Quantification of Maps to rRNA and mtDNA

Question

[dec986]

Hello,

I want to quantify how much rRNA and mtRNA has been mapped during an RNA-Seq experiment. I have heard of doing this with featureCounts, but haven't seen an exact way to do it.

My guess was to use featureCounts but this only works off of successfully assigned reads, which doesn't answer the mapped reads. So featureCounts won't work, or maybe, because the rRNA would have been identified through featureCounts anyway?

I have seen several examples of this on bioStars and elsewhere, but I don't see how I apply what they did. I can't find any rRNA databases for mm10 (or anything). My alignment is done with STAR.

How can I quantify how much rRNA and mtRNA has been mapped during an RNA-Seq experiment? -DEC

[Devon Ryan]

Do you need an exact count of reads for statistics or just an approximate percentage for QC? Which organism?

[dec986]

an approximate percentage for QC would be good, and this is for mm10 mouse.

Answer 1

[Devon Ryan]

If you just need an approximate percentage then take the mouse mm10 genome and add the Rn45S sequence for mouse that you can get from NCBI (the rRNA sequence isn't part of the GRCm38 genome). Align against that, sort, index, and use samtools idxstats to get the numbers for mtDNA and Rn45S. The real Rn45S numbers will be higher, but this gives you a decent number for quality control purposes (e.g., seeing how good your ribosomal depletion is).