I am calling variants (SNPs and Indels) on WGS data. After JointGenotyping with g.vcf files I noticed that for some of the variants in the VCF file have a missing variant call (./.) despite having reads.
1.1 271087 . A T 6094.35 . AC=6;AF=1.00;AN=6;DP=190;ExcessHet=3.0103;FS=0.000;MLEAC=6;MLEAF=1.00;MQ=49.13;QD=28.12;SOR=0.820 GT:AD:DP:GQ:PGT:PID:PL ./.:50,0:50 1/1:0,46:46:99:1|1:271079_A_T:2205,147,0 1/1:0,29:29:87:1|1:271051_A_C:1305,87,0 1/1:0,57:57:99:1|1:271079_A_T:2610,175,0
1.1 271089 . A T 6004.35 . AC=6;AF=1.00;AN=6;DP=189;ExcessHet=3.0103;FS=0.000;MLEAC=6;MLEAF=1.00;MQ=49.07;QD=33.97;SOR=0.873 GT:AD:DP:GQ:PGT:PID:PL ./.:50,0:50 1/1:0,46:46:99:1|1:271079_A_T:2160,144,0 1/1:0,29:29:87:1|1:271051_A_C:1305,87,0 1/1:0,56:56:99:1|1:271079_A_T:2565,172,0
1.1 271091 . G A 5914.35 . AC=6;AF=1.00;AN=6;DP=188;ExcessHet=3.0103;FS=0.000;MLEAC=6;MLEAF=1.00;MQ=49.20;QD=33.36;SOR=0.908 GT:AD:DP:GQ:PGT:PID:PL ./.:49,0:49 1/1:0,47:47:99:1|1:271079_A_T:2115,141,0 1/1:0,28:28:87:1|1:271051_A_C:1305,87,0 1/1:0,56:56:99:1|1:271079_A_T:2520,169,0
When you view the region of interest (Chrom 1 270713-271756) in IGV, the original alignment (Top half of attached file) indicates that there are SNPs present, but the -bamout created from the HaplotypeCaller step seems to discard the reads (Bottom half of attached file). It only seems to do this for the one sample, the -bamout for the other 3 samples does not discard the reads and calls SNPs as indicated (See above vcf).
My question is, why is the re-alignment not consistent among the samples, and is there a way to adjust the GATK parameters so that it is not discarding reads? The subspecies I am working with are expected to have many variable sites compared to the reference genome and I do not want to discard SNPs of interest.