Heads up:
We’re moving the GATK website, docs and forum to a new platform. Read the full story and breakdown of key changes on this blog.
If you happen to see a question you know the answer to, please do chime in and help your fellow community members. We encourage our fourm members to be more involved, jump in and help out your fellow researchers with their questions. GATK forum is a community forum and helping each other with using GATK tools and research is the cornerstone of our success as a genomics research community.We appreciate your help!

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.
We will be out of the office for a Broad Institute event from Dec 10th to Dec 11th 2019. We will be back to monitor the GATK forum on Dec 12th 2019. In the meantime we encourage you to help out other community members with their queries.
Thank you for your patience!

HaplotypeCaller and 1/1:0,0:0:18:383,18,0


We have a single sample which variant is called by HaplotypeCaller. We got a variant which has AD as 0,0, DP 0, however, GT is 1/1. The variant line is:

10 134726068 . TC T 345.80 PASS AC=2;AF=1.00;AN=2;DP=32;FS=0.000;MLEAC=2;MLEAF=1.00;MQ=60.00;MQ0=0;QD=10.81;SOR=0.693;set=variant2 GT:AD:DP:GQ:PL 1/1:0,0:0:18:383,18,0

I read the following link:

Just want to make sure whether this is the reason, because our case is a homozygosity variant. By the way, for clarify, the DP in the INFO field is not filtered (PCR dups, mapping quality, etc.), but the AD and DP in the sample field are all filtered? At the example you mentioned at the page, Both DP in INFO and SAMPLE field are 10. Is that the DP in SAMPLE field should be 0?



  • ying_sheng_1ying_sheng_1 Member ✭✭

    The version of GATK we used is v3.3.

  • SheilaSheila Broad InstituteMember, Broadie admin


    Can you please post IGV screenshots of the original bam file and bamout file at the position? Also, please try running HaplotypeCaller with the latest version on just that region. This may be a bug that might have been fixed in the latest version.


  • slamentslament SwedenMember

    Hi! I think I'm having a similar problem. I find a few sites with both AD= 0,0 and DP=0, but still with a genotype called. That doesn't make any sense to me. I'm using GATK v.3.5.0 in a non-model haploid organism. I used HaplotypeCaller and then GenotypeGVCF.

    An example of a site:

    chromosome_3    371180  .   A   G   1766.72 PASS    AC=5;AF=0.147;AN=34;BaseQRankSum=-0.013;ClippingRankSum=-0.445;DP=2097;FS=15.456;MLEAC=5;MLEAF=0.147;MQ=41.5;MQRankSum=0.419;QD=24.54;ReadPosRankSum=0.801;SOR=0.807    GT:AD:DP:GQ:PL  0:70,0:70:99:0,1800 0:71,0:71:99:0,1800 0:14,0:14:99:0,270  0:67,0:67:99:0,1310 0:64,0:64:99:0,713  1:0,2:2:99:135,0    1:0,0:0:45:45,0 0:45,0:45:99:0,1507 0:57,0:57:99:0,1268 0:50,0:50:99:0,1252 0:71,0:71:99:0,1482 0:49,0:49:99:0,1582 0:81,0:81:99:0,1800 0:50,0:50:99:0,1130 0:61,0:61:99:0,1598 0:68,0:68:99:0,1220 0:52,0:52:99:0,1254 0:75,0:75:99:0,1308 0:74,0:74:99:0,1800 0:64,0:64:99:0,1464 0:49,0:49:99:0,973  1:0,21:21:99:990,0  1:18,24:42:99:360,0 0:62,0:62:99:0,1616 0:83,0:83:99:0,1800 0:54,0:54:99:0,1035 0:63,0:63:99:0,1693 0:53,0:53:99:0,1414 0:34,0:34:99:0,881  0:67,0:67:99:0,1800 0:67,0:67:99:0,1800 0:62,0:62:99:0,1800 1:0,7:7:99:315,0    0:65,0:65:99:0,1708

    Notice that one sample (the 7th) has 1:0,0:0:45:45,0.

    I looked at the site in IGV and it's super nasty (attached, centered). How is it possible that such site has a PL of 45??

