added residue depth processing to generate df

meta_data_analysis/rd_df.py

@@ -0,0 +1,137 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+'''
+Created on Tue Aug  6 12:56:03 2019
+
+@author: tanu
+'''
+#=============================================================================
+# Task: Residue depth (rd) processing to generate a df with residue_depth(rd)
+# values
+
+# FIXME
+# Input: '.tsv' i.e residue depth txt file (output from .zip file manually
+# downloaded from the website). 
+# This should be integrated into the pipeline
+
+# Output: .csv with 3 cols i.e position, rd_values & 3-letter wt aa code(caps)
+#=============================================================================
+#%% load packages
+import sys, os
+import pandas as pd
+#import numpy as np
+#=============================================================================
+#%% specify input and curr dir
+homedir = os.path.expanduser('~')
+
+# set working dir
+os.getcwd()
+os.chdir(homedir + '/git/LSHTM_analysis/meta_data_analysis')
+os.getcwd()
+#=============================================================================
+#%% variable assignment: input and output 
+drug = 'pyrazinamide'
+gene = 'pncA'
+gene_match = gene + '_p.'
+
+#==========
+# data dir
+#==========
+#indir = 'git/Data/pyrazinamide/input/original'
+datadir = homedir + '/' + 'git/Data'
+
+#=======
+# input
+#=======
+#indir = 'git/Data/pyrazinamide/input/original'
+indir = datadir + '/' + drug + '/' + 'output'
+in_filename = '3pl1_rd.tsv'
+infile = indir + '/' + in_filename
+print('Input filename:', in_filename
+      , '\nInput path:', indir)
+
+#=======
+# output 
+#=======
+outdir =   datadir + '/' + drug + '/' + 'output'
+out_filename = gene.lower() + '_rd.csv'
+outfile =  outdir + '/' + out_filename
+print('Output filename:', out_filename
+      , '\nOutput path:', outdir)
+
+print('======================================================================')
+#%% end of variable assignment for input and output files
+#=======================================================================
+#%%  Read input file
+rd_data  = pd.read_csv(infile, sep = '\t')  
+print('Reading input file:', infile
+      , '\nNo. of rows:', len(rd_data)
+      , '\nNo. of cols:', len(rd_data.columns))
+
+print('Column names:', rd_data.columns)
+
+print('======================================================================')
+#========================
+# creating position col
+#========================
+# Extracting residue number from index and assigning 
+# the values to a column [position]. Then convert the position col to numeric.
+rd_data['position'] = rd_data.index.str.extract('([0-9]+)').values
+
+# converting position to numeric      
+rd_data['position'] = pd.to_numeric(rd_data['position'])
+rd_data['position'].dtype
+
+print('Extracted residue num from index and assigned as a column:'
+      , '\ncolumn name: position'
+      , '\ntotal no. of cols now:', len(rd_data.columns))
+print('======================================================================')
+
+#========================
+# Renaming amino-acid 
+# and all-atom cols
+#========================
+print('Renaming columns:'
+      ,'\ncolname==> # chain:residue: wt_3letter_caps'
+      ,'\nYES... the column name *actually* contains a # ..!'
+      ,'\ncolname==> all-atom: rd_values')
+
+print('======================================================================')
+
+rd_data.rename(columns = {'# chain:residue':'wt_3letter_caps', 'all-atom':'rd_values'}, inplace = True)
+print('Column names:', rd_data.columns)
+
+#========================
+# extracting df with the
+# desired columns
+#========================
+print('Extracting relevant columns for writing df as csv')
+
+rd_df = rd_data[['position','rd_values','wt_3letter_caps']]
+
+if len(rd_df) == len(rd_data):
+    print('PASS: extracted df has expected no. of rows'
+          ,'\nExtracted df dim:'
+          ,'\nNo. of rows:', len(rd_df)
+          ,'\nNo. of cols:', len(rd_df.columns))
+else:
+    print('FAIL: no. of rows mimatch'
+          ,'\nExpected no. of rows:', len(rd_data)
+          ,'\nGot no. of rows:', len(rd_df))
+    
+print('======================================================================')
+
+#%% write file
+print('Writing file:'
+      , '\nFilename:', out_filename
+      , '\nPath:',  outdir)
+
+rd_df.to_csv(outfile, header = True, index = False)
+
+print('======================================================================')
+print('Finished writing:', out_filename
+      , '\nNo. of rows:', len(rd_df)
+      , '\nNo. of cols:', len(rd_df.columns))
+
+#%% end of script
+#=======================================================================