This is a test version of Biostars. For the public version, visit https://www.biostars.org.
No quality in non-variant sites GATK

Heys,

I am doing the SNP calling with Haplotypecaller BP_Resolution, CombineGVCFs with convert-to-base-pair-resolution and GenotypeGVCFs with include-non-variant-sites with GATK and when I get my vcf file, the non-variant sites does not have any quality at all:

#CHROM  POS ID  REF ALT QUAL    FILTER  INFO    FORMAT  Leopards_1  Leopards_2
NC_001700.1 1   .   G   .   .   .   DP=40   GT:AD:DP:RGQ    0/0:26:26:78    0/0:14:14:42
NC_001700.1 2   .   G   .   .   .   DP=43   GT:AD:DP:RGQ    0/0:29:29:87    0/0:14:14:42
NC_001700.1 3   .   A   .   .   .   DP=43   GT:AD:DP:RGQ    ./.:0:29:0  ./.:0:14:0
NC_001700.1 4   .   C   .   .   .   DP=43   GT:AD:DP:RGQ    ./.:0:29:0  ./.:0:14:0
NC_001700.1 5   .   T   .   .   .   DP=71   GT:AD:DP:RGQ    0/0:50:50:99    0/0:21:21:63
NC_001700.1 6   .   A   .   .   .   DP=74   GT:AD:DP:RGQ    0/0:53:53:99    0/0:21:21:63
NC_001700.1 7   .   A   .   .   .   DP=85   GT:AD:DP:RGQ    0/0:59:59:99    0/0:26:26:78

Except for the variant ones:

NC_001700.1 63  .   T   A   6198.13 .   AC=4;AF=1.00;AN=4;DP=252;ExcessHet=3.0103;FS=0.000;MLEAC=4;MLEAF=1.00;MQ=48.90;QD=24.60;SOR=0.709   GT:AD:DP:GQ:PL  1/1:0,166:166:99:4057,496,0 1/1:0,86:86:99:2157,258,0

Do you know what is this happening?

gatk
• 2,128 views • link
written by gubrins gubrins

It seems there is a version problem, I just did the same runs with gatk v.4.1.3 and I got the quality parameter I was looking for, just in case it helps to anybody

• 0 views • link
written by gubrins gubrins

1 answer

IMHO GQ calculation requires an alternate allele. https://gatk.broadinstitute.org/hc/en-us/articles/360035890451-Calculation-of-PL-and-GQ-by-HaplotypeCaller-and-GenotypeGVCFs

if you have no ALT allele, this calculation is meaningless, that is why you get RGQ ( Reference GenotypeQuality ) instead of GQ.

• 0 views • link
written by Pierre Lindenbaum Pierre Lindenbaum

Thanks for your answer Pierre. But then how should I do to have a quality in the SNPs that are homozygous for my samples? Because I need them to calculate analyses as heterozygosity or ROHs.

• 0 views • link
written by gubrins gubrins

Log in to answer this question.

More posts like this