To celebrate the release of GATK 4.0, we are giving away free credits for running the GATK4 Best Practices pipelines in FireCloud, our secure online analysis portal. It’s first come first serve, so sign up now to claim your free credits worth $250. Sponsored by Google Cloud. Learn more at https://software.broadinstitute.org/firecloud/documentation/freecredits

Truth or control samples - Variant calling

Are we able to incorporate truth/control samples in addition to dbSNP when calling variants with GVCF (cohorts) or the traditional way with HaplotypeCaller. There are for example situations where the sequences are for Australian, E Asian, or African samples, and we would like to include truth/control samples for those areas, perhaps from 1000 genomes or some other source.

If possible, what arguments do we use.

Best Answer

  • Geraldine_VdAuweraGeraldine_VdAuwera Cambridge, MAMember, Administrator, Broadie
    Accepted Answer

    Ah, then sure, you can substitute whatever resources you feel are appropriate for your use case. The resources we recommend are for modern-day humans and may not be appropriate for ancient DNA. You can use any number of resources as long as you provide truth and training data (which can be the same resource or not depending on what you have available).

Answers

  • SheilaSheila Broad InstituteMember, Broadie, Moderator

    @dilawerkh4
    Hi,

    I don't think there is a way to do what you are asking. What information are you hoping to add with the extra samples?

    -Sheila

  • @Sheila said:
    @dilawerkh4
    Hi,

    I don't think there is a way to do what you are asking. What information are you hoping to add with the extra samples?

    -Sheila

    Sheila,

    I was thinking that perhaps by using truth or control samples more relevant to the test samples being processed may be better than using the generic HapMap or 1000G in the resource bundle. For example, if we are processing E Asian or Siberian aDNA, then why not just use E Asian or Siberian samples WGS samples from the Simons Diversity Project as truth samples, instead of generic HapMap and 1000G which is in part based on some irrelevant samples from Africa or Europe. Wouldn't that increase confidence in variant calls?

    Also, is it a bad idea to only use 1 truth resource instead of the 2 GATK recommends.

    Dilawer

  • Geraldine_VdAuweraGeraldine_VdAuwera Cambridge, MAMember, Administrator, Broadie

    HaplotypeCaller does not use any truth data internally. If you provide a truth file, all it does with those sites is record their IDs if you call any variants at the same sites. This is so the calling is as unbiased as possible. Afterward at the filtering step is where you would apply truth data to improve your callsetand adjust your confidence in your variants.

  • @Geraldine_VdAuwera said:
    HaplotypeCaller does not use any truth data internally. If you provide a truth file, all it does with those sites is record their IDs if you call any variants at the same sites. This is so the calling is as unbiased as possible. Afterward at the filtering step is where you would apply truth data to improve your callsetand adjust your confidence in your variants.

    Sorry Geraldine, I meant VQSR not HaplotypeCaller

    Dilawer

  • Geraldine_VdAuweraGeraldine_VdAuwera Cambridge, MAMember, Administrator, Broadie
    Accepted Answer

    Ah, then sure, you can substitute whatever resources you feel are appropriate for your use case. The resources we recommend are for modern-day humans and may not be appropriate for ancient DNA. You can use any number of resources as long as you provide truth and training data (which can be the same resource or not depending on what you have available).

Sign In or Register to comment.