AD counts in VCF file are not matching with AD based on the IGV read counts in the bam file
By comparing variants in the VCF file of two samples, which was generated by the GATK HaplotypeCaller (Version 3.5), I recognized that the AD values are not matching with the IGV read counts in the bam file.
Here is one example:
1 984302 . T C 170.55 . AC=2;AF=0.500;AN=4;DP=13;ExcessHet=3.0103;FS=0.000;MLEAC=3;MLEAF=0.750;MQ=60.00;QD=28.43;SOR=2.303 GT:AD:DP:GQ:PL 0/0:6,0:6:0:0,0,91 1/1:0,6:6:18:199,18,0
The ADs are 6,0 (sample 1) and 0,6 (sample 2) whereas the AD of this variants in sample 1 is 4,2 according to the IGV visualized BAM.
Do you have any idea why this is the case?