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.
Attention:
We will be out of the office on November 11th and 13th 2019, due to the U.S. holiday(Veteran's day) and due to a team event(Nov 13th). We will return to monitoring the GATK forum on November 12th and 14th respectively. Thank you for your patience.

Is there a tool the can convert all-sites gvcf to banded gvcf?

I have create 'all-sites' gvcf for a large number of genomes.
But I discovered that for one analysis I want to apply the input must be banded gvcf.
Is there a tool within the gatk suite or elsewhere that can achieve this conversion?

Thanks for any help.
Cheers,
Chuck

Best Answer

Answers

  • SheilaSheila Broad InstituteMember, Broadie admin

    @chlangley
    Hi Chuck,

    Hmmm. I don't think there is a tool to do that. What analysis are you doing that needs a banded GVCF?

    -Sheila

  • chlangleychlangley UCDMember

    Sheila:

    Thanks for the quick responses.

    I had issues running snpEff on the unbanded *.g.vcf s.
    Pablo C. suggested creating banded versions.
    I am now running a couple examples to see if the issues
    are resolved by his suggestion.

    But if that were so I will then be faced with
    rerunning several hundred genomes (> 1 month of my computing
    resources). So i was looking for a way to convert the already created
    g.vcf s to banded.

    BTW: The reason to create the unbanded g.vcf s was to preserve
    the details the go into the quality assessment so that I might explore
    potential dataset-specific quality assessment.

    Cheers,
    Chuck

  • Geraldine_VdAuweraGeraldine_VdAuwera Cambridge, MAMember, Administrator, Broadie admin

    Hi Chuck, why are you running snpeff on the GVCFs? You should really finish the genotyping + filtering before proceeding with post-processing and downstream analyses.

  • chlangleychlangley UCDMember

    Hello Geraldine:

    A fair question I'll try to address.

    Our data are from large samples of deeply sequenced haploid Drosophila melanogaster genomes. The most important research questions to be addressed with these data concern the theoretical models of the processes that shape genomic polymorphism and divergence. So the invariant sites are critical and we must strive to have an robustly comparable quality for all calls. The robustness and completeness of the best practices pipeline through the GVCF is a great foundation. The frequency spectra of SNP and indels are critical in many downstream analyses. As the GenotypeGVCF documentation discusses the results (especially the numbers of rare variants) can vary with assumed prior spectrum. I would be inclined to explores this functionality in GenotypeGVCF (--input_prior) if haploidy were supported (curious why it is not). But because of the squirrelly nature of variant ascertainment bias, as well as lingering concerns about reference bias (not to mention the strong impact of demographic history), I am inclined to keep the default as genotyping on the single genome basis and then (exploring and) applying filters to meet the specifics and rigor of each question.

    In a similar spirit, it seems appropriate to get the snpEff annotation into the gvcf so the user could incorporate functional considerations at the genotyping and filtering stage (at least in exploring/disproving possible interactions between functional effects and genotyping biases). For the time being the file format is (g)vcf and that format strongly shapes how population genomics can be pursued.

    Cheers,
    Chuck

Sign In or Register to comment.