Separately calculate MAP based p-value for inosine and m6A

[kamaloxfordpathology]

Hi,

Hope you are well. Is there a way to calculate MAP based p-value separately for inosine and m6A.Right now it seems to pool both residues for calculation. Kindly let me know

Best, Kamal

[ArtRand]

Hello @kamaloxfordpathology

Sorry for the terribly long response time (new Modkit features are coming!).

Do you mean calculate a metric for m6A vs ¬m6A (so Inosine and unmodified will be combined)? If you give me a little more information on the biological question maybe I can give you more advice. There is also modkit adjust-mods --ignore <mod_code> which will re-distribute the probability of the --ignoreed modification.

[kamaloxfordpathology]

Hi @ArtRand ,

Thanks for the reply and also for the new features!. I wanted to explore the nascent transcriptome-wide changes for different types RNA modification upon treatment of cells. Therefore, shall I I split the modkit pileup data using adjust mods separately for m6A and Inosine and then perform differential modification analysis for each of them?. Thanks!

[ArtRand]

Hello @kamaloxfordpathology

shall I I split the modkit pileup data using adjust mods separately for m6A and Inosine and then perform differential modification analysis for each of them?

Actually, you can't do this and Modkit should give you an error. Are you trying to perform a "one-vs-all" test?

[kamaloxfordpathology]

Hi @ArtRand,

Thanks, I want to answer the following questions:

1. Where are loci (single base) which show significant change in m6A modification upon treatment?
2. Where are loci (single base) which show significant change in Inosine modification upon treatment?. Using DMR, I can find the A sites that shows significant change in overall modification but it does not tell the p-value for each of m6A and I?. I wrote a beta distribution test to model the probability of modification distribution at each loci for m6A. However, it takes too long to run. Kindly help. Thanks