    Another example:

    chromosome_5    1066363 .   C   T,* 7125.5  PASS    AC=2,2;AF=0.059,0.059;AN=34;BaseQRankSum=0.647;ClippingRankSum=0.903;DP=3826;FS=1.872;MLEAC=2,2;MLEAF=0.059,0.059;MQ=49.98;MQRankSum=1.43;QD=28.14;ReadPosRankSum=0.866;SOR=0.318   GT:AD:DP:GQ:PL  0:86,0,0:86:99:0,102,102    0:103,0,0:103:99:0,1800,1800    0:107,0,0:107:99:0,1800,1800    0:92,0,0:92:99:0,1693,1693  0:130,0,0:130:99:0,1800,1800    1:0,84,0:84:99:3955,0,3955  1:5,82,0:87:99:2881,0,3139  2:0,0,0:0:90:90,90,0    2:0,0,4:4:99:270,270,0  0:100,0,0:100:99:0,1800,1800    0:121,0,0:121:99:0,1800,1800    0:121,0,0:121:99:0,1800,1800    0:102,0,0:102:99:0,1800,1800    0:130,0,0:130:99:0,1800,1800    0:117,0,0:117:99:0,1800,1800    0:132,0,0:132:99:0,1800,1800    0:127,0,0:127:99:0,1800,1800    0:143,0,0:143:99:0,1800,1800    0:193,0,0:193:99:0,1800,180 0:103,0,0:103:99:0,1800,1800    0:92,0,0:92:99:0,1800,1800  0:90,0,0:90:99:0,1800,1800  0:131,0,0:131:99:0,1800,1800    0:135,0,0:135:99:0,1800,1800    0:160,0,0:160:99:0,1800,1800    0:138,0,0:138:99:0,1800,1800    0:103,0,0:103:99:0,1800,1800    0:107,0,0:107:99:0,1800,1800    0:79,0,0:79:99:0,1800,1800  0:141,0,0:141:99:0,1800,1800    0:111,0,0:111:99:0,1800,1800    0:87,0,0:87:99:0,424,424    0:138,0,0:138:99:0,1800,1800    0:104,0,0:104:99:0,1800,1800

    The 8th sample is 2:0,0,0:0:90:90,90,0.

    Thanks in advance!

  • SheilaSheila Broad InstituteMember, Broadie admin


    Can you please post the bamout and GVCF record for that sample at those positions?


  • slamentslament SwedenMember

    Hi again Sheila! Thanks for your answers. I got the bamout for the first site and indeed, it looks SO much better than the original BAM, which is expected of course but I forgot about this... (thank you for remind me!). I cannot attach the bamout itself...(I get an error "Uploaded file type is not allowed").

    Here is the gvcf for that site:

    chromosome_3    371180  .   A   G,<NON_REF> 15.14   .   DP=64;MLEAC=1,0;MLEAF=1.00,0.00;MQ=38.52    GT:AD:DP:GQ:PL:SB   1:0,0,0:0:45:45,0,45:0,0,0,0

    I attached an IGV figure comparing the original BAM with the bamout, and I see that the offending site (centred) has two sequences with the annotation "ArtificialHaplotype", each with a different allele (REF A, and ALT G) but no extra reads for that position. I guess that AD is 0,0,0 because the (filtered out) reads in the site are terrible anyway so they don't count? But why DP is 0 as well? Where are the reads supporting the artificial haplotypes?

    Thank youuuu!

  • slamentslament SwedenMember

    Here it is for the second example, the gvcf:

    chromosome_5    1066362 .   ACG A,<NON_REF> 50.97   .   DP=49;MLEAC=1,0;MLEAF=1.00,0.00;MQ=56.30    GT:AD:DP:GQ:PL:SB   1:0,0,0:0:90:90,0,90:0,0,0,0

    And the attached screenshot of IGV. In this case the read count is 32 C (REF) and 19 indel (ALT) for the site 1066363.

    Sorry for the many questions and thanks again!

    Issue · Github
    by Sheila

    Issue Number
    Last Updated
    Closed By
  • SheilaSheila Broad InstituteMember, Broadie admin


    I am looking into a similar case. I will get back to you soon with some insights.


  • slamentslament SwedenMember
  • SheilaSheila Broad InstituteMember, Broadie admin


    There is a new feature in the nightly build that outputs the filtered reads. Can you try outputting the bamout using the latest nightly build and post it here?

    Also, try adding --forceActive and --disableOptimizations and --dontTrimActiveRegions to your command when you run on the interval. This will force all the sites to be tagged as active and be output to the bamout file.


  • slamentslament SwedenMember

    Hi @Sheila! I'm answering this terribly late, so sorry! Exactly what nightly build I should use? And what is the feature that you recommend me to try?

    Thanks and sorry again for taking so long!

Sign In or Register to comment.