Hi all,

I am currently doing alternative splicing analyses using SUPPA. I have transcript-level quantification of Illumina and nanopore reads, each normalised with TPM and CPM respectively. My questions may be naive so please bear with me.

I am looking at individual events at each developmental stage - I understand that each sample has different RNA amount/compositions which may influence comparisons. For example, the number of a given AS events (e.g. Skipping Exon, Intron Retention) at one given stage as opposed to the other. My question is should I and how I could normalise the potential influence in a way that makes it comparable between different developmental stages? In addition to TPM/CPM normalisation or maybe use something else instead before passing it through SUPPA? And would it make sense to treat it as gene-level quantification when applying normalisation?

Also, this paper (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6678158/) normalizes the distribution of each AS event by the overall number of that event in the mouse transcriptome. Not exactly sure how to achieve this. I would greatly appreciate any insight on how that could be achieved.

Many thanks in advance.