Extract_sequences.py für mehrere CSV-Dateien

extract_sequences_multi.py

#! /usr/bin/env python2


import argparse
import csv


def write_sequence(name, sequence, fh):
    fh.write("> " + name + "\n")
    fh.write(sequence + "\n")
    fh.write("\n")


def construct_sequences(file):
    fh = open(file)
    csv_data = csv.DictReader(fh, delimiter=',')
    tmp_seqid = None
    fullseq = ''
    for row in csv_data:
        seqid = row['protein.key']
        if (seqid != tmp_seqid) and (tmp_seqid is not None):
            yield (name, fullseq)
            fullseq = ''

        if name == 'Crap':
            name = row['protein.Accession']
        else:
            name = row['protein.Entry']
        seq = row['peptide.seq']
        seqstart = int(row['peptide.seqStart'])
        seqlen = len(seq)
        seqend = seqstart + seqlen

        if len(fullseq) < seqend:
            # Add missing residues as gaps
            fullseq += '-' * (seqend - len(fullseq))
        # Replace gaps at sequence positions with seq
        fullseq = fullseq[:seqstart] + seq + fullseq[seqend:]
        tmp_seqid = seqid
    fh.close()
    # Yield the last sequence as well
    yield (name, fullseq)


def parse_args():
    p = argparse.ArgumentParser()
    p.add_argument("csv_infile", metavar='INFILE', nargs='+', help="CSV input data")
    p.add_argument("fasta_outfile", metavar='OUTFILE', help="FASTA output file")
    return p.parse_args()


def main():
    args = parse_args()
    outfile = open(args.fasta_outfile, 'w')
    for infile in args.csv_infile:
        for name, seq in construct_sequences(csv_infile):
            write_sequence(name, seq, outfile)
    outfile.close()


if __name__ == '__main__':
    main()