I have a BAM file with 10 different read groups all from the same sample:
- 2 of those were test libraries sequenced in a MiSeq.
- The remaining 8 are from a HiSeq and are much more abundant.
Can GATK recognize that the input quality scores will have a different profile for each read group? Should I split my BAM file with respect to sequencing platform or even on a per read group basis?