output from comb script & electrostatic mut changes calculated
This commit is contained in:
parent
96ebb85069
commit
de1822f491
4 changed files with 250 additions and 167 deletions
|
|
@ -9,9 +9,9 @@ Created on Tue Aug 6 12:56:03 2019
|
|||
# FIXME: include error checking to enure you only
|
||||
# concentrate on positions that have structural info?
|
||||
|
||||
# FIXME: import dirs.py to get the basic dir paths available
|
||||
|
||||
#%% load libraries
|
||||
###################
|
||||
# load libraries
|
||||
import os, sys
|
||||
import pandas as pd
|
||||
#import numpy as np
|
||||
|
|
@ -52,19 +52,19 @@ from reference_dict import my_aa_dict #CHECK DIR STRUC THERE!
|
|||
#========================================================
|
||||
|
||||
#%% variable assignment: input and output paths & filenames
|
||||
|
||||
drug = 'pyrazinamide'
|
||||
gene = 'pncA'
|
||||
gene_match = gene + '_p.'
|
||||
|
||||
#=======
|
||||
#==========
|
||||
# input dir
|
||||
#=======
|
||||
#==========
|
||||
#indir = 'git/Data/pyrazinamide/input/original'
|
||||
indir = homedir + '/' + 'git/Data'
|
||||
#=========
|
||||
|
||||
#===========
|
||||
# output dir
|
||||
#=========
|
||||
#===========
|
||||
# several output files
|
||||
# output filenames in respective sections at the time of outputting files
|
||||
#outdir = 'git/Data/pyrazinamide/output'
|
||||
|
|
|
|||
|
|
@ -1,13 +1,12 @@
|
|||
#!/usr/bin/env python3
|
||||
# -*- coding: utf-8 -*-
|
||||
"""
|
||||
'''
|
||||
Created on Tue Aug 6 12:56:03 2019
|
||||
|
||||
@author: tanu
|
||||
"""
|
||||
'''
|
||||
|
||||
# FIXME: include error checking to enure you only
|
||||
# concentrate on positions that have structural info
|
||||
# FIXME: import dirs.py to get the basic dir paths available
|
||||
|
||||
#%% load libraries
|
||||
###################
|
||||
|
|
@ -16,147 +15,160 @@ import os, sys
|
|||
import pandas as pd
|
||||
#import numpy as np
|
||||
|
||||
#from pandas.api.types import is_string_dtype
|
||||
#from pandas.api.types import is_numeric_dtype
|
||||
|
||||
#====================================================
|
||||
# TASK: calculate how many mutations result in
|
||||
# electrostatic changes wrt wt
|
||||
# Input: mcsm and AF_OR file
|
||||
# output: mut_elec_changes_results.txt
|
||||
# Output: mut_elec_changes_results.txt
|
||||
#========================================================
|
||||
#%%
|
||||
####################
|
||||
#%% specify homedir as python doesn't recognise tilde
|
||||
homedir = os.path.expanduser('~')
|
||||
|
||||
# my working dir
|
||||
os.getcwd()
|
||||
homedir = os.path.expanduser('~') # spyder/python doesn't recognise tilde
|
||||
os.chdir(homedir + '/git/LSHTM_analysis/meta_data_analysis')
|
||||
os.getcwd()
|
||||
#%%
|
||||
from reference_dict import my_aa_dict #CHECK DIR STRUC THERE!
|
||||
#%%
|
||||
############# specify variables for input and output paths and filenames
|
||||
drug = "pyrazinamide"
|
||||
gene = "pnca"
|
||||
|
||||
datadir = homedir + "/git/Data"
|
||||
basedir = datadir + "/" + drug + "/input"
|
||||
#========================================================
|
||||
#%% variable assignment: input and output paths & filenames
|
||||
drug = 'pyrazinamide'
|
||||
gene = 'pncA'
|
||||
gene_match = gene + '_p.'
