR took 37.49 seconds for the same procedure. In the analysis of ovarian cancer data, the KM-MDR method took 5.577 hours to search over all two-way models with 565 SNPs and 432 patients, while Surv-MDR took 6.003 hours. Thus, KM-MDR is computationally more efficient than Surv-MDR. Furthermore, the information of each cell is used once in KM-MDR, but it is used multiple times in Surv-MDR because the log-rank test compares the survival time between samples with and without the genotype, which may cause the distinction between the high-and low-risk groups to be contaminated. However, when only a few events are observed due to either a small size of the sample or heavy censoring, the classification for highand low-risk groups may be not available in KM-MDR. In fact, there were a few cases in which a median survival time was not reached in the process of real data analysis with 10fold cross-validation. Even if the sample is large enough, the median survival time may not be obtained when most patients are cured of a disease. In cases where overall median survival time is not available due to heavy censoring, KM-MDR cannot be applicable, whereas Surv-MDR has no such restriction. However, for the case that the median survival time of a specific cell is not obtained, we proposed an alternative method using complementary samples. As shown in the simulation studies, KM-MDR has greater power than Cox-MDR and AFT-MDR when there is no covariate effect and seems to be robust even under heavy censoring, whereas the power of AFT-MDR decreases very rapidly when the censoring fraction becomes greater than 0.3. In addition, KM-MDR has moderate power even when there is a covariate effect and has a similar power to that of Cox-MDR as the censoring fraction increases. Although KM-MDR has the weakness of not being able to In this paper, we did not perform the simulation study for the higher-order interaction model due to heavy computation time for the higher-order model. However, we implemented the simulation study by considering only a 3-way interaction model which referred in. We found that the power trend of all methods is similar to that shown in the 2-way interaction model though not given here. In the analysis of a dataset of ovarian cancer patients from TCGA, the two methods showed similar results to those obtained in the simulation study. Four pairs of SNPs overlapped, and the testing scores were also similar over all pairs. In addition, all the permutation p values were very low, which implies that all of the top three pairs were significant interaction models. Furthermore, we plotted the survival curves between the high-and low-risk groups identified by the KM-MDR and Surv-MDR attributes of SNP pairs using 10fold cross-validation. All survival curves showed a significant separation of the two groups in terms of the p value of the log-rank test. Conclusion We propose KM-MDR, a new extension of the MDR algorithm that enables an efficient identification of gene-gene interactions in a survival setting. Because it is simple and requires less computation, it is highly advantageous for dealing with high-dimensional biological data. We expect that KM-MDR will contribute to the identification of gene-gene interactions associated with numerous human diseases. Data Availability The real data used to support the findings of this study are available at https://gdc.xenahubs.net. |
The Health Minister Simon Harris is considering extending hours for some hospital services, to deal with the latest spike in hospital overcrowding.
He was responding to new figures, which showed a record number of 612 people on trolleys in emergency wards Tuesday morning.
Minister Harris says the surge is the result of a strain of influenza, which is affecting older people particularly harshly.
He says the best way to stop the problem escalating is to get vaccinated.
He has also outlined some of the options available to help reduce the numbers on trolleys.
"One is the idea of extending access to diagnostics in our hospitals until 8pm several nights a week from now until the Spring.
"A second is reviewing the length of stay for all of the patients who may be were admitted during the Christmas period, and are now fit to go home.
"A third is working with our nursing homes - particularly in relation to our frail elderly".
Trolley and ward watch figures from the Irish Nurses and Midwives Organisation (INMO) show University Hospital Limerick is the most overcrowded, with 46 people in need of a bed.
The Midland Regional Hospitals in Portlaoise and Tullamore, St Luke's Hospital in Kilkenny, and University Hospital Galway all have 40 or more patients on trolleys.
The INMO also says there were a record 93,621 admitted patients on trolleys over the course of 2016 - a record figure for a calendar year.
INMO General Secretary Liam Doran said: "612 patients, admitted for care, for whom there is no bed, is a truly shocking figure. The compromising of care, not to mention the loss of privacy and dignity, cannot go unchallenged and must be acknowledged and addressed by health management.
"We cannot allow this to become just another statistic and it must result in a fundamentally new approach to our health system as overcrowding, as the 2016 figures confirm, continues to grow."
He added that his organisation fears the situation may deteriorate further in the coming days, due to the current flu situation and closed beds due to staff shortages.
"An emergency response is now required and this must be forthcoming, from health management, immediately," Mr Doran said.
He added that "it is proof, if it were needed, that the health service is still too small to cater for demand".
Fianna Fáil's spokesperson on health, Billy Kelleher, suggested that "hospitals in Cork, Limerick, the Midlands and Donegal are creaking with the pressure being put on them".
In a statement responding to today's figures, he said: "It has now gotten to a stage where the HSE believes that 300-400 patients every day are lying on a trolley waiting for admission is acceptable. It’s not, and the Minister needs to bring forward proposals that increase capacity in our hospital wards and ensure those that need a ward bed get one as quickly as possible.
"Right across the country, there are community hospitals with fully functioning wards that could be used to treat non-acute patients, thereby freeing up acute beds in our major hospitals. It’s time for the Minister for think outside the box to find a permanent solution to hospital overcrowding."
Sinn Féin's Louise O'Reilly, meanwhile, argued: "It is clear that the government has no plan to deal with this escalating crisis and no strategy that will reassure overworked staff that this will not be another year of unacceptable overcrowding.
