GenotypeGVCFs: unexpected output in PGT and PID fields

Follow

Johnny Li

• March 31, 2020 21:35
• Edited

Can you please provide
a) GATK version used
b) Exact GATK commands used
c) The entire error log if applicable.

4.0.11.0

b) Exact GATK commands used

I'm running this as a Databricks job, and I'm not sure how to find the exact command. I'm essentially running their standard Joint genotyping pipeline on Azure. 

c) The entire error log if applicable.

The pipeline runs without error. But the output VCF from joint genotyping contains special characters which result in an incorrectly formatted VCF. Here's an example:

chr1 101389 . A C 861.80 . AC=22;AF=0.14666666666666667;AN=150;BaseQRankSum=0.027;ClippingRankSum=0.175;DP=1999;ExcessHet=7.3032;FS=31.074;InbreedingCoeff=-0.1071;MLEAC=22;MLEAF=0.14666666666666667;MQ=8.01;MQ0=0;MQRankSum=0.198;QD=1.95;RAW_MQ=28330.0;ReadPosRankSum=0.899;SOR=7.172 GT:AB:AD:DP:GQ:PGT:PID:PL 0/0:.:41,4:45:6: ^C^@:^@^@^@^@^@^@^@^@^@^@:0,6,1210
0/0:.:40,0:40:51:.:.:0,51,1232 0/1:0.860:36,6:42:62:0|1:101389_A_C:62,0,1136 0/1:0.890:17,2:19:9:.:.:9,0,517 0/0:.:34,0:34:99:.:.:0,99,1485 0/0:.:36,0:36:99:.:.:0,99,1211

These kinds of insertions occur in about 5-10% of lines, and always affect the PGT and PID fields. I've checked the original input VCFs (produced by HaplotypeCaller), and they appear normal. Here's the corresponding line for the first individual above: 

chr1 101389 . A C,<NON_REF> 0 . BaseQRankSum=-0.140;ClippingRankSum=-0.419;DP=45;ExcessHet=3.0103;MLEAC=0,0;MLEAF=0.00,0.00;MQ0=0;MQRankSum=0.578;RAW_MQ=45721.00;ReadPosRankSum=1.177 GT:AD:DP:GQ:PGT:PID:PL:SB 0/0:41,4,0:45:6:0|1:101389_A_C:0,6,1210,123,1222,1339:21,20,4,0

Any ideas what's going on? 

Thanks!

0

• 

  Bhanu Gandham

  • April 01, 2020 20:25

  Hi Johnny Li

   

  You seem to be running a pretty old version of GATK. Can you please try and run this on a more recent version of GATK and let us know if the issue persists? Here is the latest version: https://github.com/broadinstitute/gatk/releases/tag/4.1.6.0

   

  Also we do not support pipelines run on databricks. And without the exact commands used it is difficult to figure out what is casing the issue.

  0

  Comment actions Permalink
• 

  Johnny Li

  • April 02, 2020 03:36

  Thanks for your reply, Bhanu. Upgrading seemed to do the trick! 

  0

  Comment actions Permalink
• 

  Bhanu Gandham

  • April 02, 2020 16:57

  Glad it worked! :)

  0

  Comment actions Permalink

Please sign in to leave a comment.