602panel_test/vcfsplit_v0_20220929_finish.py

import pandas as pd
import os,sys
import argparse



def vcfsplit(inputdata):
	#inputdata = '/cgdata/bioproject/pancancer602gene/CGB0183/1SV_varscan_pair/Lib26611T.merge.Somatic.hc.nohead.vcf'
	dir = inputdata.split('.')
	vcfdata = pd.read_table(inputdata, sep='\t', header=None, low_memory=False)
	chrom = pd.DataFrame(vcfdata[0].drop_duplicates(keep='first'))
	chrom.reset_index(drop=True, inplace=True)
	for i in range(len(chrom)):
		chrid = chrom.loc[i, 0]
		chr_vcf = vcfdata[vcfdata[0] == chrid]
		chr_vcf.reset_index(drop=True, inplace=True)
		outputname = ('.'.join(dir[0:(len(dir) - 1)])) + '_' + str(chrid) + '.vcf'
		chr_vcf.to_csv(outputname, sep='\t', index=False, header=None)

if __name__=='__main__':
	parser = argparse.ArgumentParser(description='filter the MSI')
	parser.add_argument('-i', '--inputdata', type=str, help='the inputdata of vcf')
	parser.add_argument('-s', '--sample', type=str, help='the sample name')
	args = parser.parse_args()
	Inputdata = args.inputdata
	vcfsplit(Inputdata)