"This is utterly unacceptable and a bad start to the New Year for patients and health workers," she added. |
How and how not to develop IR theory This paper starts from the fact that there is a substantial gap in terms of IR theory production, between the West and the rest. Its aim is to investigate how that gap might be closed, and for this purpose, the paper takes a broad view of what counts as theory. Its method is comparative history: to observe how IR theory has developed not just in the West, which is well-studied, but also in the periphery, which is not. The idea is to identify what material conditions and motivations in both locations were associated with the emergence of theoretical thinking about international relations, and how and why theoretical differentiations emerged, particularly within the West. It also looks at conditions and circumstances that seem to work against the successful production of IR theory. The paper concludes with a brief consideration of IR theory development in China on the basis of the lessons drawn from the history of IR theory development. |
<reponame>bovender/unity-ip-indicator
import os
import logging
def enable():
log = logging.getLogger(__name__)
log.info("Writing autostart file")
path = __get_autostart_path()
if path is None: return
log.debug(path)
if not os.path.isfile(path):
print "Enabling autostart."
try:
dir = os.path.dirname(path)
if not os.path.exists(dir):
os.makedirs(dir)
with open(path, 'w') as f:
f.write("""[Desktop Entry]
Type=Application
Exec=indicator-ip
Hidden=false
NoDisplay=false
X-GNOME-Autostart-enabled=true
Name=IP Indicator
Name[en_US]=IP Indicator
Name[de_DE]=IP-Indicator
Comment=Displays IP addresses.
Comment[en_US]=Displays IP addresses.
Comment[de_DE]=Zeigt die aktuellen IP-Adressen.""")
except (OSError, IOError) as e:
print "Could not write file; error: {} ({})".format(e.strerror, e.errno)
log.warn("Failed to write file; error: {} ({})".format(e.strerror, e.errno))
else:
print "Autostart is already enabled."
print "If it does not work, please inspect the file {}".format(path)
log.info("Autostart is already enabled")
def disable():
log = logging.getLogger(__name__)
path = __get_autostart_path()
log.info("Disabling autostart")
log.debug(path)
if not os.path.isfile(path):
print "Autostart is not enabled at the moment."
log.info('Autostart .desktop file not found')
else:
print "Disabling autostart."
log.info("Removing autostart .desktop file")
try:
os.remove(path)
except OSError as e:
print "Could not remove file; error: {} ({})".format(e.strerror, e.errno)
log.warn("Failed to delete file; error: {} ({})".format(e.strerror, e.errno))
def __get_autostart_path():
desktop_file = 'indicator-ip.desktop'
if os.geteuid() == 0:
return os.path.join('/etc', 'xdg', 'autostart', desktop_file)
else:
# When updating via apt-get, $HOME may be unset, causing os.path.join
# to crash with a null reference exception.
try:
return os.path.join(
os.getenv('HOME'), '.config', 'autostart', desktop_file)
except Exception:
log.warning("Could not determine autostart path")
pass
|
<reponame>usgin/aasgtrack
from django.conf.urls.defaults import patterns, include, url
urlpatterns = patterns('aasgtrack.views',
url(r'^(?P<state>[a-zA-Z]{2})/(?P<context>[a-zA-Z]+)/?(?P<category>[a-zA-Z]+)?/?', 'state_progress'),
url(r'^map/?$', 'progress_map'),
url(r'^map/js/(?P<js_file_name>.+\.js)$', 'map_scripts'),
url(r'^admin/js/(?P<js_file_name>.+\.js)$', 'admin_scripts'),
url(r'^proxy.*$', 'service_proxy')
)
urlpatterns += patterns('aasgtrack.reports',
url(r'^report/all-data', 'all_data'),
url(r'^report/data/online', 'online_state_data'),
url(r'^report/data', 'state_data'),
url(r'^report/?$', 'full_report'),
url(r'^report/(?P<state>[a-zA-Z]{2})/?$', 'state_report')
) |
Comfort air-conditioning control for building energy-saving In buildings such as offices and department stores, the requirement is that the room air-conditioning control system has to provide a comfortable thermal feeling for the inhabitant. While energy saving may sometime conflict with comfort, it is possible to eliminate the wasteful use of energy associated its overuse without prejudice to comfort. The basic concept is that the target set values for the room temperature should be changed dynamically so as to suit the constantly changing room environment. This paper shows that the well-known comfort index PMV is useful to air-conditioning control systems aimed at both energy-saving and comfort. |
#include "veellipse.h"
#include <QPainter>
#include <QDebug>
#include <QCursor>
#include <QGraphicsScene>
#include <QGraphicsSceneMouseEvent>
#include <QGraphicsEllipseItem>
#include <math.h>
#include "dotsignal.h"
static const double Pi = 3.14159265358979323846264338327950288419717;
static double TwoPi = 2.0 * Pi;
static qreal normalizeAngle(qreal angle)
{
while (angle < 0)
angle += TwoPi;
while (angle > TwoPi)
angle -= TwoPi;
return angle;
}
VEEllipse::VEEllipse(QObject *parent) :
QObject(parent),
m_cornerFlags(0),
m_actionFlags(ResizeState)
{
setAcceptHoverEvents(true);
setFlags(ItemIsSelectable|ItemSendsGeometryChanges);
