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.

Calling Variants with more than one tool necessary?

sp580sp580 GermanyMember


I am currently executing GATK's best practices pipeline. However, I am trying to decide if thereafter I should call variants using other tools, as it is suggested in this 2017's review.

I am working with mouse data (60 samples, ~20x coverage per sample) and most papers use one variant calling tool (GATK or SAM/BCFtools), except for two studies using 3 callers (GATK, SAM/BCFtools, QCALL), but they are relatively "old" (2011 & 2012).

I am looking for some help to make up my mind between the choices:

  • Only GATK
  • GATK plus SAM/BCFtools
  • GATK plus SAM/BCFtools plust Tool3

Since I have only three weeks to finish this part of the project, this decission is critical.

Looking forwards for your suggestions!


  • aazizaaziz Member
    Take my advice with a grain of salt, because I'm an inexperienced PhD student. Also, you've probably already made your decision as you asked this question a few weeks ago.

    I think choosing a tool you're most familiar with, have the most in depth knowledge with, is the most important factor. Learning all 3 tools, the ins and outs, best practices is a lot of work etc is a lot of work, so chances are you will end up half assing 2 of them anyway.

    The other aspect you haven't discussed is how to deal with the output of 3 tools, do you take only the intersecting variants (only those found in all 3 tools)? or do you take all variants? Maybe you accept variants found in 2/3 tools. Are you trying to minimise false positives, reduce false negatives, or increase variant discovery (at the cost of false positives)? etc etc.

    I have seen this approach used rarely, even though it's been recommended in several different reviews of different omics fields.

    Good luck and hope I've helped.
Sign In or Register to comment.