137 lines
5.2 KiB
Python
137 lines
5.2 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
|
|
import numpy as np
|
|
from pandas.api.types import is_string_dtype
|
|
from pandas.api.types import is_numeric_dtype
|
|
#%%#####################################################################
|
|
def format_dynamut2_output(dynamut_output_csv):
|
|
"""
|
|
@param dynamut_output_csv: file containing dynamut2 results for all muts
|
|
which is the result of combining all dynamut2_output batch results, and using
|
|
bash scripts to combine all the batch results into one file.
|
|
Dynamut2ran manually from batches
|
|
Formatting df to a pandas df and output as csv.
|
|
@type string
|
|
|
|
@return (not true) formatted csv for dynamut output
|
|
@type pandas df
|
|
|
|
"""
|
|
#############
|
|
# Read file
|
|
#############
|
|
dynamut_data_raw = pd.read_csv(dynamut_output_csv, sep = ',')
|
|
|
|
# strip white space from both ends in all columns
|
|
dynamut_data = dynamut_data_raw.apply(lambda x: x.str.strip() if x.dtype == 'object' else x)
|
|
|
|
dforig_shape = dynamut_data.shape
|
|
print('dimensions of input file:', dforig_shape)
|
|
|
|
#%%============================================================================
|
|
#####################################
|
|
# create binary cols for ddg_dynamut2
|
|
# >=0: Stabilising
|
|
######################################
|
|
outcome_cols = ['ddg_dynamut2']
|
|
|
|
# col test: ddg_dynamut
|
|
#len(dynamut_data[dynamut_data['ddg_dynamut'] >= 0])
|
|
#dynamut_data['ddg_dynamut_outcome'] = dynamut_data['ddg_dynamut'].apply(lambda x: 'Stabilising' if x >= 0 else 'Destabilising')
|
|
#len(dynamut_data[dynamut_data['ddg_dynamut_outcome'] == 'Stabilising'])
|
|
|
|
print('\nCreating classification cols for', len(outcome_cols), 'columns'
|
|
, '\nThese are:')
|
|
|
|
for cols in outcome_cols:
|
|
print(cols)
|
|
|
|
tot_muts = dynamut_data[cols].count()
|
|
print('\nTotal entries:', tot_muts)
|
|
|
|
outcome_colname = cols + '_outcome'
|
|
print(cols, ':', outcome_colname)
|
|
c1 = len(dynamut_data[dynamut_data[cols] >= 0])
|
|
dynamut_data[outcome_colname] = dynamut_data[cols].apply(lambda x: 'Stabilising' if x >= 0 else 'Destabilising')
|
|
c2 = len(dynamut_data[dynamut_data[outcome_colname] == 'Stabilising'])
|
|
if c1 == c2:
|
|
print('\nPASS: outcome classification column created successfully'
|
|
, '\nColumn created:', outcome_colname
|
|
#, '\nNo. of stabilising muts: ', c1
|
|
#, '\nNo. of DEstabilising muts: ', tot_muts-c1
|
|
, '\n\nCateg counts:\n', dynamut_data[outcome_colname].value_counts() )
|
|
|
|
else:
|
|
print('\nFAIL: outcome classification numbers MISmatch'
|
|
, '\nexpected length:', c1
|
|
, '\nGot:', c2)
|
|
|
|
#%%=====================================================================
|
|
################################
|
|
# scale all ddg_dynamut2 values
|
|
#################################
|
|
# Rescale values in all ddg_dynamut2 col col b/w -1 and 1 so negative numbers
|
|
# stay neg and pos numbers stay positive
|
|
|
|
outcome_cols = ['ddg_dynamut2']
|
|
|
|
for cols in outcome_cols:
|
|
#print(cols)
|
|
col_max = dynamut_data[cols].max()
|
|
col_min = dynamut_data[cols].min()
|
|
print( '\n===================='
|
|
, '\nColname:', cols
|
|
, '\n===================='
|
|
, '\nMax: ', col_max
|
|
, '\nMin: ', col_min)
|
|
|
|
scaled_colname = cols + '_scaled'
|
|
print('\nCreated scaled colname for', cols, ':', scaled_colname)
|
|
col_scale = lambda x : x/abs(col_min) if x < 0 else (x/col_max if x >= 0 else 'failed')
|
|
|
|
dynamut_data[scaled_colname] = dynamut_data[cols].apply(col_scale)
|
|
|
|
col_scaled_max = dynamut_data[scaled_colname].max()
|
|
col_scaled_min = dynamut_data[scaled_colname].min()
|
|
print( '\n===================='
|
|
, '\nColname:', scaled_colname
|
|
, '\n===================='
|
|
, '\nMax: ', col_scaled_max
|
|
, '\nMin: ', col_scaled_min)
|
|
|
|
#%%=====================================================================
|
|
#############
|
|
# reorder columns
|
|
#############
|
|
dynamut_data.columns
|
|
dynamut_data_f = dynamut_data[['mutationinformation'
|
|
, 'chain'
|
|
, 'ddg_dynamut2'
|
|
, 'ddg_dynamut2_scaled'
|
|
, 'ddg_dynamut2_outcome']]
|
|
|
|
if len(dynamut_data.columns) == len(dynamut_data_f.columns) and sorted(dynamut_data.columns) == sorted(dynamut_data_f.columns):
|
|
print('\nPASS: outcome_classification, scaling and column reordering completed')
|
|
else:
|
|
print('\nFAIL: Something went wrong...'
|
|
, '\nExpected length: ', len(dynamut_data.columns)
|
|
, '\nGot: ', len(dynamut_data_f.columns))
|
|
sys.exit()
|
|
|
|
return(dynamut_data_f)
|
|
#%%#####################################################################
|
|
|