121 lines
3.7 KiB
Python
121 lines
3.7 KiB
Python
#!/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
|
|
#%% homedir
|
|
homedir = os.path.expanduser('~')
|
|
print('My homedir is:', homedir)
|
|
#%%
|
|
def request_calculation(pdb_file
|
|
, mutation_list
|
|
, batch_no
|
|
, chain
|
|
, my_email
|
|
, prediction_url
|
|
, output_dir
|
|
#, gene_name
|
|
#, out_url_file
|
|
):
|
|
"""
|
|
Makes a POST request for dynamut predictions.
|
|
|
|
@param pdb_file: valid path to pdb structure
|
|
@type string
|
|
|
|
@param mutation_list: list of mutations (1 per line) of the format: {WT}<POS>{Mut}
|
|
@type string
|
|
|
|
@param batch_no: batch no so it can be added as a suffix to the the outfile
|
|
@type int
|
|
|
|
@param chain: single-letter(caps)
|
|
@type chr
|
|
|
|
@param prediction_url: dynamut url for prediction
|
|
@type string
|
|
|
|
@param output_dir: output dir
|
|
@type string
|
|
|
|
@param gene_name: name of gene
|
|
@type string
|
|
|
|
#@param out_url_file: name of output file with batch no. as suffix
|
|
@type string
|
|
|
|
@return txt file containing batch no. of snps processed (i.e out_url_file)
|
|
@type string
|
|
"""
|
|
|
|
with open(pdb_file, "rb") as pdb_file, open (mutation_list, "rb") as mutation_list:
|
|
files = {"wild": pdb_file
|
|
, "mutation_list": mutation_list}
|
|
body = {"chain": chain
|
|
, "email": my_email}
|
|
|
|
response = requests.post(prediction_url, files = files, data = body)
|
|
print(response.status_code)
|
|
if response.history:
|
|
print('\nPASS: valid submission. Fetching result url')
|
|
url_match = re.search('/dynamut/results_prediction/.+(?=")', response.text)
|
|
url = host + url_match.group()
|
|
print('\nURL for snp batch no ', str(batch_no), ':', url)
|
|
|
|
#===============
|
|
# writing file: result urls
|
|
#===============
|
|
dynamut_temp_dir = outdir + '/dynamut_temp'
|
|
if not os.path.exists(dynamut_temp_dir):
|
|
print('\nCreating dynamut_temp in outdir', outdir )
|
|
os.makedirs(dynamut_temp_dir)
|
|
|
|
out_url_file = dynamut_temp_dir + '/dynamut_result_url_batch_' + str(batch_no) + '.txt'
|
|
print('\nWriting output url file:', out_url_file)
|
|
myfile = open(out_url_file, 'a')
|
|
myfile.write(url)
|
|
myfile.close()
|
|
#%%globals!?
|
|
host = 'http://biosig.unimelb.edu.au'
|
|
#prediction_url = f"{host}/dynamut/prediction_list"
|
|
#print(prediction_url)
|
|
|
|
#gene = 'gid'
|
|
drug = 'streptomycin'
|
|
datadir = homedir + '/git/Data'
|
|
indir = datadir + '/' + drug + '/input'
|
|
outdir = datadir + '/' + drug + '/output'
|
|
#outdir = homedir + '/git/LSHTM_analysis/dynamut' # for example
|
|
|
|
my_chain = 'A'
|
|
my_pdb_file = homedir + '/git/Data/streptomycin/input/gid_complex.pdb'
|
|
|
|
# batch 1: 00.txt
|
|
#my_mutation_list = homedir + '/git/Data/streptomycin/output/snp_batches/50/snp_batch_00.txt'
|
|
#my_batch = 1
|
|
|
|
# batch 2: 01.txt
|
|
my_mutation_list = outdir + '/snp_batches/50/snp_batch_01.txt'
|
|
my_batch = 2
|
|
|
|
# %% call this function
|
|
request_calculation (pdb_file = my_pdb_file
|
|
, mutation_list = my_mutation_list
|
|
, chain = my_chain
|
|
, my_email = 'tanushree.tunstall@lshtm.ac.uk'
|
|
, prediction_url = f"{host}/dynamut/prediction_list"
|
|
, output_dir = outdir
|
|
, batch_no = my_batch)
|