I'm looking for some help when analysing RNA-Seq data for differential exon usage/splicing. I've previously used edgeR for DE analysis and was happy to see diffSpliceDGE as part of that package. What I don't know is:
- how to obtain exon level counts
- whether there is a normalisation step involved like there is in a DGE pipeline (i.e. TMM)
For the first point, I have BAM files for the samples I'd like to analyse that were mapped using STAR to the GRCh38 reference. Is there a way to extract exon-level counts from these files? In case it matters I'm analysing data from an exon skipping experiment and so will be looking at differential splicing for our gene of interest as well as any off-target effects.
For the second point, I've been through the example on the documents page (https://rdrr.io/bioc/edgeR/man/diffSpliceDGE.html) and while that made perfect sense I noticed there was no normalisation step used. How, if at all, should the exon level count data be normalised prior to analysis?
Many thanks for your help and time.