Heads up:
We’re moving the GATK website, docs and forum to a new platform. Read the full story and breakdown of key changes on this blog.
If you happen to see a question you know the answer to, please do chime in and help your fellow community members. We encourage our fourm members to be more involved, jump in and help out your fellow researchers with their questions. GATK forum is a community forum and helping each other with using GATK tools and research is the cornerstone of our success as a genomics research community.We appreciate your help!

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.

Interpreting alt_allele_in_normal filter in Mutect2

darioberdariober Cambridge UKMember

Hi- If I understand correctly Mutect2, the alt_allele_in_normal flag is assigned when variant reads in the normal exceed the --max_alt_alleles_in_normal_count setting (which has default=1).

I wonder if this generates false negatives since you can have only few variant reads in normal out of many total reads. For example, in normal you can have REF:ALT=1000:3 and in tumour REF:ALT=1000:700. These numbers would pass the NLOD and TLOD filters (right?) but the variant would be flagged as alt_allele_in_normal. So I wonder whether one should rescue alt_allele_in_normal variants by, for example, applying a simple fisher test to see if the ALT counts are convincingly larger in tumour vs normal.

Just to make a concrete example, this is a variant that potentially may be rescued. Here tumour REF:ALT=35:25 and normal REF:ALT=51:2:

chr17 39524885 . G A . alt_allele_in_normal ECNT=1;HCNT=1;MAX_ED=.;MIN_ED=.;NLOD=6.35;TLOD=76.43 GT:AD:AF:ALT_F1R2:ALT_F2R1:FOXOG:QSS:REF_F1R2:REF_F2R1 0/1:35,25:0.396:7:18:0.28:1386,995:12:23 0/0:51,2:0.039:2:0:1:2028,82:26:25

(This comment is based on Mutect2 in GATK 3.8.0, I would switch to GATK4 but I see Mutect is still in beta there)

Many thanks


Sign In or Register to comment.