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.

Amplicon-Seq Edge Effects in MuTect2


I've been mostly successful in adapting MuTect2 for use with amplicon-based panels. The one issue I continue to run in to is at the edges of the regions that we are targeting. I've done a great deal of parameter manipulation to try and get this to work properly, but no luck much of the time.

In the attached bamout shot, you can see the strong evidence for the variant, while the reassembly stage seems to have another idea. This particular example is a little farther from the end than most of the examples I could dig up.

I've seen bits and pieces of this same issue by looking through the forums, and have tried to employ any methods that have been discussed. As another idea, might it be worth creating synthetic reference-based sequences that overlap amplicon ends and trying to call variants with those spiked in to the input BAM file? If you have a good reason to think this won't work, I wouldn't want to sink a bunch of time in to it.

Thanks for any ideas, much appreciated!


Best Answer


  • SheilaSheila Broad InstituteMember, Broadie admin

    Hi John,

    Are the blue reads the reads from your sample, or are they the Artificial Haplotypes?


  • pachewychomppachewychomp OregonMember

    The blue one's are my reads, and that variant at 7579801 has been vetted.

    How does manipulating interval_padding affect haplotype determination?


  • pachewychomppachewychomp OregonMember

    To extend the previous comment, I have noticed a couple of situations where calling with a list of let's say 5000 intervals, and then calling with a subset of those intervals, like 5, yield different results. This is taking every parameter to be identical other than the interval lists that are fed to the -L option. Could you also comment on potential reasons for this behavior?


Sign In or Register to comment.