for(int i = 0; i < 8; i++){
cornerGrabber[i] = new DotSignal(this);
}
setPositionGrabbers();
}
VEEllipse::~VEEllipse()
{
for(int i = 0; i < 8; i++){
delete cornerGrabber[i];
}
}
QPointF VEEllipse::previousPosition() const
{
return m_previousPosition;
}
void VEEllipse::setPreviousPosition(const QPointF previousPosition)
{
if (m_previousPosition == previousPosition)
return;
m_previousPosition = previousPosition;
emit previousPositionChanged();
}
void VEEllipse::setElli(qreal x, qreal y, qreal w, qreal h)
{
setElli(QRectF(x,y,w,h));
}
void VEEllipse::setElli(const QRectF &rect)
{
QGraphicsEllipseItem::setRect(rect);
if(brush().gradient() != 0){
const QGradient * grad = brush().gradient();
if(grad->type() == QGradient::LinearGradient){
auto tmpRect = this->rect();
const QLinearGradient *lGradient = static_cast<const QLinearGradient *>(grad);
QLinearGradient g = *const_cast<QLinearGradient*>(lGradient);
g.setStart(tmpRect.left() + tmpRect.width()/2,tmpRect.top());
g.setFinalStop(tmpRect.left() + tmpRect.width()/2,tmpRect.bottom());
setBrush(g);
}
}
}
void VEEllipse::mouseMoveEvent(QGraphicsSceneMouseEvent *event)
{
QPointF pt = event->pos();
if(m_actionFlags == ResizeState){
switch (m_cornerFlags) {
case Top:
resizeTop(pt);
break;
case Bottom:
resizeBottom(pt);
break;
case Left:
resizeLeft(pt);
break;
case Right:
resizeRight(pt);
break;
case TopLeft:
resizeTop(pt);
resizeLeft(pt);
break;
case TopRight:
resizeTop(pt);
resizeRight(pt);
break;
case BottomLeft:
resizeBottom(pt);
resizeLeft(pt);
break;
case BottomRight:
resizeBottom(pt);
resizeRight(pt);
break;
default:
if (m_leftMouseButtonPressed) {
setCursor(Qt::ClosedHandCursor);
auto dx = event->scenePos().x() - m_previousPosition.x();
auto dy = event->scenePos().y() - m_previousPosition.y();
moveBy(dx,dy);
setPreviousPosition(event->scenePos());
emit signalMove(this, dx, dy);
}
break;
}
} else {
switch (m_cornerFlags) {
case TopLeft:
case TopRight:
case BottomLeft:
case BottomRight: {
rotateItem(pt);
break;
}
default:
if (m_leftMouseButtonPressed) {
setCursor(Qt::ClosedHandCursor);
auto dx = event->scenePos().x() - m_previousPosition.x();
auto dy = event->scenePos().y() - m_previousPosition.y();
moveBy(dx,dy);
setPreviousPosition(event->scenePos());
emit signalMove(this, dx, dy);
}
break;
}
}
QGraphicsItem::mouseMoveEvent(event);
}
void VEEllipse::mousePressEvent(QGraphicsSceneMouseEvent *event)
{
if (event->button() & Qt::LeftButton) {
m_leftMouseButtonPressed = true;
setPreviousPosition(event->scenePos());
emit clicked(this);
}
QGraphicsItem::mousePressEvent(event);
}
void VEEllipse::mouseReleaseEvent(QGraphicsSceneMouseEvent *event)
{
if (event->button() & Qt::LeftButton) {
m_leftMouseButtonPressed = false;
}
QGraphicsItem::mouseReleaseEvent(event);
}
void VEEllipse::mouseDoubleClickEvent(QGraphicsSceneMouseEvent *event)
{
m_actionFlags = (m_actionFlags == ResizeState)?RotationState:ResizeState;
setVisibilityGrabbers();
QGraphicsItem::mouseDoubleClickEvent(event);
}
void VEEllipse::hoverEnterEvent(QGraphicsSceneHoverEvent *event)
{
setPositionGrabbers();
setVisibilityGrabbers();
QGraphicsItem::hoverEnterEvent(event);
}
void VEEllipse::hoverLeaveEvent(QGraphicsSceneHoverEvent *event)
{
m_cornerFlags = 0;
hideGrabbers();
setCursor(Qt::CrossCursor);
QGraphicsItem::hoverLeaveEvent( event );
}
void VEEllipse::hoverMoveEvent(QGraphicsSceneHoverEvent *event)
{
QPointF pt = event->pos(); // 当前位置
qreal drx = pt.x() - rect().right(); // 与右侧距离
qreal dlx = pt.x() - rect().left(); // 于左侧距离
qreal dby = pt.y() - rect().top(); // 上
qreal dty = pt.y() - rect().bottom(); // 下
//如果在四个顶点半径为9的范围内，就设置flag标志
m_cornerFlags = 0;
if( dby < 9 && dby > -9 ) m_cornerFlags |= Top; // Top side
if( dty < 9 && dty > -9 ) m_cornerFlags |= Bottom; // Bottom side
if( drx < 9 && drx > -9 ) m_cornerFlags |= Right; // Right side
if( dlx < 9 && dlx > -9 ) m_cornerFlags |= Left; // Left side
if(m_actionFlags == ResizeState){
QPixmap p(":/icons/arrow-up-down.png");
QPixmap pResult;
QTransform trans = transform();
switch (m_cornerFlags) {
case Top:
case Bottom:
pResult = p.transformed(trans);
setCursor(pResult.scaled(24,24,Qt::KeepAspectRatio));
break;
case Left:
case Right:
trans.rotate(90);
pResult = p.transformed(trans);
setCursor(pResult.scaled(24,24,Qt::KeepAspectRatio));
break;
case TopRight:
case BottomLeft:
trans.rotate(45);
pResult = p.transformed(trans);
setCursor(pResult.scaled(24,24,Qt::KeepAspectRatio));
break;
case TopLeft:
case BottomRight:
trans.rotate(135);
pResult = p.transformed(trans);
setCursor(pResult.scaled(24,24,Qt::KeepAspectRatio));
break;
default:
setCursor(Qt::CrossCursor);
break;
}
} else {
switch (m_cornerFlags) {
case TopLeft:
case TopRight:
case BottomLeft:
case BottomRight: {
QPixmap p(":/icons/rotate-right.png");
setCursor(QCursor(p.scaled(24,24,Qt::KeepAspectRatio)));
break;
}
default:
setCursor(Qt::CrossCursor);
break;
}
}
QGraphicsItem::hoverMoveEvent( event );
}
QVariant VEEllipse::itemChange(QGraphicsItem::GraphicsItemChange change, const QVariant &value)
{
switch (change) {
case QGraphicsItem::ItemSelectedChange:
m_actionFlags = ResizeState;
break;
default:
