How to interprete "NAs" in the psi.gz file?

[SiyiWanggou]

Hi, Timbitz: I'm using "Whippet" to compare alternative splicing of different samples from RNAseq data. The idea to identify AS events from event-level is really brilliant. When I finished the Whippet analysis, I got a psi.gz file from each sample. In the file, there are a lot of "NA"s. How to interpret these "NA"s? Is it "0" or can just be filtered out in the following analysis? On the other hand, I found you used Mann-Whitney U test to compare the difference between samples in the published paper, but the Whippet package provides whippet-delta.jl for sample comparison. What's the differences between these two methods? Is Mann-Whitney U test integrated in whippet-delta.jl? Or, should should we use Mann-Whitney U test to find differences independently? Thanks

[meettel]

Hi, I am resuming this old issue since I had the same problem in interpreting NAs in the .psi output, do you have any suggestion about this? Thank you

[OlgaVT]

Hi,

I have the same question. What are 'NA's' in the Type column? And how to interpret 'NA's' in the Psi column? Thank you!

[timbitz]

NAs are nodes without a simple type, which would usually be caused by combinations of start/end sites embedded within spliced nodes. They aren't quantified by default, and so ignoring them is a simple solution. If you are specifically interested in an exonic interval that is given an NA type then the only way to remedy that is to simplify the input reference annotation over that specific region/gene.