gen_gen.py

import re
import subprocess
from subprocess import PIPE
import sys
import time
import numpy as np
import numpy


impute_file = sys.argv[1]
sample_file= sys.argv[2]
info_file = sys.argv[3]

def dosage_convert(genos):
	out=(genos[0]*0)+(genos[1]*1)+(genos[2]*2)
	return(str(out))

#[0*p(AA)]+[1*p(AB)]+[2*p(BB)]

## debugging scripts
snps = {}
with open('test_impute.txt') as f_in:
	for line in f_in:
		if line:
			line_parse = line.strip('\r\n').split(' ')
			make_keys = ','.join(line_parse[:5])
			for i in range(5, len(line_parse), 3):
				print line_parse[i:i+3]
				genos = np.array(line_parse[i:i+3], dtype=float)
				genos1=dosage_convert(genos=genos)
				### function to convert genos into dosages goes here
				if make_keys in snps:
					get_geno = snps.get(make_keys)
					snps[make_keys] = str(get_geno)+','+str(genos1)
				else:
					snps[make_keys]= str(genos1)