break;
}
return QGraphicsItem::itemChange(change, value);
}
void VEEllipse::resizeLeft(const QPointF &pt)
{
QRectF tmpRect = rect();
// 如果在右侧返回
if( pt.x() > tmpRect.right() )
return;
qreal widthOffset = ( pt.x() - tmpRect.right() );
// 限制最小宽度
if( widthOffset > -10 )
return;
// 设置宽度
if( widthOffset < 0 )
tmpRect.setWidth( -widthOffset );
else
tmpRect.setWidth( widthOffset );
//拖动左侧，保持右侧不变
tmpRect.translate( rect().width() - tmpRect.width() , 0 );
prepareGeometryChange();
//设置边框
setRect( tmpRect );
update();
setPositionGrabbers();
}
void VEEllipse::resizeRight(const QPointF &pt)
{
QRectF tmpRect = rect();
if( pt.x() < tmpRect.left() )
return;
qreal widthOffset = ( pt.x() - tmpRect.left() );
if( widthOffset < 10 )
return;
if( widthOffset < 10)
tmpRect.setWidth( -widthOffset );
else
tmpRect.setWidth( widthOffset );
prepareGeometryChange();
setRect( tmpRect );
update();
setPositionGrabbers();
}
void VEEllipse::resizeBottom(const QPointF &pt)
{
QRectF tmpRect = rect();
if( pt.y() < tmpRect.top() )
return;
qreal heightOffset = ( pt.y() - tmpRect.top() );
if( heightOffset < 11 )
return;
if( heightOffset < 0)
tmpRect.setHeight( -heightOffset );
else
tmpRect.setHeight( heightOffset );
prepareGeometryChange();
setRect( tmpRect );
update();
setPositionGrabbers();
}
void VEEllipse::resizeTop(const QPointF &pt)
{
QRectF tmpRect = rect();
if( pt.y() > tmpRect.bottom() )
return;
qreal heightOffset = ( pt.y() - tmpRect.bottom() );
if( heightOffset > -11 )
return;
if( heightOffset < 0)
tmpRect.setHeight( -heightOffset );
else
tmpRect.setHeight( heightOffset );
tmpRect.translate( 0 , rect().height() - tmpRect.height() );
prepareGeometryChange();
setRect( tmpRect );
update();
setPositionGrabbers();
}
void VEEllipse::rotateItem(const QPointF &pt)
{
QRectF tmpRect = rect();
QPointF center = boundingRect().center();
QPointF corner;
switch (m_cornerFlags) {
case TopLeft:
corner = tmpRect.topLeft();
break;
case TopRight:
corner = tmpRect.topRight();
break;
case BottomLeft:
corner = tmpRect.bottomLeft();
break;
case BottomRight:
corner = tmpRect.bottomRight();
break;
default:
break;
}
QLineF lineToTarget(center,corner);
QLineF lineToCursor(center, pt);
//获取旋转角度
qreal angleToTarget = ::acos(lineToTarget.dx() / lineToTarget.length());
qreal angleToCursor = ::acos(lineToCursor.dx() / lineToCursor.length());
if (lineToTarget.dy() < 0)
angleToTarget = TwoPi - angleToTarget;
angleToTarget = normalizeAngle((Pi - angleToTarget) + Pi / 2);
if (lineToCursor.dy() < 0)
angleToCursor = TwoPi - angleToCursor;
angleToCursor = normalizeAngle((Pi - angleToCursor) + Pi / 2);
//计算角度
auto resultAngle = angleToTarget - angleToCursor;
QTransform trans = transform();
trans.translate( center.x(), center.y());
trans.rotateRadians(rotation() + resultAngle, Qt::ZAxis);
trans.translate( -center.x(), -center.y());
setTransform(trans);
}
void VEEllipse::setPositionGrabbers()
{
QRectF tmpRect = rect();
cornerGrabber[GrabberTop]->setPos(tmpRect.left() + tmpRect.width()/2, tmpRect.top());
cornerGrabber[GrabberBottom]->setPos(tmpRect.left() + tmpRect.width()/2, tmpRect.bottom());
cornerGrabber[GrabberLeft]->setPos(tmpRect.left(), tmpRect.top() + tmpRect.height()/2);
cornerGrabber[GrabberRight]->setPos(tmpRect.right(), tmpRect.top() + tmpRect.height()/2);
cornerGrabber[GrabberTopLeft]->setPos(tmpRect.topLeft().x(), tmpRect.topLeft().y());
cornerGrabber[GrabberTopRight]->setPos(tmpRect.topRight().x(), tmpRect.topRight().y());
cornerGrabber[GrabberBottomLeft]->setPos(tmpRect.bottomLeft().x(), tmpRect.bottomLeft().y());
cornerGrabber[GrabberBottomRight]->setPos(tmpRect.bottomRight().x(), tmpRect.bottomRight().y());
}
void VEEllipse::setVisibilityGrabbers()
{
cornerGrabber[GrabberTopLeft]->setVisible(true);
cornerGrabber[GrabberTopRight]->setVisible(true);
cornerGrabber[GrabberBottomLeft]->setVisible(true);
cornerGrabber[GrabberBottomRight]->setVisible(true);
if(m_actionFlags == ResizeState){
cornerGrabber[GrabberTop]->setVisible(true);
cornerGrabber[GrabberBottom]->setVisible(true);
cornerGrabber[GrabberLeft]->setVisible(true);
cornerGrabber[GrabberRight]->setVisible(true);
} else {
cornerGrabber[GrabberTop]->setVisible(false);
cornerGrabber[GrabberBottom]->setVisible(false);
cornerGrabber[GrabberLeft]->setVisible(false);
cornerGrabber[GrabberRight]->setVisible(false);
}
}
void VEEllipse::hideGrabbers()
{
for(int i = 0; i < 8; i++){
cornerGrabber[i]->setVisible(false);
}
}
|
Remarks by B. Ko-Yung Tung: I come here with two hats on, one of Edward Martin, who was unable to be here, and the other is my own, which is that of one straight out of law school and practicing only for a few months. Mr. Martin asked me to impress upon you the fact that it is very difficult to practice transnational law without fully understanding the culture and the laws of the foreign nation with which you are dealing. It is difficult to sit in New York and keep current with the law and thinking and certain forecasting that you have to do in the practice of law. |
Hibs head coach Neil Lennon did not take training this morning and failed to appear for a scheduled press conference this afternoon.
