Coverage after Base Quality Recalibration: enough for variant calling?
Dear GATK community
I used Picardtools' CollectWgsMetrics to check the mean, median and sd of coverage for each of my samples (see attached figure).
For that I used base-quality-recalibrated BAMs and let them run through CollectWgsMetrics with default settings (incl. min mapping quality 20 and min base quality 20).
I was expecting to get coverage close to 30, as this is what we requested, however for many samples the coverage is much lower than that.
My general question is the same as in the title, but my specific questions are:
1/ am I using the right tool to check the coverage?
2/ did I run CollectWgsMetrics too stringently?
3/ should I evaluate coverage before base quality recalibration, and consider that as confirmation that the requested coverage was met?
4/ shoud we increase coverage for some of the samples by sequencing them again?
Any help and suggestions on how to go about this are greatly welcome.
Thanks in advace!