I've been using a pipeline including novoalign and GATK best practices for alignment, variant calling, joint genotyping, vqsr, etc. I recently gained access to a DRAGEN box and would love to switch over to it because it is much faster. I am looking to do some quality control. I was thinking I would process a genome cohort with each pipeline to get g.vcfs. From these, I would look at all variants and see if the same variants are being called by both pipelines. Also would look at allele depths to try to identify where reads were mapped differently. Option 2 would certainly be to do some comparison of bam files but I thought this may be more straightforward. I am wondering if (1) anyone has ideas on a good way to do this comparison? or (2) if anyone knows of an existent tool that will provide me with these stats from either g.vcf or bam files?