His assistant Garry Parker told journalists that Lennon was fine, but had been struggling with the flu.
Lennon, who said after Wednesday’s 2-1 defeat to Hearts that he would consider his future at Easter Road over the summer, is understood to have held constructive talks with Chief Executive Leeann Dempster and Head of Football Operations George Craig.
However, the players haven’t seen Lennon since Wednesday night, with Marvin Bartley admitting that they were as surprised as anyone to hear the Northern Irishman’s comments after the match at Tynecastle.
Lennon branded some of his players “unprofessional” following the derby defeat, which cost the Easter Road side the chance of finishing second in the Scottish Premiership.
Speaking after the match, Lennon said: “I have been here two years and made great strides but I have to think about myself and how I feel about the whole thing.
“Maybe I am a bit over emotional, but obviously fourth is not good enough. I will reconsider my position now over the summer.
Speaking at today’s media conference, midfielder Bartley revealed the players haven’t had a meeting with Lennon.
Bartley said: “We haven’t had a meeting, I haven’t seen the gaffer since we left the hotel [on Wednesday]. |
/**
* Add a new tab. Should only do this if have checked showIfOpen to avoid dupes being opened.
*
* @param tabname The displayed tab name.
* @param widget The contents.
* @param keys An array of keys which are unique.
*/
public void addTab(final String tabname,
IsWidget widget,
final String[] keys) {
final String panelId = (keys.length == 1 ? keys[0] + id++ : Arrays.toString( keys ) + id++);
ScrollPanel localTP = new ScrollPanel();
localTP.add( widget );
tabLayoutPanel.add( localTP,
newClosableLabel( localTP,
tabname ) );
tabLayoutPanel.selectTab( localTP );
/* localTP.ad( new PanelListenerAdapter() {
public void onDestroy(Component component) {
Panel p = openedTabs.remove( keys );
if ( p != null ) {
p.destroy();
}
openedAssetEditors.remove( panelId );
openedPackageEditors.remove( tabname );
}
} );
*/
if ( widget instanceof GuvnorEditor ) {
this.openedAssetEditors.put( panelId,
(GuvnorEditor) widget );
} else if ( widget instanceof PackageEditorWrapper ) {
this.getOpenedPackageEditors().put( tabname,
(PackageEditorWrapper) widget );
}
openedTabs.put( keys,
localTP );
itemWidgets.put( localTP,
keys );
} |
from flask_wtf import FlaskForm
from wtforms import StringField, TextAreaField, SelectField
from wtforms.validators import DataRequired
from models import CategoryEnum
class QuestionForm(FlaskForm):
category = SelectField('Категория', choices=CategoryEnum.choices(), coerce=CategoryEnum.coerce)
question = StringField('Вопрос', validators=[DataRequired()])
detailed_description = TextAreaField('Детальное описание')
|
Six weeks into the Massachusetts football team’s season, it appeared that UMass coach Mark Whipple’s time in Amherst was over. The team was 0-6 and, given the strength of its remaining schedule, looked destined for another two-win season at best.
But the Minutemen turned it around, winning four of their last six to finish the season 4-8, the most wins they’ve had since becoming an FBS team.
Following the Minutemen’s 63-45 defensive collapse against Florida International University, Athletic Director Ryan Bamford gave a hint toward Whipple’s future via Twitter.
He later confirmed to Matt Vautour of the Daily Hampshire Gazette that Whipple will be getting an extension on his contract.
The UMass hockey team is 14 games into its regular season, less than halfway through, and have already surpassed their win total from last season and tied that of the 2015-16 season.
The Minutemen’s most recent win, a 4-2 victory over University of Connecticut at the Mullins Center, not only featured UMass’ largest attendance of the season at 3,776, but also helped them earn six votes in the most recent USCHO.com rankings.
Though not yet ranked, the six votes are an increase from last week when the Minutemen received one for the first time all season.
UConn will play host to the Minutemen tomorrow in Storrs, Connecticut. The Huskies are well below .500 at 6-11-2, and 4-7-1 in the Hockey East. UMass is currently 3-3 in the Hockey East.
It started with Minnesota, then continued with Brigham Young, followed by Quinnipiac and now South Carolina. The once 3-1 Minutemen are now 3-5 following their loss to the Gamecocks on Saturday.
Despite leading at halftime, UMass couldn’t overcome South Carolina’s second-half push. However, the most notable discrepancy on the stat sheet was in the free throw column.
Seventeen of the Gamecocks’ points came from the charity stripe out of twenty-nine attempts, but the same could not be said for the Minutemen.
UMass went to the line only five times. Getting more fouls was a point of emphasis by UMass coach Matt McCall in the postgame press conference.
“We have to find ways to get fouled and be more physical,” McCall said to the Daily Hampshire Gazette following Saturday’s game.
Though the Minutemen weren’t on the receiving end of many fouls, they did hand a lot out. Five Minutemen finished with four or more fouls by the game’s end.
UMass will have its next chance to snap its losing streak Wednesday night vs. Holy Cross at the Mullins Center.
Three weeks ago, the UMass women’s basketball team traveled north, and this past weekend they traveled south.
With games lined up against Incarnate Word, Texas Rio Grande Valley and Mississippi Valley State, the Minutewomen took two of the three, their only loss coming to UTRGV.
The team’s most dominant win came in its final game against Mississippi Valley State on Saturday, where UMass dominated 85-46. Hailey Leidel led with 25 points while Maggie Mulligan recorded a double-double. The Minutewomen ended up shooting over 40 percent overall and from beyond the 3-point line.
