I am using a large VCF file (ftp://biodisk.org/Store/Genome/African/Mota_man/Bam_and_VCF/GB20_sort_merge_dedup_l30_IR_q30_mapDamage_Entire.vcf.gz), which I need to filter before converting it to Plink.

I was able to use a custom script to filter it down to the sites I have comparative data for, and I wrote a quick python script to filter out trialleic sites and Qual < 50.

#! /usr/bin/env python

OutFile = open('Mota_2.vcf', 'w')
BadFile = open('Mota_bad.vcf', 'w')
InFile = open('GB20_sort_merge_dedup_l30_IR_q30_mapDamage_Entire_filtered.vcf', 'r')

#filtering the LowQual sites and Qual < 50

import re
for Line in InFile:
    Line=Line.strip('\n')
    first = Line[0]
    if first == '#':
        OutFile.write(Line + '\n')
        #BadFile.write(Line + '\n')
    else:
        ElementListQ = Line.split('\t')
        Alt = ElementListQ[4]
        Qual = ElementListQ[5]
        if len(Alt) != 1:
            BadFile.write(Line + '\n')
        elif ElementListQ[6] == 'LowQual':
            BadFile.write(Line + '\n')
        elif float(Qual) >= 50:
            OutFile.write(Line + '\n')
        else:
            BadFile.write(Line + '\n')
InFile.close()
OutFile.close()
BadFile.close()

The authors talk about only using sites with MQ > 30.

The MQ is tied in the INFO field. Here is an example of one of those

DP=16;VDB=0.958132;SGB=-0.683931;RPB=1;MQB=1;MQSB=0.852144;BQB=1;MQ0F=0;AF1=1;AC1=2;DP4=0,1,10,3;MQ=35;FQ=-33.9875;PV4=0.285714,0.460053,0.350463,1

I was wondering what would be a good of filtering for this? Thanks!