Dirty python script to merge fasta files

Posted on July 24, 2021

an AI generated picture (Midjourney) with prompt; 'cat :: pop art :: fun -'. You can share and adapt this image following a CC BY-SA 4.0 licence

Motivation & Requirements

Here is a Dirty python script to look in a directory, find fasta files (ext. ".fa"), and modify the header and merge them into a single fasta file. This will only look one directory down. It is not recursive. It won't even check if the directory records are directories, so it is pretty fragile.

I wrote this in a HURRY.

Requires:

Python 3.7
Biopython module

Code

import os
from Bio import SeqIO, Seq

input_dir = "/home/ubuntu/output_dir"

all_fasta = []
for dir_name in os.listdir(input_dir):
    if dir_name.startswith('EBRE'):
        output_dir = os.path.join(input_dir, dir_name)
        fasta_consensus = [os.path.join(output_dir, y)
            for y in os.listdir(output_dir) if y.endswith('.fa')]
        if len(fasta_consensus) == 1:
            rec = SeqIO.parse(open(fasta_consensus[0]), 'fasta')
            for fas in rec:
                fas.id = fas.id.split('_')[1]
                fas.decription = ''
                all_fasta.append(fas)

with open("merged_output.fasta", "w") as output_handle:
    SeqIO.write(all_fasta, output_handle, "fasta")