esm_variants / app.py
vasilisNt's picture
Update app.py
f575805
raw
history blame
1.93 kB
import streamlit as st
st.set_page_config(layout="wide")
import pandas as pd
import numpy as np
from zipfile import ZipFile
import plotly.express as px
import plotly.graph_objs as go
LLR_FILE='sample gene scores.zip'
with ZipFile(LLR_FILE) as myzip:
uids=[i.split('/')[-1].split('_')[0] for i in myzip.namelist()][1:]
df=pd.read_csv('isoform_list.csv',index_col=0)
def load_LLR(uniprot_id):
'''Loads the LLRs for a given uniprot id. Returns a 20xL dataframe
rows are indexed by AA change,
(AAorder=['K','R','H','E','D','N','Q','T','S','C','G','A','V','L','I','M','P','Y','F','W'])
columns indexed by WT_AA+position e.g, "G 12"
Usage example: load_LLR('P01116') or load_LLR('P01116-2')'''
with ZipFile(LLR_FILE) as myzip:
data = myzip.open(myzip.namelist()[0]+uniprot_id+'_LLR.csv')
return pd.read_csv(data,index_col=0)
def plot_interactive(uniprot_id):
primaryLLR = load_LLR(uniprot_id)
template='plotly_white'
fig = px.imshow(primaryLLR.values, x=primaryLLR.columns, y=primaryLLR.index, color_continuous_scale='Viridis_r',zmax=0,zmin=-20,
labels=dict(y="Amino acid change", x="Protein sequence", color="LLR"),
template=template,
title='uniprot_id: '+uniprot_id)
fig.update_xaxes(tickangle=-90,range=[0,99],rangeslider=dict(visible=True),dtick=1)
fig.update_yaxes(dtick=1)
fig.update_layout({
'plot_bgcolor': 'rgba(0, 0, 0, 0)',
'paper_bgcolor': 'rgba(0, 0, 0, 0)',
},font={'family':'Arial','size':11},
hoverlabel=dict(font=dict(family='Arial', size=14)))
fig.update_traces(
hovertemplate="<br>".join([
"<b>%{x} %{y}</b>"+
" (%{z:.2f})",
])+'<extra></extra>'
)
return fig
uid = st.selectbox("uniprot_id:", df)
fig = plot_interactive(uid)
fig.update_layout(width = 800, height = 600, autosize = False)
st.plotly_chart(fig, use_container_width=True)