REQUIRED for all errors and issues:
a) GATK version used: 4.3.0.0
b) Exact command used: shell('{config[gatk][path]} --java-options "-Xmx90g" HaplotypeCaller --disable-read-filter GoodCigarReadFilter --disable-read-filter PassesVendorQualityCheckReadFilter --disable-read-filter MappingQualityAvailableReadFilter --disable-read-filter NotDuplicateReadFilter --disable-read-filter MappingQualityReadFilter --disable-read-filter NotSecondaryAlignmentReadFilter -I {sIn} -O {output} -R {config[samtools][reference]}  2> {log}')

c) Entire program log: N/A

Hi everyone,

It's one of my first times messing around with the Haplotypecaller, and I need to dig into a few things to figure out why some reads are not counted or not.

The issue is I have a variant called (actually tons of them) from a RNAseq sample for HG001. Many of them are not in the WGS for this sample. While having a closer look, it seems that most of them are called because secondary alignments are not counted.

My example: One variant has in total 36 reads, of which 7 are primary, 29 are secondary. Within the 7 primary are 3 reads with SNPs, therefore a SNP gets called (which is not in the WGS). In the 29 secondary there are no SNPs. With the command outlined above I have now managed to get the secondary alignments called, the VCF file shows a DP of 36, which is correct. The SNP is still called, based on only the 7 reads. For testing, I'd like to have it called on all the reads. I thought I'd have everything relevant deactivated with my command above. But why is it not counting the 36 reads for the SNP calling?

Relevant VCF entry:

1       14522   .       G       A       77.64   .       AC=1;AF=0.500;AN=2;BaseQRankSum=2.189;DP=36;ExcessHet=3.0103;FS=0.000;MLEAC=1;MLEAF=0.500;MQ=26.47;MQRankSum=0.000;QD=11.09;ReadPosRankSum=0.000;SOR=0.495      GT:AD:DP:GQ:PL  0/1:4,3:7:85:85,0,88

Any insight would be appreciated.

Thanks,

B