Get n genes

get_n_genes.py

#!/usr/bin/env python3
# -*- coding: UTF-8 -*-

import sys
import argparse
import pandas as pd 

def main(table: str, n_genes: int, target_colname: str) -> None:
    data = (
        pd.read_csv(
            table,
            sep='\t',
        )
        .loc[
            lambda df_: df_.padj.notna(),
            target_colname,
        ]
        .sort_values(ascending=False)
        .head(n_genes)
        .index
    )

    print(
       *data,
        sep='\n',
        file=sys.stdout 
    )

if __name__ == '__main__':
    parser = argparse.ArgumentParser(
        prog='get_top_n',
        description='Reads a table and extract the N top expressed genes.',
    )

    parser.add_argument(
        'table',
    )

    parser.add_argument(
        '-n', '--n_genes',
        help='Number of top genes to take.',
        default=5,
        type=int
    )

    parser.add_argument(
        '-t', '--target_colname',
        help='Target column to sort.',
        type=str
    )

    args = vars(parser.parse_args())

    main(
        **args
    )

Instructions

Download test table

# md5sum -> 9fa6acda7b96d7b1bf423d5f843425f5  DESeq2_results.txt
wget https://raw.githubusercontent.com/snandiDS/prokseq-v2.0/master/example_output/Output/DiffExpResults/DESeq2_results.txt

Run the script

This command will pull a python image with pandas and run the script. Fell free to replace the variable values.

docker run --rm -it \
    -v $PWD:/tmp/run \
    -w /tmp/run \
    amancevice/pandas:alpine-2.1.2 \
        python3 ./get_n_genes.py  \
            -n 10 \
            -t 'padj' \
            DESeq2_results.txt