Ad hoc filter script

filter_contigs.py

from random import randint
import pandas as pd

def main():
    with open("/home/chponte/Descargas/Bmedia_15MD_002MAF_filtered.new.vcf", "x") as fout:
        with open("/home/chponte/Descargas/Bmedia_15MD_002MAF_filtered.vcf", "r") as fin:
            l = fin.readline()
            while not l.startswith("#CHROM"):
                fout.write(l)
                l = fin.readline()
            fout.write(l)
            df = pd.read_csv(fin, sep="\t", names=l.strip().split("\t"))
        for _, sub_df in df.groupby("#CHROM", sort=False):
            index = randint(0, sub_df.shape[0] - 1)
            # index = 0
            s = sub_df.iloc[index:index+1, :]
            s.to_csv(fout, sep="\t", header=False, index=False)

if __name__ == "__main__":
    main()