LSHTM_analysis/dynamut/get_results_dynamut.py

98 lines
4.6 KiB
Python
Executable file

#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Created on Wed Aug 19 14:33:51 2020
@author: tanu
"""
#%% load packages
import os,sys
import subprocess
import argparse
import requests
import re
import time
from bs4 import BeautifulSoup
import pandas as pd
from pandas.api.types import is_string_dtype
from pandas.api.types import is_numeric_dtype
#%%#####################################################################
def get_results(url_file, host_url, output_dir, outfile_suffix):
# initilialise empty df
dynamut_results_out_df = pd.DataFrame()
with open(url_file, 'r') as f:
for count, line in enumerate(f):
line = line.strip()
print('URL no.', count+1, '\n', line)
#batch_response = requests.get(line, headers=headers)
batch_response = requests.get(line)
batch_soup = BeautifulSoup(batch_response.text, features = 'html.parser')
# initilialise empty df
#dynamut_results_df = pd.DataFrame()
for a in batch_soup.find_all('a', href=True, attrs = {'class':'btn btn-default btn-sm'}):
print ("Found the URL:", a['href'])
single_result_url = host_url + a['href']
snp = re.search(r'([A-Z]+[0-9]+[A-Z]+$)', single_result_url).group(0)
print(snp)
print('\nGetting results from:', single_result_url)
result_response = requests.get(single_result_url)
if result_response.status_code == 200:
print('\nFetching results for SNP:', snp)
# extract results using the html parser
soup = BeautifulSoup(result_response.text, features = 'html.parser')
#web_result_raw = soup.find(id = 'predictions').get_text()
ddg_dynamut = soup.find(id = 'ddg_dynamut').get_text()
ddg_encom = soup.find(id = 'ddg_encom').get_text()
ddg_mcsm = soup.find(id = 'ddg_mcsm').get_text()
ddg_sdm = soup.find(id = 'ddg_sdm').get_text()
ddg_duet = soup.find(id = 'ddg_duet').get_text()
dds_encom = soup.find(id = 'dds_encom').get_text()
param_dict = {"mutationinformation" : snp
, "ddg_dynamut" : ddg_dynamut
, "ddg_encom" : ddg_encom
, "ddg_mcsm" : ddg_mcsm
, "ddg_sdm" : ddg_sdm
, "ddg_duet" : ddg_duet
, "dds_encom" : dds_encom
}
results_df = pd.DataFrame.from_dict(param_dict, orient = "index").T
print('Result DF:', results_df, 'for URL:', line)
#dynamut_results_df = dynamut_results_df.append(results_df)#!1 too many!:-)
dynamut_results_out_df = dynamut_results_out_df.append(results_df)
#print(dynamut_results_out_df)
#============================
# Writing results file: csv
#============================
dynamut_results_dir = output_dir + 'dynamut_results/'
if not os.path.exists(dynamut_results_dir):
print('\nCreating dir: dynamut_results within:', output_dir )
os.makedirs(dynamut_results_dir)
print('\nWriting dynamut results df')
print('\nResults File:'
, '\nNo. of rows:', dynamut_results_out_df.shape[0]
, '\nNo. of cols:', dynamut_results_out_df.shape[1])
print(dynamut_results_out_df)
#dynamut_results_out_df.to_csv('/tmp/test_dynamut.csv', index = False)
# build out filename
out_filename = dynamut_results_dir + 'dynamut_output_' + outfile_suffix + '.csv'
dynamut_results_out_df.to_csv(out_filename, index = False)
# TODO: add as a cmd option
# Download .tar.gz file
prediction_number = re.search(r'([0-9]+$)', line).group(0)
tgz_url = f"{host_url}/dynamut/results_file/results_" + prediction_number + '.tar.gz'
tgz_filename = dynamut_results_dir + outfile_suffix + '_results_' + prediction_number + '.tar.gz'
response_tgz = requests.get(tgz_url, stream = True)
if response_tgz.status_code == 200:
print('\nDownloading tar.gz file:', tgz_url
, '\n\nSaving file as:', tgz_filename)
with open(tgz_filename, 'wb') as f:
f.write(response_tgz.raw.read())
#%%#####################################################################