|
||||
|
||||
# input
|
||||
inpath = "/processed"
|
||||
#==========
|
||||
# data dir
|
||||
#==========
|
||||
#indir = 'git/Data/pyrazinamide/input/original'
|
||||
datadir = homedir + '/' + 'git/Data'
|
||||
|
||||
# uncomment as necessary
|
||||
in_filename = "/meta_data_with_AFandOR.csv"
|
||||
#in_filename = "/mcsm_complex1_normalised.csv" # probably simpler
|
||||
#==========
|
||||
# input dir
|
||||
#==========
|
||||
indir = datadir + '/' + drug + '/' + 'input'
|
||||
|
||||
infile = basedir + inpath + in_filename
|
||||
#print(infile)
|
||||
#============
|
||||
# output dir
|
||||
#============
|
||||
# several output files
|
||||
outdir = datadir + '/' + drug + '/' + 'output'
|
||||
|
||||
# output file
|
||||
outpath = "/output"
|
||||
outdir = datadir + "/" + drug + outpath
|
||||
out_filename = "/mut_elec_changes_results.txt"
|
||||
outfile = outdir + out_filename
|
||||
# specify output file
|
||||
out_filename = 'mut_elec_changes.txt'
|
||||
outfile = outdir + '/' + out_filename
|
||||
print('Output path: ', outdir)
|
||||
|
||||
#print(outdir)
|
||||
#%% end of variable assignment for input and output files
|
||||
#=============================================================
|
||||
#%% Read input files
|
||||
#in_filename = gene.lower() + '_meta_data_with_AFandOR.csv'
|
||||
in_filename = 'merged_df3.csv'
|
||||
infile = outdir + '/' + in_filename
|
||||
print('Reading input file (merged file):', infile)
|
||||
|
||||
if not os.path.exists(datadir):
|
||||
print('Error!', datadir, 'does not exist. Please ensure it exists. Dir struc specified in README.md')
|
||||
os.makedirs(datadir)
|
||||
exit()
|
||||
comb_df = pd.read_csv(infile, sep = ',')
|
||||
|
||||
if not os.path.exists(outdir):
|
||||
print('Error!', outdir, 'does not exist.Please ensure it exists. Dir struc specified in README.md')
|
||||
exit()
|
||||
|
||||
else:
|
||||
print('Dir exists: Carrying on')
|
||||
|
||||
################## end of variable assignment for input and output files
|
||||
#%%
|
||||
#==============================================================================
|
||||
############
|
||||
# STEP 1: Read file
|
||||
############
|
||||
meta_pnca = pd.read_csv(infile, sep = ',')
|
||||
print('Input filename: ', in_filename,
|
||||
'\nPath :', outdir,
|
||||
'\nNo. of rows: ', len(comb_df),
|
||||
'\nNo. of cols: ', infile)
|
||||
|
||||
# column names
|
||||
list(meta_pnca.columns)
|
||||
list(comb_df.columns)
|
||||
|
||||
#========
|
||||
# Step 2: iterate through the dict, create a lookup dict that i.e
|
||||
# lookup_dict = {three_letter_code: aa_prop_polarity}
|
||||
# Do this for both wild_type and mutant as above.
|
||||
#=========
|
||||
# initialise a sub dict that is lookup dict for three letter code to aa prop
|
||||
lookup_dict = dict()
|
||||
|
||||
for k, v in my_aa_dict.items():
|
||||
lookup_dict[k] = v['aa_calcprop']
|
||||
#print(lookup_dict)
|
||||
wt = meta_pnca['mutation'].str.extract('pnca_p.(\w{3})').squeeze() # converts to a series that map works on
|
||||
meta_pnca['wt_calcprop'] = wt.map(lookup_dict)
|
||||
mut = meta_pnca['mutation'].str.extract(r'\d+(\w{3})$').squeeze()
|
||||
meta_pnca['mut_calcprop'] = mut.map(lookup_dict)
|
||||
|
||||
# added two more cols
|
||||
|
||||
# clear variables
|
||||
del(k, v, wt, mut, lookup_dict)
|
||||
del(in_filename, infile, inpath)
|
||||
del(in_filename, infile)
|
||||
|
||||
#%%
|
||||
###########
|
||||
# Step 3: subset unique mutations
|
||||
###########
|
||||
meta_pnca_muts = meta_pnca.drop_duplicates(['Mutationinformation'], keep = 'first')
|
||||
non_struc = meta_pnca_muts[meta_pnca_muts.position == 186]
|
||||
#%% subset unique mutations
|
||||
df = comb_df.drop_duplicates(['Mutationinformation'], keep = 'first')
|
||||
|
||||
# remove pos non_struc 186 : (in case you used file with AF and OR)
|
||||
df = meta_pnca_muts[meta_pnca_muts.position != 186]
|
||||
total_muts = df.Mutationinformation.nunique()
|
||||
#df.Mutationinformation.count()
|
||||
print('Total mutations associated with structure: ', total_muts)
|
||||
|
||||
###########
|
||||
# Step 4: combine cols
|
||||
###########
|
||||
#%% combine aa_calcprop cols so that you can count the changes as value_counts
|
||||
# check if all muts have been categorised
|
||||
print('Checking if all muts have been categorised: ')
|
||||
if df['wt_calcprop'].isna().sum() == 0 & df['mut_calcprop'].isna().sum():
|
||||
print('PASS: No. NA detected i.e all muts have aa prop associated')
|
||||
else:
|
||||
print('FAIL: NAs detected i.e some muts remain unclassified')
|
||||
|
||||
df['aa_calcprop_combined'] = df['wt_calcprop']+ '->' + df['mut_calcprop']
|
||||
df['aa_calcprop_combined']
|
||||
df['wt_calcprop'].head()
|
||||
df['mut_calcprop'].head()
|
||||
|
||||
print('Combining wt_calcprop and mut_calcprop...')