After defeating Incarnate Word handily, UMass fell to UTRGV on Friday 67-59, despite outscoring the Vaqueros 21-15 in the fourth quarter. |
Radiative transfer in a Solar CPC Photoreactor using the First-Order Scattering Method Abstract This manuscript analyzes the suitability of a recently proposed numerical method, the First-Order Scattering Method (FOS), to describe radiation transfer in a Solar Compound Parabolic Collector Photoreactor (CPCP). The study considers five different irradiance conditions ranging from fully diffuse to fully direct solar radiation, with 90 and 45° angled rays. Three photocatalysts at different loadings were considered: Evonik P25, Graphene Oxide, and Goethite, selected due to their relevance in photocatalytic applications and the availability of optical transport properties in the open literature. The study shows that the method is efficient and free of statistical noise, while its accuracy is not affected by the boundary conditions complexity. The methods accuracy is very high for photocatalysts with low to moderate albedos, such as Goethite and Graphene Oxide, displaying Normalized Absoluted Mean Error below 3%, i.e., comparable to the Monte Carlo (MC) Methods statistical fluctuations. |
#include <iostream>
#define maxLength 100
using namespace std;
class queue {
private:
int front = 0;
int rear = 0;
int queue[maxLength];
public:
int isEmpty(){
if (front == rear){
return 1;
}
return 0;
}
void enqueue(int ele){
if (maxLength == rear){
cout << "Queue Overflow" << endl;
}
else{
queue[rear] = ele;
rear++;
}
}
void dequeue(){
if (isEmpty() == 1){
cout << "Queue Underflow" << endl;
}
else{
for (int i = 0; i < (rear - 1); i++){
queue[i] = queue[i + 1];
}
rear--;
}
}
void display(){
if (isEmpty() == 1){
cout << "Queue Empty." << endl;
}
else {
for (int i = 0; i < rear; i++){
cout << queue[i] << " <- ";
}
cout << endl;
}
}
};
int main(){
cout << "Name: <NAME>\nReg No: 20BEE1004" << endl;
queue q;
q.dequeue();
q.enqueue(4);
q.enqueue(8);
q.enqueue(12);
q.display();
q.dequeue();
q.display();
q.enqueue(16);
q.display();
q.dequeue();
q.display();
return 0;
} |
<reponame>skerkewitz/appreview
package de.skerkewitz.appreview.service;
import org.w3c.dom.Node;
import org.w3c.dom.NodeList;
import javax.xml.xpath.XPathConstants;
import javax.xml.xpath.XPathExpression;
import javax.xml.xpath.XPathExpressionException;
import javax.xml.xpath.XPathFactory;
import java.text.ParseException;
import java.text.SimpleDateFormat;
/**
* Class to parse a single review entry.
*
* @author skerkewitz, 2016-03-16
*/
public class EntryParser {
private final XPathFactory xPathfactory = XPathFactory.newInstance();
private final XPathExpression exprId;
private final XPathExpression exprTitle;
private final XPathExpression exprAuthor;
private final XPathExpression exprRating;
private final XPathExpression exprText;
private final XPathExpression exprDate;
private final XPathExpression exprVersion;
private final XPathExpression exprEntries;
public EntryParser() throws XPathExpressionException {
exprEntries = xPathfactory.newXPath().compile("//entry");
exprId = xPathfactory.newXPath().compile("./id");
exprDate = xPathfactory.newXPath().compile("./updated");
exprVersion = xPathfactory.newXPath().compile("./version");
exprTitle = xPathfactory.newXPath().compile("./title");
exprAuthor = xPathfactory.newXPath().compile("./author/name");
exprRating = xPathfactory.newXPath().compile("./rating");
exprText = xPathfactory.newXPath().compile("./content[@type='text']");
}
public ReviewEntry parseEntry(Node node, AppStoreCC appStoreCC) throws XPathExpressionException, ParseException {
final Double id = (Double) exprId.evaluate(node, XPathConstants.NUMBER);
final String date = (String) exprDate.evaluate(node, XPathConstants.STRING);
final String version = (String) exprVersion.evaluate(node, XPathConstants.STRING);
final String title = (String) exprTitle.evaluate(node, XPathConstants.STRING);
final String name = (String) exprAuthor.evaluate(node, XPathConstants.STRING);
final Double rating = (Double) exprRating.evaluate(node, XPathConstants.NUMBER);
final String text = (String) exprText.evaluate(node, XPathConstants.STRING);
final SimpleDateFormat sdf = new SimpleDateFormat("yyyy-MM-dd'T'HH:mm:ssXXX");
return new ReviewEntry(id.intValue(), appStoreCC, sdf.parse(date), name, rating.intValue(), version, title, text);
}
public NodeList parseEntries(Object doc) throws XPathExpressionException {
return (NodeList) exprEntries.evaluate(doc, XPathConstants.NODESET);
}
}
|
/*
* generate 37 different valid_data_mask
* 8 from 0xff to 0xff00000000000000
* 7 from 0xffff to 0xffff000000000000
* ...
