convert-to-csv.py

import pandas as pd
import csv
import math
import argparse


parser = argparse.ArgumentParser()

parser.add_argument('input_filename')
parser.add_argument('file_name')
args = parser.parse_args()

df = pd.read_csv(args.input_filename, skipinitialspace=True, sep=' ')  # read file for csv processing
print("file processed Successfully")


def convert_to_csv_clustered():
    df['-log'] = list(map(lambda x: -math.log10(x), list(df['P'])))
    p = df.loc[df['-log'] >= 5, '-log']
    chr = df.loc[df['-log'] >= 5, 'CHR']
    a1 = df.loc[df['-log'] >= 5, 'A1']
    a2 = df.loc[df['-log'] >= 5, 'A2']
    snp = df.loc[df['-log'] >= 5, 'SNP']
    bp = df.loc[df['-log'] >= 5, 'BP']
    data = list(zip(chr, p, snp, a1, a2, bp))

    with open(f'clustered-{args.file_name}', 'w') as out:
        out_csv = csv.writer(out)
        out_csv.writerow(['CHR', 'P', 'SNP', 'A1', 'A2', 'BP'])
        for row in data:
            out_csv.writerow(row)


def convert_to_csv():
    temp = list(map(lambda x: -math.log10(x), list(df['P'])))
    p = list(map(lambda x: round(x, 3), temp))
    chr = list(df['CHR'])
    if len(df) >= 50000:
        data = list(set(zip(chr, p)))
    else:
        data = list((zip(chr, p)))
    with open(f'full-{args.file_name}', 'w') as out:
        out_csv = csv.writer(out)
        out_csv.writerow(['CHR', 'P'])
        for row in data:
            out_csv.writerow(row)


convert_to_csv_clustered()
convert_to_csv()