Genome Analysis Toolkit

Variant Discovery in High-Throughput Sequencing Data

GATK process banner

Need Help?

Search our documentation

Community Forum

Hi, How can we help?

Developed in the Data Sciences Platform at the Broad Institute, the toolkit offers a wide variety of tools with a primary focus on variant discovery and genotyping. Its powerful processing engine and high-performance computing features make it capable of taking on projects of any size. Learn more

DRAGEN-GATK Update: Let's get more specific Follow

6 comments

  • Avatar
    Kurt Hetrick

    Is this single sample calling only? Is there a 2ish step process still involved (creating a gvcf like file) and then joint call?

    0
    Comment actions Permalink
  • Avatar
    Geraldine Van der Auwera

    Hi Kurt, this applies to both the single sample case and the multisample (joint calling) cases. Here we show the evaluation results from the single sample case because they're more straightforward to interpret, but the modifications we describe will benefit joint calling as well. So if you're calling variants on cohorts, you will still do the 2-step process of creating a GVCF for each sample separately then joint calling all your samples together. 

    0
    Comment actions Permalink
  • Avatar
    MICHAEL MCMANUS

    Is there a plan for improving and merging Mutect for both the Dragen accelerated and non-accelerated pipelines? 

    1
    Comment actions Permalink
  • Avatar
    Shahryar Alavi

    What about duplicated genes?

    To my knowledge, GATK does not call variants if reads are mapped to more than one locus on the genome; e.g. we never see variants of SMN1 and SMN2 genes in our annotated variants list.

    Does the new DRAGEN-GATK pipeline have a solution to this challenge?

    1
    Comment actions Permalink
  • Avatar
    Geraldine Van der Auwera

    MICHAEL MCMANUS Mutect2 and the somatic short variants pipeline are on the list of use cases we want to work on together, but we haven't yet decided which will be next after the germline short variants. We'd love to hear from you all on what would be most valuable to the research community, so don't hesitate to comment. 

    Shahryar Alavi You're correct that GATK will only take into account uniquely mapped reads; dealing with duplicated genes is more a matter of mapping strategy. The updated DRAGEN-GATK pipeline does not change anything compared to the current behavior in that respect.

    2
    Comment actions Permalink
  • Avatar
    Geraldine Van der Auwera

    FYI our colleagues in the Genomics Platform (who do the sequencing at the Broad Institute) are presenting a poster at the AGBT meeting this week about their testing of the DRAGEN-GATK pipeline. For the lucky ducks over there, it's poster 1010 and will be presented on Tuesday Feb 25th from 4:40 pm to 6:10pm. For the rest of the world, here is the PDF: https://drive.google.com/open?id=1fa0RG9RjoPOBlqvPoH6CXL121NldqAph

    0
    Comment actions Permalink

Please sign in to leave a comment.

Powered by Zendesk