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.

again about the AD field

BogdanBogdan Palo Alto, CAMember ✭✭

Dear Sheila, and Geraldine,

also please could you advise on understanding the data in the VCF files of NORMAL and TUMOR for the variant at this position (from MUTECT2) : chr21 46689620 : especially the FS, SOR, and the AD.

chr21 46689620 . G A . PASS BaseCounts=7,0,27,0;ECNT=1;FS=0.000;HCNT=1;HRun=4;MAX_ED=.;MIN_ED=.;MQ=33.87;NLOD=6.16;SOR=2.636;TLOD=10.90;VariantType=SNP GT:AD:AF:ALT_F1R2:ALT_F2R1:FOXOG:QSS:REF_F1R2:REF_F2R1 0/1:20,6:0.240:5:1:0.833:408,145:8:12 0/0:21,0:0.00:0:0:.:435,0:11:10

I have also attached the images from samtools tview - for NORMAL and TUMOR : will describe the questions and the calculations below ;)

Answers

  • BogdanBogdan Palo Alto, CAMember ✭✭

    now about the computations for the G to A substitution :smile:

    for the alternative allele - in NORMAL (1st image) : AD=21,0 : that looks correct;
    for the alternative allele - in TUMOR (2nd image) : AD= 20,6, although after the visual count it shall be 30,7.

    why MUTECT reports 20,6 instead of 30,7. Then, another question is : A nucleotide at the position chr21:46689620 seems to be highly strand-biased, but the FS calculation is 0 ? why ? Then the SOR value is 2.6, although this number does not fit the formula described at :smile:

    https://www.broadinstitute.org/gatk/guide/tooldocs/org_broadinstitute_gatk_tools_walkers_annotator_StrandOddsRatio.php

    Am I misunderstanding the fields and is there anything else missing ? thanks again !

  • BogdanBogdan Palo Alto, CAMember ✭✭

    Regarding the AD and DP, I understand that AD includes the NON-FILTERED COUNTS, and DP includes the FILTERED COUNTS,
    I am just surprised that FS calculation is 0, although there is an obvious strand bias, Also how the SOR is computed to 2.6 ? thanks !

Sign In or Register to comment.