(up) need to add other columns

liftover_bed.py

#!/usr/bin/python3
from pyliftover import LiftOver
import pandas as pd
import argparse
import mapply
import sys
import os

mapply.init(
    n_workers=8,
    chunk_size=100,
    max_chunks_per_worker=8,
    progressbar=False
)

def get_coordinate(row, lo):
    """ Translate chr, start and end postions from hg19 to hg38 """
    chr = row[0]
    start = row[1]
    end = row[2]
    try:
        # We make start and end separated
        coord_start = lo.convert_coordinate(chr, start)[0]
        coord_end = lo.convert_coordinate(chr, end)[0]
        return [coord_start[0], coord_start[1], coord_start[2], coord_end[1]]
    except:
        return ["not available", 0, "-", 0]

def main(args):
    """ Args function to run liftover from hg19 to hg38 """
    bed = os.path.abspath(args.bed)
    genome = args.genome
    output_dir = os.path.abspath(args.output_dir)
    if genome == "hg38":
        lo = LiftOver('hg19', 'hg38')
    else:
        lo = LiftOver('hg38', 'hg19')
    df = pd.read_csv(bed, header=None, sep="\t")

    # Create new column with the region translated
    df["region_notation"] = df.mapply(
        lambda row: get_coordinate(row, lo), axis=1
    )
    # Split the "region_notation" content into its individual columns
    df[f"{genome}_chr"] = df.apply(
        lambda row: row["region_notation"][0], axis=1
    )
    df[f"{genome}_start"] = df.apply(
        lambda row: row["region_notation"][1], axis=1
    )
    df[f"{genome}_end"] = df.apply(
        lambda row: row["region_notation"][3], axis=1
    )
    # filter the rows that wasn't translated and that start is greater than end
    coord_df = df[~df[f"{genome}_chr"].str.contains("not")]
    filtered_df = coord_df[coord_df[f"{genome}_start"]<coord_df[f"{genome}end"]]
    filtered_df = filtered_df.sort_values([f"{genome}_chr", f"{genome}_start", f"{genome}_end"])
    del df, coord_df

    # Create new bed file with converted genomic coordinates
    filtered_df.to_csv(
        bed.replace(".bed", f"_{genome}.bed"),
        sep="\t",
        index=False,
        header=False,
        columns=[f"{genome}_chr", f"{genome}_start", f"{genome}_end"]
    )

if __name__ == "__main__":
    """ Arguments for the main function """
    parser = argparse.ArgumentParser(
        description="Script to liftover BED", usage='''Usage:
        python liftover_bed.py -b file.bed
        ''')
    parser.add_argument("-b", "--bed", help="BED file", type=str, required=True)
    parser.add_argument("-o", "--output_dir", help="Path to output directory", type=str)
    parser.add_argument("-g", "--genome", help="Genome you want to translate to (default: hg38 [hg19 > hg38])", type=str, required=False, default="hg38")
    parser.set_defaults(func=main)
    if len(sys.argv) == 1:
        parser.print_help()
        sys.exit(1)
    args = parser.parse_args()
    args.func(args)