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.

Outputting REF calls in Mutect

Jeff_GaitherJeff_Gaither Math Biosciences Institute, Columbus, OHMember
edited May 2017 in Ask the GATK team

Hi. I wonder if there's any way to output REF calls with MuTect, so that I can distinguish them from no-calls? What I'd like to obtain is a confidence-score for ref calls, i.e.
GT:PL 0/0:0,255,255

Here's the background. My goal is to apply MuTect (1.1.7) to a bunch of different samples and combine the outputs site-by-site. However, if MuTect doesn't call a site in a given sample, I don't see any way of telling if that was because
1. MuTect thinks the site is REF
2. the quality at the site is too poor for MuTect to tell, one way or the other.

I understand that it would be impracticable for MuTect to output every single REF call, but perhaps it could so if I gave it a bedfile of specific sites to look at. Or maybe another tool like HaplotypeCaller/mpile would be better suited for this? Or maybe there's some functionality in MuTect that I'm completely missing?

Thanks for any help you can offer.

Issue · Github
by Sheila

Issue Number
Last Updated
Closed By

Best Answers


  • Jeff_GaitherJeff_Gaither Math Biosciences Institute, Columbus, OHMember

    Thanks for the update, Sheila. It's helpful to know that this is impossible in MuTect1.

  • Jeff_GaitherJeff_Gaither Math Biosciences Institute, Columbus, OHMember

    Thanks, Sheila. It's nice to know that we'll have this capability before too long. And as you suggest, HaplotypeCaller*O(1) is totally feasible for individual sites.

Sign In or Register to comment.