Notice:
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 bamout to read post-realignment coverage including non-variants

Hi,

I want to output a bamout to see the post-realignment coverage (a la -A DepthPerAlleleBySample) for a position, regardless of whether that site is a variant site. I'm sure I'm need to turn on -disableOptimizations and --emitDroppedReads, but should I also turn on --forceActive and --dontTrimActiveRegions? In addition, would these flags affect the results of the variant calling itself?

Thanks for the help!

Tagged:

Best Answers

Answers

  • valentinvalentin Cambridge, MAMember, Dev ✭✭

    I think that --dontTrimActiveRegions may well change the results like the PL/QUAL values somewhat however only in rare cases it will change GT call or allele list would be rare though.

  • johnmajohnma Member

    But I thought AD counts unfiltered reads as well. Aren't informative, but filtered, reads not output to standard bamout?

    In addition, I have looked at the source code, and noticed the possible reasons for being tagged as dropped include Trimmed, Downsampled, Failed realignment, Filtered, and Not informative. I suppose only Failed realignment and Not informative reads won't be counted into allele, right?

  • valentinvalentin Cambridge, MAMember, Dev ✭✭

    For a read to count towards AD at a site, it must overlap the variant after realignment and it must have exhibit some difference in likelihoods between the top two alleles (the so called "informative" reads).

    I think the user may ask for with some option, which I don't remember right now, to use filtered out (and perhaps trimmed?) reads for calculating annotations (but never for PL/GQ or QUAL), but the way they are re-added they won't be considered informative anyway.

    Down sampled reads won't ever be considered for AD or another annotation and I reckon that they will be displayed with the original alignment.

  • johnmajohnma Member

    @valentin
    I take that to mean -disableOptimizations --forceActive --dontTrimActiveRegions should emit all realigned reads that are used to count AD, including in non-variant regions. Is that right?

Sign In or Register to comment.