Entering edit mode
12.7 years ago
Johan
▴
890
I have a problem with my vcf files.
I have called SNPs using the GATK UnifiedGenotyper with the AlleleBalance annotator. For some of the SNPs the allele balance has been annotated, and for others it hasn't. (See the example below) I would really appreciate any help in understanding why this is the case.
chr5 112176559 rs866006 T G 2598.39 PASS **AB=0.502;AC=1**;AF=0.50;AN=2;BaseQRankSum=-9.353;DB;DP=329;Dels=0.00;FS=1.005;HRun=0;HaplotypeScore=24.1819;MQ=32.65;MQ0=18;MQRankSum=-8.933;QD=7.90;ReadPosRankSum=-0.599 GT:AD:DP:GQ:PL 0/1:164,163:329:99:2628,0,3234
chr5 112176756 rs459552 T A 88430.88 PASS **AC=2**;AF=1.00;AN=2;BaseQRankSum=3.447;DB;DP=2419;Dels=0.00;FS=0.000;HRun=0;HaplotypeScore=30.4979;MQ=34.64;MQ0=100;MQRankSum=-3.456;QD=36.56;ReadPosRankSum=-2.099 GT:AD:DP:GQ:PL 1/1:5,2409:2419:99:88431,6901,0
As far as I know AB= (Ref/Ref+ALT) and it is mentioned just for heterozygous variants.But for homozygous variants it is recorded zero.It was what I have seen in my out put file.
I think it is the opposite AB=(ALT/REF+ALT) but I have seen it written both ways... I noticed in my VCFs if AB=1 or AB=0 it still means homozygous variant. If AB=1 it is usually an indel and if it's AB=0 it's a SNV. If AB = 0.### then it is heterozygous. It is not explained in the VCF docs. But here is a helpful snippet from a paper
https://www.nature.com/articles/ncomms12824