Holiday Notice:
The Frontline Support team will be offline December 17-18 due to an institute-wide retreat and offline December 22- January 1, while the institute is closed. Thank you for your patience during these next few weeks as we get to all of your questions. Happy Holidays!

GT, PL and GQ inconsistent

hangdhangd Baylor College of medicineMember

From the GATK documentation I found that for PL field, the most likely genotype (assigned in the GT field) is 0. However, from the vcf file called by GATK, I found that this is not always the case. Below are a few examples:

GT:AD:DP:GQ:PL
0/1:0,232:444:99:0,382,4800
1/1:1,0:1:2:1,0,0
1/1:1,0:1:99:0,3,35
1/1:1,0:1:87:0,292,247

Also, GQ is equal to the second smallest PL, unless that PL is greater than 99. However, I also found cases like the following, where the GQ is 1.76:

GT:AD:DP:GQ:PL
0/1:9,14,23:1.76:30,3,0

I am wondering that whether they have specific meanings or it is due to calling error, or due to the low quality of calling. Thank you very much.

Answers

  • SheilaSheila Broad InstituteMember, Broadie, Moderator admin

    @hangd
    Hi,

    Can you tell us the exact command you ran to generate the VCF and the version of GATK you are using?

    Thanks,
    Sheila

  • hangdhangd Baylor College of medicineMember

    Hi, thank you very much. These variants were not called by me, and I know very little about GATK, so I attached the meta info in vcf files.

    For the following calling, GATK 2.0 was used. The meta information is in vcfheader1.txt.
    GT:AD:DP:GQ:PL
    0/1:0,232:444:99:0,382,4800
    1/1:1,0:1:2:1,0,0
    1/1:1,0:1:99:0,3,35
    1/1:1,0:1:87:0,292,247

    For the following calling, I am not sure which version of GATK was used. The meta information is in vcfheader2.txt.
    GT:AD:DP:GQ:PL
    0/1:9,14,23:1.76:30,3,0

    Thank you again.

  • Geraldine_VdAuweraGeraldine_VdAuwera Cambridge, MAMember, Administrator, Broadie admin

    These all look like errors. The calling was done with an old tool (UnifiedGenotyper) in a very old version, so I'm not surprised there would be some problematic sites. If you have the opportunity to go back to the read data and re-call all of these that's what I would recommend.

Sign In or Register to comment.