adaptive file read+merge

vzv.py

import os
import pandas as pd
import platform
import re
import parmap
from tcrdist.swap_gene_name import adaptive_to_imgt

path= "" #path to key file
key = "" #list of adaptive files to include

files=pd.read_csv(os.path.join(path, key), sep="\t")

files=sorted(files['file'])
print(files)

count=1
vzv_list=list()

for file in files:
	if not file.endswith(".tsv"):
		continue
	df = pd.read_csv(os.path.join(path, file),
		sep="\t")
	df['file']=file
	df.dropna(subset = ["aminoAcid"], inplace=True)
	df.dropna(subset = ['vMaxResolved'], inplace=True)
	print(df.head(2))
	vzv_list.append(df)
	count=count+1
#	if count>5: break

vzv_tcrs = pd.concat(vzv_list, axis=0)
vzv_tcrs = vzv_tcrs[vzv_tcrs['sequenceStatus']=='In'].reset_index(drop = True)

vzv_tcrs.columns

cols = {'nucleotide': 'cdr3_b_nucseq',
		'aminoAcid': 'cdr3_b_aa',
		'count (templates/reads)': 'templates',
		'frequencyCount (%)': 'freq',
		'vMaxResolved': 'vMaxResolved',
		'jMaxResolved': 'jMaxResolved',
		'file': 'file'}

vzv_tcrs = vzv_tcrs[cols].rename(columns = cols)

vzv_tcrs['file'] = vzv_tcrs['file'].str.replace("gE_CD4","gE")

vzv_tcrs[['person', 'cell', 'type', 'TCR']] = vzv_tcrs.file.str.split('_', n=3, expand =True)
vzv_tcrs['source'] = vzv_tcrs['cell'] + "_" + vzv_tcrs['type']

vzv_tcrs['v_b_gene'] = vzv_tcrs['vMaxResolved'].apply(lambda x: adaptive_to_imgt['human'].get(x.split("*")[0]) if isinstance(x,str) else None)
vzv_tcrs['j_b_gene'] = vzv_tcrs['jMaxResolved'].apply(lambda x: adaptive_to_imgt['human'].get(x.split("*")[0]) if isinstance(x,str) else None)

cols = {'cdr3_b_nucseq': 'cdr3_b_nucseq',
		'cdr3_b_aa': 'cdr3_b_aa',
		'templates': 'count',
		'freq': 'freq',
		'person': 'person',
		'source': 'source',
		'v_b_gene': 'v_b_gene',
		'j_b_gene': 'j_b_gene'}
vzv_tcrs = vzv_tcrs[cols].rename(columns = cols)

## frequency wide version 
dfp = vzv_tcrs.pivot(index=['cdr3_b_nucseq', 'cdr3_b_aa', 'v_b_gene', 'j_b_gene', 'person'],
		columns='source', values='freq')
dfp = dfp.reset_index(drop = False)

## counts wide version
dft = vzv_tcrs.pivot(index=['cdr3_b_nucseq', 'cdr3_b_aa', 'v_b_gene', 'j_b_gene', 'person'],
		columns='source', values='count')
dft = dft.reset_index(drop = False)