* 0xffffffffffffffff and 0
*/
static int initLegalValidMasks(u64a validMasks[]) {
u64a data = ONES64;
int num = 0;
for (int i = 0; i < 64; i += 8) {
for (int j = 0; j <= i; j += 8) {
validMasks[num] = data << j;
num++;
}
data >>= 8;
}
validMasks[num] = 0;
num++;
return num;
} |
# -*- coding: utf-8 -*-
"""Tests for compression/decompression."""
import graphtransliterator
import graphtransliterator.compression as compression
from graphtransliterator import GraphTransliterator
import json
import pytest
test_config = """
tokens:
a: [class_a]
b: [class_b]
c: [class_c]
" ": [wb]
d: []
Aa: [contrained_rule]
rules:
a: A
b: B
<class_c> <class_c> a: A(AFTER_CLASS_C_AND_CLASS_C)
(<class_c> b) a: A(AFTER_B_AND_CLASS_C)
(<class_c> b b) a a: AA(AFTER_BB_AND_CLASS_C)
a <class_c>: A(BEFORE_CLASS_C)
a b (c <class_b>): AB(BEFORE_C_AND_CLASS_B)
c: C
c c: C*2
a (b b b): A(BEFORE_B_B_B)
d (c <class_a>): D(BEFORE_C_AND_CLASS_A)
(b b) a: A(AFTER_B_B)
<wb> Aa: A(ONLY_A_CONSTRAINED_RULE)
d d: "<DD>"
d: "<D>"
" ": " "
onmatch_rules:
-
<class_a> <class_b> + <class_a> <class_b>: "!"
-
<class_a> + <class_b>: ","
whitespace:
default: ' '
consolidate: True
token_class: wb
"""
def test_compression():
gt = GraphTransliterator.from_yaml(test_config)
compressed_config = compression.compress_config(gt.dump())
decompressed_config = compression.decompress_config(compressed_config)
gt_from_decompressed = GraphTransliterator.load(decompressed_config)
# Compare JSON dumps with sorted keys.
assert (
json.dumps(gt.dump(), sort_keys=True) ==
json.dumps(gt_from_decompressed.dump(), sort_keys=True)
)
# Test bad compression level
with pytest.raises(ValueError):
gt.dump(compression_level=graphtransliterator.HIGHEST_COMPRESSION_LEVEL+1)
# Test compression at level 0 (should likely not be called)
assert "compressed_settings" not in compression.compress_config(
gt.dump(), compression_level=0
)
# Test compression levels
assert '"tokens": ' in gt.dumps(compression_level=0)
assert '"compressed_settings"' in gt.dumps(compression_level=1)
assert '"compressed_settings"' in gt.dumps(compression_level=2)
for i in range(0, graphtransliterator.HIGHEST_COMPRESSION_LEVEL+1):
x = gt.dumps(compression_level=i)
y = gt.loads(x)
assert y.transliterate("a") == "A"
|
/**
* Registers the instance of {@link BatchStepCollections} as bean.
*
* @return The instance of {@link BatchStepCollections}
*/
@Bean
public BatchStepCollections batchStepCollections() {
final BatchStepCollections.BatchStepCollectionsBuilder batchStepCollectionsBuilder = BatchStepCollections
.builder();
batchStepCollectionsBuilder.executeDeleteFileStep(this.executeDeleteFileStep());
batchStepCollectionsBuilder.notifyResultReportStep(this.notifyResultReportStep());
batchStepCollectionsBuilder.closeSessionStep(this.closeSessionStep());
return batchStepCollectionsBuilder.build();
} |
TORONTO – One of the largest gatherings of Western Muslims, the Annual Reviving the Islamic Spirit Convention (RIS) set to begin on Friday in Toronto, will be marking its 15th Anniversary this year.
“The convention showcases Islamic leadership from across the globe sharing a common platform before the widest cross-section of our community,” stated RIS in a statement obtained by AboutIslam.net.
“This program hopes to empower the youth across North America and inspire a true revival.”
Thousands of Muslims from across North America and beyond are expected to attend the convention.
They will come to listen to leading Muslim scholars, academics, and motivational speakers, meet and mingle with fellow Muslims, and discuss matters of faith and contemporary issues.
Themed ‘Promise of God: Conditions of Revival’, the convention is set to run from Friday, December 23, through Sunday, December 25, at the Metro Toronto Convention Center in the heart of downtown Toronto.
“It will be a timely discussion on many challenges faced by us collectively and, inshaAllah, find answers to,” said RIS in a release.
“It will be a hopeful and positive response, one that will uplift our soul, inspire our heart and revive our spirit.”
The Reviving the Islamic Spirit convention was first launched fifteen years ago by Muslim youth to tackle the backlash on Islam and Muslims after the 9/11 and to build a bridge of understanding with non-Muslims.
The event has become a staple in Toronto’s downtown core during the Christmas holiday season and an estimated 20,000 attendees are expected from all over North America and around the world, adding a boost to the local economy.
Busy Schedule
The RIS annual convention is regarded a premiere global event among Muslims in the Western world and an annual meeting place for leading thinkers and theologians.
This year’s speakers include an impressive list of world-renowned scholars and speakers such as Shaykh Suhaib Webb, Dr. Seyyed Hossein Nasr, Prof. Sibghatullah Mojaddedi, Dr. Umar F. Abd Allah, Shaykh Hamza Yusuf, Shaykh Mokhtar Maghraoui, Imam Zaid Shakir, Ustadh Nouman Ali Khan, Dr. Jamal Badawi, and Shaykh Sulaiman Mulla.
Dr. Recep Senturk, Imam Khalid Latif, Imam Siraj Wahhaj, Dr. Abdal Hakim Jackson, Muslema Purmul, Yasmin Mogahed, Shaykh Abdul Nasir Jangda, Imam Yasir Fahmy, Pastor Bob Roberts, Rabbi Michael Lerner, Sr. Salma Yaqoob, Mehdi Hasan, Shaykh Muhammad Ninowy, Linda Sarsour, Khizr and Ghazala Khan, Ibtihaj Muhammad, Lonnie Ali, Rasheda Ali, Maryum Ali, Hamza Abdullah and Husain Abdullah, will also be among attendants.
A special tribute to the late boxing legend Muhammad Ali will be held during the program with his wife, Lonnie Ali, and daughters, Rashed and Maryum Ali, who will be special guests.
A number of international entertainers and performance artists will be appearing at the convention including Isam Bachiri and Dawud Wharnsby Ali
A bustling grand bazaar will be a major feature during the three-day event showcasing products, as well as charities and services from around North America and the Muslim world.
