SeqMonk RNAseq QC plot error (>100 exon?)

[JamieHeather]

I've just started exploring SeqMonk, just looking at human total RNA-seq data to check the distribution of different types of RNA.

However, when I run the RNA-seq QC plot option I get the following error:

Percent in exons was >100 for accepted_hits.bam

I imagine this is systemic of some larger discrepancy in my analysis (or at least I hope so, because this was an rRNA-depleted data-set that SeqMonk is claiming has ~75% of reads in rRNA!), but I can't seem to find what that may be.

The steps up to here basically involved using the iGenome hg19 as a custom genome, importing my accepted hits from TopHat, setting probes from the default running window generator settings and then doing basic read count quantitation.

Have I just done something silly, or is something strange afoot?

[JamieHeather]

OK problem solved! It was something to do with the way that I'd built the custom genome - when I just imported an Ensembl genome via SeqMonk and re-ran the QC the error went away (as did the presence of rRNA mapping reads, thankfully!).

[GenoMax]

Did your rRNA-depleted samples actually have rRNA (if SeqMonk was reading your data right) or was that just an artifact?

[JamieHeather]

I believe that was just an artifact.

I'm not sure exactly what it was about using the custom hg19 genome (I'd just thought that was the sensible choice given that's what I used for my reference with TopHat), but everything now looks as I'd expected it to.