Hi everyone,

Im fairly new to Bioinformatics and working on a project on aberrant splicing in cancer. For that I wanted to calculate the healthy background expression of splicing events of interest using the GTEX db.

So with Snapcount I managed to calculate PSIs for the events for the samples that contain junction overlaping reads. To account for sequencing depths etc. between samples I want to calculate junction reads per million = ( junction reads / total reads per sample) * 10^6.

To do so I built a named vector with by getting all the gtex samples reads and rail_id from the gtex samples' metadata with recount3.

However, somehow the rail_ids Snapcounts QueryBuilder function is returning don't match any of the rail_ids in my vector. (obv. they have the same str rail_XXXXXXXX).

Is that because Snapcount uses recount2? Does anyone have experience with this?

Thank you in advance! Henry