I would like to run my GATK snp calling on two bam files as if they were one biological sample.
There are a two recommended ways of doing this (mentioned here), and since I have already done BQSR I have decided to try the second way.
I just need to change the read groups of one of my files so that they say the belong to the same sample. That way HaplotypeCaller will recognize them as a single sample.
To do this I'm using Picard AddOrReplaceReadGroups
Basically I'm making two new bam file (from ones that has previously been BQSRed) with a read group SM tag that both files will share.
Then I will feed both of the "samples" to HaplotypeCaller.
It took about 21 min per bam file, btw.
Note: I originally did :"on-the-fly" BQSR with HC so that I didn't actually have the calibrated bam files - it saves a step and some file space. So I had to make these calibrated bam files in order to change the Read Groups. Otherwise I would have had to option one in the link above and realign everything..