How to calculate PSI value from rMATS like this?

[Nipan Hazong]

I've taken this screenshot from someone else supplementary file. I wanted to know how can I obtain PSI value just like this from rMATS txt files.

[kutsc...@gmail.com]

In that screenshot it looks like they took the values from the regular rMATS output file and just renamed the columns (IncLevel1 -> PSI_CK), (IncLevel2 -> PSI_IBA), (IncLevelDifference -> delta PSI)

Eric

[Nipan Hazong]

Thank you so so much Eric.

[Nipan Hazong]

If the IncLevelDifference is positive, it means that the alternative splicing event is more prevalent in your second condition (e.g., treatment or disease state), suggesting up-regulation. If it’s negative, the event is less prevalent in the second condition, suggesting down-regulation.

Eric, is this correct?

[kutsc...@gmail.com]

Saying the event is more prevalent in one sample or the other is a little confusing. IncLevelDifference is: average(IncLevel1) - average(IncLevel2). The inclusion level is a ratio of the counts for the two isoforms (inclusion isoform and skipping isoform)

It's not clear what it means for one condition to have more of the event, but you can say that one condition is more likely to use one of the isoforms. See this post: https://github.com/Xinglab/rmats2sashimiplot/issues/68

Eric