We've moved!
This site is now read-only. You can find our new documentation site and support forum for posting questions here.
Be sure to read our welcome blog!

Can likelihood be skip?

FatenFaten MalaysiaMember

Dear author,

I wondering if i can skip likelihood step for GATK variant calling? Is there any set of parameters that can be change or do i need to skip certain steps?

Thank you very much.


  • Geraldine_VdAuweraGeraldine_VdAuwera Cambridge, MAMember, Administrator, Broadie admin

    I don't understand your question, sorry. Can you please clarify what you want to skip?

    In general, our recommendations are designed to produce the best possible results within certain practical constraints. If you change or skip any steps you may get suboptimal results.

  • FatenFaten MalaysiaMember

    I want to try to find rare variant for none human genome. If i run using the likelihood, most probably the rare variant couldn't be found. Is it possible way that i can do to obtain the rare variant?

  • Geraldine_VdAuweraGeraldine_VdAuwera Cambridge, MAMember, Administrator, Broadie admin

    Every valid way of calling variants involves likelihoods. Perhaps you mean to refer to the joint calling workflow, which uses GVCFs as an intermediate to store likelihoods of all genotypes at all sites?

    If that is what you mean, the answer is this: that workflow does allow you to identify rare variants. That is one of its main purposes. This is described in our documentation. If you are worried, you can also use the traditional multisample calling workflow, or even call variants directly on your one sample using HaplotypeCaller in its default mode. These are less powerful and less scalable methods, but you can try them and compare results.

Sign In or Register to comment.