Attention: Want an end-to-end pipelining solution for GATK Best Practices?
A Sanger verified variant is not called by HaplotypCaller
We have a trio which all three members were whole exome sequenced. We processed the samples by the standard pipeline (bwa mem alignment, picard mark duplicates and gatk joint variant calling by HaplotypeCaller). The version of GATK we used is v3.8.1. We can see there is a variant both from mother and the proband from IGV (see here) and the variant is Sanger verified both from proband and the mother. The position is well covered in both proband bam(94x from vcf FORMAT) and mother bam (207x from vcf FORMAT). However, in the variant file, the variant is called as de novo, so there is no variant called from mother.
17 71189251 . C G 1581.68 . AC=1;AF=0.167;AN=6;BaseQRankSum=1.59;ClippingRankSum=0.00;DP=343;ExcessHet=3.0103;FS=2.970;MLEAC=1;MLEAF=0.167;MQ=60.00;MQRankSum=0.00;QD=16.83;ReadPosRankSum=1.27;SOR=1.095 GT:AD:DP:GQ:PGT:PID:PL 0/1:46,48:94:99:0|1:71189182_A_G:1610,0,1851 0/0:41,0:41:99:.:.:0,99,1485 0/0:207,0:207:0:.:.:0,0,117
I have both bam file and bamoutput file from HaplotypeCaller sliced on this variant. I could upload if you tell me where to upload.
Could you help us to check what is the reason for the caller to miss this variant? Thanks.