Test-drive the GATK tools and Best Practices pipelines on Terra
Check out this blog post to learn how you can get started with GATK and try out the pipelines in preconfigured workspaces (with a user-friendly interface!) without having to install anything.
Help me understand why this variant was called
This variant was called with MuTect2 T/N mode :
chr7 152008861 . A T . PASS \
If i read this correctly :
AF for this SNP (A->T) is 0.112 (11.2%) in the tumor sample.
AF for this SNP (A->T) is 0.055 in the normal sample.
Depth at this position (AD) in tumor sample : 201 (ref), 0 (alt).
Depth at this position (AD) in normal sample : 445 (ref), 0 (alt).
This seems strange to call a SNP with AF = 0.112 when no reads at all show up for the alt_allele in the AD field.
We have nothing to support alt_allele in the QSS field either (for both tumor and normal sample, the sum of base quality scores for alt_allele equals 0).
Ps : We wanted to have some clue about strand biaisfor our variants (strand_artifact flag nether appears in MuTect2 VCF), so we requested the SAC to be present in the VCF output. No reads at all in the SAC field are supporting alternate allele on +/- strands.
I know that it's stated in the doc that "the allele counts provided by AD and SAC should not be used to make assumptions about the called genotype", but still, i think it's very strange to have no supporting allele count in both AD & SAC when AF = 11.2%. And given that AF = AD[format] / DP[format], i was expecting some correlation between AF for a given variant and its AD field.
What am I missing ? Did i misread how AF is calculated ?