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!

Doesn't Base Quality Score Recalibration degrade sensitivity in heavily mutated cancers?

From the GATK FAQs - "The base quality score recalibrator treats every reference mismatch as indicative of machine error"... which is why we have to give it a list of dbSNPs to skip over. But what about somatic SNPs? Wouldn't hypermutated tumors from uterine, colorectal, melanoma, or lung cancers be re-calibrated to a lower quality than data from AML or breast. And variant caller sensitivity would drop accordingly.

Should we ideally do some high-confidence SNP calling on un-calibrated BAMs, and append those to the dbSNP VCF for the BQSR?


Best Answer


  • EugenieEugenie Member

    I have same question.

    In BQSR documentation it was mentioned that even for cancer samples the difference from ref. is not so high to spoil the recalibrationand and Mutect paper suggests to go through all best practice steps including BQSR.

    But I wonder if it's applicable to highly mutated tumors: if there is a risk to underestimate the quality of somatic mutation calls by this way?
    For melanoma for example, with over 10 SSNVs/Mb...

    Thank you!

  • Hi

    I know this post is 4 years-old but is there an update as to how BQSR behaves with somatic datasets, and whether there is a risk that BQSR can lower the sensitivity of somatic variant calling, due to the fact that it can lower the base qualities of real somatic variants?


  • SheilaSheila Broad InstituteMember, Broadie ✭✭✭✭✭

    Hi Severine,

    Perhaps this thread and this thread which are more recent will help.


Sign In or Register to comment.