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.
Different amount of unfiltered variants when using different PON's
Erik84
Member ❠
Hi,
I'm analyzing the impact of two different panel of normals on the final results in tumor-only mode with Mutect2 4.0.6.0.
I found out that the vcf files on these two analyses differ in the amount of unfiltered variants. I didn't expect this, because the panel of normals is also used in the filter steps later on to indicate variants as non-somatic (FilterMutectCalls). I also checked what happens when I don't use a panel of normals at all. These are the results:
pon A: 53958 variants
pon B: 50768 variants
No pon: 121485 variants
The only difference in these analyses are the panel of normals. Other settings are kept the same.
My question is: why does it seems that the panel of normals has impact on both the amount of unfiltered variants as on the filtersteps in FilterMutectCalls?
Erik
Best Answer
-
shlee
Cambridge ✭✭✭✭✭
Hi @Erik84,
For your ponA and ponB cases to be comparable to your No Pon case, you should use the
--genotype-pon-sitesflag. Mutect2 will pre-filter, i.e. not emit at all, variant sites obviously in the PoN. There are some sites where only upon reassembly of reads can the tool see that the final variants match those in the PoN. In this case, these are emitted to the callset for downstream filtering by FilterMutectCalls. To include all sites in the callset, even if obviously in the PoN, again, you can add the--genotype-pon-sitesparameter to your Mutect2 command.Boolean Call sites in the PoN even though they will ultimately be filtered. Default value: false.
Possible values: {true, false}
I hope this helps to clarify your question.
Answers
Hi @Erik84,
For your ponA and ponB cases to be comparable to your No Pon case, you should use the
--genotype-pon-sitesflag. Mutect2 will pre-filter, i.e. not emit at all, variant sites obviously in the PoN. There are some sites where only upon reassembly of reads can the tool see that the final variants match those in the PoN. In this case, these are emitted to the callset for downstream filtering by FilterMutectCalls. To include all sites in the callset, even if obviously in the PoN, again, you can add the--genotype-pon-sitesparameter to your Mutect2 command.Possible values: {true, false}
I hope this helps to clarify your question.
Hi Shlee,
Thank you very much for your reply. Now I understand what is happening.
Erik