The convention will also host an Appreciation Dinner on Friday, December 23rd to honor the visiting scholars and speakers. Dignitaries, community leaders, and political and civic leaders are expected to attend the elegant dinner event.
A parallel children’s program runs every day of the event and a food and coat drive to assist the needy in the Greater Toronto Area will be held.
“Reviving the Islamic Spirit Convention is an attempt by the youth to help overcome new challenges of communication and integration,” noted RIS.
“The convention aims to promote stronger ties within the North American Society through reviving the Islamic tradition of education, tolerance and introspection, and across cultural lines through points of commonality and respect.” |
/*
* The MIT License (MIT)
*
* Copyright (c) 2015 CodeRevisited.com
*
* Permission is hereby granted, free of charge, to any person obtaining a copy
* of this software and associated documentation files (the "Software"), to deal
* in the Software without restriction, including without limitation the rights
* to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
* copies of the Software, and to permit persons to whom the Software is
* furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included in all
* copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
* AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
* OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*
*/
package com.coderevisited.arrays.rotation;
import com.coderevisited.math.utils.MathUtils;
/**
* http://www.cs.bell-labs.com/cm/cs/pearls/s02b.pdf
*/
public class ArrayRotationJuggling
{
public static void main(String[] args)
{
int[] array = new int[]{1, 2, 3, 4, 5, 6, 7, 8, 9, 10};
rotateArray(array, 4);
System.out.println("contents after rotation : ");
for (int i : array) {
System.out.print(i + " ");
}
}
private static void rotateArray(int[] array, int d)
{
int gcd = MathUtils.gcd(array.length, d);
for (int i = 0; i < gcd; i++) {
int pivot = array[i];
int j = i;
int k;
while (true) {
k = j + d;
if (k >= array.length)
k = k - array.length;
if (k == i)
break;
array[j] = array[k];
j = k;
}
array[j] = pivot;
}
}
}
|
In October, the federal government conducted the largest ever release of federal prisoners, letting 6,000 drug offenders out into the world. As bipartisan support pushes criminal justice reform forward at the state and federal levels, Americans should expect these types of releases to continue. But many people, such as Fox News host Bill O'Reilly, are worried this will lead to more crime and chaos.
But there's a good reason not to worry: There's little indication most federal drug offenders are violent or dangerous, even if they were involved in drug trafficking.
A new report from the Urban Institute makes that point: It found that a majority of drug offenders in federal prison have no serious, violent criminal history. A little more than one in five have a minor history, such as simple assault and other crimes that don't typically lead to serious injury. Fewer than one in four drug offenders in federal prison has a serious history.
The Urban Institute also found that very few of those convicted for drug offenses even had leading or violent roles in drug trafficking organizations. Only 14 percent were sentenced for being a manager, supervisor, leader, or organizer in an offense. Fewer than 14 percent were sentenced for using violence, making a credible threat to use violence, or directing the use of violence during the offense. And more than 75 percent didn't have or weren't in the presence of a weapon during the offense.
The result: There are a lot of people serving long prison sentences — on average, more than nine years, according to Urban Institute — for what amounts to small-time drug dealing. The general consensus among criminal justice reformers is that there's little reason to think this population is a serious risk to society and should be in prison at all or for very long, so maybe releasing them a little early isn't a bad idea.
Correction: The article originally suggested that most federal drug offenders have no criminal history, but they actually have no serious criminal history. |
Newt Gingrich for vice president?
I don’t think we’ve commented yet on the report by Eliana Johnson that Newt Gingrich is under serious consideration as Donald Trump’s running mate. Elaina’s report is consistent with what I’ve been hearing.
Gingrich would, I think, be an odd choice. But then, this is an odd year.
If Trump needs the support of any one group to win the presidency, that group is white women. The billionaire can expect virtually no support from African-Americans. From Hispanics, he’ll be lucky to get to 30 percent. He therefore will have to carry the white vote overwhelmingly, but cannot if he doesn’t do reasonably well with white women.
Is there a potential Trump running mate who white women like less than Newt Gingrich? I can’t think of one.
Trump also needs to win over traditional conservatives. Would Gingrich help with this cohort? Perhaps. He was once Mr. Conservative. Though too flaky to maintain this title for long, many conservatives are likely to find Gingrich palatable, especially compared to some of the other alleged VP prospects.
Thus, selecting Gingrich might enable Trump to convince conservatives he’s likely to govern mostly from the right. I’m no Gingrich fan, but if Trump picks him I will be a bit less worried about the tycoon from a policy perspective.
Picking Gingrich would be a good sign in another sense. The former Speaker reportedly is bombarding Trump with policy proposals and other forms of advice. If Trump taps Gingrich for the vice presidency, it would be an indication that he is willing to learn and unafraid to work with an egotistical policy wonk. Say what you want about Gingrich, he is not a yes man.
What Trump needs, though, is someone who can help him appeal to white women. His running mate need not be female, but he shouldn’t be someone who puts women off.
In my opinion, he shouldn’t be Newt Gingrich. |
package phoenixTeam.component.render;
import com.badlogic.ashley.core.Component;
import com.badlogic.gdx.graphics.Texture;
import com.badlogic.gdx.graphics.g2d.TextureRegion;
import com.badlogic.gdx.utils.Array;
import phoenixTeam.util.specific.LoadableComponent;
public class RenderComponent extends Component{
public TextureRegion region;
public float xSize;
public float ySize;
public Array<LoadableComponent<Texture>> toLoad = new Array<LoadableComponent<Texture>>();
}
|
def do_sentences(self, args):
sentence_token = lambda t: t[-1] in ".!?"
try:
print(self.markov.generate(args["<len>"], args["--seed"],
args["--prob"], args["--offset"],
startf=sentence_token,
endchunkf=sentence_token,
prefix=args["<prefix>"]))
except markovstate.MarkovStateError as e:
print(e.value) |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.