Hello,
When a read has mutli-hits on a reference genome, the bam file would record mutli positions in XA tag. So, would GATK consider the XA tag when calling variations?
After getting filtered VCF using gatk, I found that some sites were covered by many reads (for example, more than ten). But when I check the bam file in the IGV viewer, these sites didn't have any reads but the consensus sequence was intact. So, I wonder whether GATK also count the reads in the XA tag when calling variations?
Thanks a lot！