differences between intron excision ratios and psi values

[zhipeng...@gmail.com]

Hi,

Thanks for developing leafcutter. After checking the tutorial, I feel a little bit confused about the differences between "intron excision ratios" (calculated by "prepare_phenotype_table.py") and PSI values (calculated by "quantify_psi" function). Are they conceptually the same? I am assuming PSI was used for PCA, while intron excision ratios were used for splicing QTL, but could you please explain why that is the case? 

Also, I noticed that the sum of the "intron excision ratios" for a single cluster is not equal to 1 (larger than 1 in my data). Could you please explain why? 

Thanks,

Zhipeng

[Yang Li]

The intron excision ratio table are standardized and normalized (row
and column respectively), so they don't have to sum to 1. You can
treat PSI and intron excision ratios to be conceptually the same. The
differential expression analysis takes count as input since the
Dirichlet-multinomial is a discrete distribution, we found that QTL
mapping works better by simply using the ratio instead of modeling
counts. Hope this answers your questions.

Best,
Yang

> --
> You received this message because you are subscribed to the Google Groups "leafcutter-users" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to leafcutter-use...@googlegroups.com.
> To post to this group, send email to leafcutt...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/leafcutter-users/28441f2a-1e68-4f85-a53f-fb0253e8863c%40googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.

[yin wang]

hello,

l read the answer, but i want to know how the intron excision ratio are calculated and how it are standardized and normalized.

thanks,

Yin