What is the difference between ApplyRecalibration and ApplyVQSR

I am trying to do quality control for germline SNP & indels in 12 cancer patients. I have completed upto the VariantRecalibrator step. Now I am trying to figure out which tool do I use next ApplyRecalibration or ApplyVQSR. I looked at the documentation of both these functions and they seem to do the same thing. I am looking at this pipeline. Here ApplyVQSR has been used and ApplyRecalibrator function is missing. But in the corresponding documentation ApplyRecalibration has been discussed and ApplyVQSR is missing. I am confused, can somebody explain which one should I be using fpr characterizing germline SNP and indels.

Secondly, I noticed that in the wdl file referenced above, the vcf which is output from the indel mode of ApplyVQSR is being fed into ApplyVQSR snp mode (search for the word " tmp.indel.recalibrated.vcf" in the wdl file). But, I thought snp and indels are evaluated separately. Is this correct?

Thank you.



  • SheilaSheila Broad InstituteMember, Broadie, Moderator admin

    Hi Robin,

    Sorry for the confusion. ApplyRecalibration is the GATK3 tool and ApplyVQSR is the GATK4 tool for use after VariantRecalibrator. I think the point is to make it easier to distinguish between ApplyBQSR and ApplyVQSR in GATK4 :smile:

    For the ApplyVQSR step in the WDL, I think VariantRecalibrator is run twice (once for SNPs and once for INDELs), then ApplyVQSR is run twice (one for SNPs and once for INDELs). I hope that makes sense.


Sign In or Register to comment.