Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Why are these events not detected? #161

Open
lennartraman opened this issue Nov 18, 2021 · 1 comment
Open

Why are these events not detected? #161

lennartraman opened this issue Nov 18, 2021 · 1 comment

Comments

@lennartraman
Copy link

lennartraman commented Nov 18, 2021

Hi devs,

We've been using rmats-turbo quite extensively and we are very satisfied the results! There are however some instances where certain events are not detected (for our application, sensitivity is very important). Could you provide some insight on why these are not detected, and possibly, how these might be detected if we alter some parameters?

I've been using v4.1.1, with one bam file, --statoff, and --novelSS. Below are some screenshots from IGV with 'missed events'. Each time you can see (from top to bottom): a coverage histogram, a splice junction diagram, a couple of mapped reads, and the used gtf annotation.

  1. no events are detected in this region, eg, the large exon skip

image

  1. undetected exon skip

image

  1. undetected intron retention (even for non-de novo events): seems to be a recurrent problem

image

  1. undetected A5SSs

image

Your input is very much appreciated!

Thanks a lot, Lennart

@EricKutschera
Copy link
Contributor

rmats will only detect an event if it has evidence for all the exons and splice junctions in the event. The evidence can either be from the gtf annotation or from the aligned reads

Here is an issue that explains the requirements for detecting a retained intron event: #17

The exons and junctions that rmats needs for each event type are shown in this diagram from the README: https://github.com/Xinglab/rmats-turbo/tree/v4.1.1#output

In the second example you posted (undetected exon skip), it looks like the gtf annotation only includes 2 of the exons required for the event definition

In the third example (undetected intron retention), it looks like the gtf has a transcript where 2 consecutive introns are retained. But rmats requires a transcript that retains just the single intron so that it looks like the diagram in the README

In the fourth example (undetected A5SSs), it looks like the splice junctions match with the gtf so rmats would not define a novel exon, but the exons in the gtf differ at both endpoints. rmats requires that the short and long exons in A5SS share the same endpoint on one side

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants