app.py

import datasets
import streamlit as st
import numpy as np
import pandas as pd
import altair as alt

st.markdown("""
# CryptoCEN Top50 co-expressed partners

**CryptoCEN** is a co-expression network for *Cryptococcus neoformans* built on 1,524 RNA-seq runs across 34 studies.
A pair of genes are said to be co-expressed when their expression is correlated across different conditions and
is often a marker for genes to be involved in similar processes. 

To Cite:
MJ O'Meara, JR Rapala, CB Nichols, C Alexandre, B Billmyre, JL Steenwyk, A Alspaugh,
TR O'Meara CryptoCEN: A Co-Expression Network for Cryptococcus neoformans reveals
novel proteins involved in DNA damage repair
* Code available at https://github.com/maomlab/CalCEN/tree/master/vignettes/CryptoCEN
* Full network and dataset: https://huggingface.co/datasets/maomlab/CryptoCEN

## Look up top-coexpressed partners:
Put in the ``CNAG_#######`` gene_id for a gene and expand the table to get the top 50 co-expressed genes.
``coexp_score`` ranges between ``[0-1]``, where ``1`` is the best and greater than ``0.85`` can be considered significant.
""")

top_coexp_hits = datasets.load_dataset(
    path = "maomlab/CryptoCEN",
    data_files = {"top_coexp_hits": "top_coexp_hits.tsv"})
top_coexp_hits = top_coexp_hits["top_coexp_hits"].to_pandas()

gene_id = st.text_input(
    label = "Gene 1",
    value = "CNAG_04365",
    max_chars = 10,
    help = "CNAG Gene ID e.g. CNAG_04365")

top_coexp_hits = top_coexp_hits[
    top_coexp_hits.gene_id_1 == gene_id,]
top_coexp_hits = top_coexp_hits[[
    'gene_id_1', 'gene_symbol_1', 'description_1',
    'gene_id_2', 'gene_symbol_2', 'description_2',
    'coexp_score', 'blastp_EValue']]

st.table(top_coexp_hits)