search

williamritchie / IRFinder

Detecting intron retention from RNA-Seq experiments
53 stars 25 forks source link

IRFinder on total RNAseq libraries #161

Open ojziff opened 2 years ago

ojziff commented 2 years ago

Dear @dg520

I wondered if IRFinder could be used to quantify splicing in total RNA libraries ± Ribo-Zero RNAseq experiments e.g. with TruSeq Stranded total RNA? i.e. not undergone oligoDT enrichment for polyA mRNA?

I note a previous issue asking about comparing IR in polyA vs Total but my query here is about comparing IR between Total libraries. Should the output be filtered e.g. by raising the minimal intron coverage threshold?

Many thanks! Oliver

dg520 commented 2 years ago

@ojziff Sorry for a late reply. Applying IRFinder on Ribo-Zero RNAseq experiment is do-able. The commands to run is the same as those for TruSeq. As you have already noticed, the issue sits in how to interpret the results and get meaningful biological findings out of it. I have never used Ribo-Zero derived result for biological discovery so far. But raising the minimal intron coverage threshold makes sense to me as there should be quite a lot of nascent transcripts. Comparing nascent IR between conditions is quite tricky as well. Those potential difference observed might be derived from batch effect. This is because splicing is a dynamic process and RNASeq is just a snapshot at a "random" time point of it. It might be impossible to convince ourselves we capture the two conditions at exactly the same time point in a splicing period. You might need additional support/evidence for controlling such confounding factors, or use some other data cleaning approaches if they exist for this specific purpose. I hope this information is helpful.

Best, Dadi