|
||||
#df['aa_calcprop_combined'] = df['wt_calcprop']+ '->' + df['mut_calcprop']
|
||||
df['aa_calcprop_combined'] = df.wt_calcprop.str.cat(df.mut_calcprop, sep = '->')
|
||||
df['aa_calcprop_combined'].head()
|
||||
|
||||
mut_categ = df["aa_calcprop_combined"].unique()
|
||||
print('Total no. of aa_calc properties: ', len(mut_categ))
|
||||
print('Categories are: ', mut_categ)
|
||||
|
||||
# counting no. of muts in each mut categ
|
||||
|
||||
# way1: count values within each combinaton
|
||||
df.groupby('aa_calcprop_combined').size()
|
||||
#df.groupby('aa_calcprop_combined').count()
|
||||
|
||||
# way2: count values within each combinaton
|
||||
#df['aa_calcprop_combined'].value_counts()
|
||||
df['aa_calcprop_combined'].value_counts()
|
||||
|
||||
# comment: the two ways should be identical
|
||||
# groupby result order is similar to pivot table order
|
||||
# groupby result order is similar to pivot table order,
|
||||
# I prefer the value_counts look
|
||||
|
||||
#assign to variable: count values within each combinaton
|
||||
all_prop = df.groupby('aa_calcprop_combined').size()
|
||||
# assign to variable: count values within each combinaton
|
||||
all_prop = df['aa_calcprop_combined'].value_counts()
|
||||
|
||||
# convert to a df from Series
|
||||
ap_df = pd.DataFrame({'aa_calcprop': all_prop.index, 'mut_count': all_prop.values})
|
||||
|
||||
# subset df to contain only the changes in prop
|
||||
all_prop_change = ap_df[ap_df['aa_calcprop'].isin(['neg->neg','non-polar->non-polar','polar->polar', 'pos->pos']) == False]
|
||||
|
||||
|
||||
elec_count = all_prop_change.mut_count.sum()
|
||||
print('Total no.of muts with elec changes: ', elec_count)
|
||||
|
||||
# calculate percentage of electrostatic changes
|
||||
elec_changes = (elec_count/total_muts) * 100
|
||||
|
||||
print("Total number of electrostatic changes resulting from Mutation is (%):", elec_changes)
|
||||
|
||||
print('Total number of electrostatic changes resulting from Mutation is (%):', elec_changes)
|
||||
|
||||
# check no change muts
|
||||
no_change_muts = ap_df[ap_df['aa_calcprop'].isin(['neg->neg','non-polar->non-polar','polar->polar', 'pos->pos']) == True]
|
||||
|
||||
no_change_muts.mut_count.sum()
|
||||
|
||||
|
||||
###########
|
||||
# Step 5: output from console
|
||||
###########
|
||||
#sys.stdout = open(file, 'w')
|
||||
sys.stdout = open(outfile, 'w')
|
||||
|
||||
print(df.groupby('aa_calcprop_combined').size() )
|
||||
print("=======================================================================================")
|
||||
print("Total number of electrostatic changes resulting from Mutation is (%):", elec_changes)
|
||||
print("=======================================================================================")
|
||||
#print(no_change_muts, '\n',
|
||||
# all_prop_change)
|
||||
|
||||
print('======================\n'
|
||||
,'Unchanged muts'
|
||||
,'\n=====================\n'
|
||||
, no_change_muts
|
||||
,'\n=============================\n'
|
||||
, 'Muts with changed prop:'
|
||||
, '\n============================\n'
|
||||
, all_prop_change)
|
||||
|
||||
#print('======================================================================')
|
||||
#print('Total number of electrostatic changes resulting from Mutation is (%):', elec_changes)
|
||||
#print('Total no. of muts: ', total_muts)
|
||||
#print('Total no. of changed muts: ', all_prop_change.mut_count.sum())
|
||||
#print('Total no. of unchanged muts: ', no_change_muts.mut_count.sum() )
|
||||
#print('=======================================================================')
|
||||
|
||||
print('========================================================================'
|
||||
, '\nTotal number of electrostatic changes resulting from Mtation is (%):', elec_changes
|
||||
, '\nTotal no. of muts: ', total_muts
|
||||
, '\nTotal no. of changed muts: ', all_prop_change.mut_count.sum()
|
||||
, '\nTotal no. of unchanged muts: ', no_change_muts.mut_count.sum()
|
||||
, '\n=========================================================================')
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue