Closed bbimber closed 3 months ago
Hi, thanks for the feedback.
genome1K.phase3.SNP_AF5e2.chr1toX.hg38.vcf.gz
file, here I assume the genome version is hg38.--maxDepth
is the highest possible value (currently INT_MAX), while we have little experience about how to set a reasonable value, which should be sample/BAM-specific. We would suggest using the default value in this case.ok, thank you
Hello,
I'm processing 10x scRNAseq data using mode 2b. We do not have any non-multiplexed / single-donor data for mode 2b. I am giving it what I think is a fairly vanilla command:
The job seems to be moving, but it reports many messages about high depth (which i think are from pileup itself):
I have a couple of questions:
In the situation where the only available data is the multiplexed 10x BAM (containing cells from 2 donors), is there an option besides mode 2b? This is one lane of 10x with about 20K cells, and ~250m paired reads.
Setting --maxDepth seems like it might be a good idea here. Do you have any suggestions or experience on reasonable values? My assumption is that we simply want to avoid extremely high-depth loci, and therefore a fairly high value is OK, but this is purely a guess.
Are there other optimizations we should consider?
Thanks for any help.