is it possible to filter small RNA from a regular transcriptome?

[stvos]

Dear all,

I was wondering if anyone has ever tried filtering (potential) small RNA from a regular RNAseq transcriptome .
I have 38 assembled transcriptomes of a particular organism in different circumstances. I was thinking that these should contain small RNA as well, and that I could filter them out, based on transcript size, but that it might be difficult to make the difference between degraded regular RNA and real small RNA. The upside is that I also have a well-assembled genome of the organism in question, so I figure I could align the transcripts with the genome and remove most of the degraded transcripts this way.
Has anyone ever tried this and if so, with success or not?


Eager to use all the info in my data, but still wanting to use the time wisely!

Cheers,

Stephanie

[GenoMax]

Someone with experimental expertise will chime in but it is my understanding that small RNAs are eliminated from a regular RNAseq prep because of size selection. You may see some that may have escaped but in general you would need to do a small RNA prep to capture those.

[kerplunk412]

Someone with experimental expertise will chime in but it is my understanding that small RNAs are eliminated from a regular RNAseq prep because of size selection. You may see some that may have escaped but in general you would need to do a small RNA prep to capture those.

Yes, with most RNA-Seq protocols the small RNAs are eliminated in the cleanup steps. Also, the typical first and second strand synthesis strategies will rarely capture the entirety of an RNA molecule, which is a big problem if your target RNA is only ~20 nt to start.

[Brian Bushnell]

You could try finding these via adapter-trimming then filtering for reads <30bp or so. If these map to the genome and pile up in specific locations with no mismatches, they sound like small RNAs.

[HESmith]

In addition to the size difference, most RNA-Seq preps employ poly(A)+ selection to enrich for mRNAs. Small RNAs are not poly-adenylated and therefore excluded from the library.

[dpryan]

In my experience, <0.01% of a normal polyA-enriched mRNAseq library ends up being small RNAs. I wouldn't trust that for statistics.

[Melissa]

I agree with previous comments that ~20bp miRNA will not be sequenced. However, you can find primary miRNA transcripts in your transcriptomes because they are polyadenylated. Just download the hairpin sequences from miRBase and search against your transcriptomes.