predicted CDS regions in CTAT-LR-Fusion

41 views
Skip to first unread message

Asher Preska Steinberg

unread,
Feb 3, 2025, 1:44:15 PM2/3/25
to Trinity_CTAT_users
Hi Brian,

Thanks again for your help getting CTAT-LR-Fusion setup with my data. I have begun exploring some of the results, and I was wondering if you could explain to me how the CDS regions are predicted that are included in the output tsv? My apologies if I missed this in the manuscript. 

The reason I ask is I am working on a project where we are trying to match some of these fusions to proteomics data we have, and I'm wondering if I can use these predicted CDS regions to provide protein sequences for the fusions, or if perhaps it is better to use something like your tool transdecoder to predict CDS regions instead.

Thanks for your time and help.

Best,
Asher

Brian Haas

unread,
Feb 4, 2025, 7:51:03 AM2/4/25
to Asher Preska Steinberg, Trinity_CTAT_users
Hi Asher,

In this case, it's only making a guess based on how the breakpoint matches up with the reference transcript structures in each case.  If there's a combination of reference isoforms that when stitched together at that breakpoint provide for an in-frame situation, then it will report that in preference to having an out-of-frame scenario.  There's certainly a lot of room for improvement in this approach given that long reads provide more isoform-specific info that's not being leveraged here at the moment.

Hope this helps

--
You received this message because you are subscribed to the Google Groups "Trinity_CTAT_users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to trinity_ctat_us...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/trinity_ctat_users/76af6aaa-1ac6-4bca-8a12-176cede46b9fn%40googlegroups.com.


--
--
Brian J. Haas
The Broad Institute
http://broadinstitute.org/~bhaas

 

Asher Preska Steinberg

unread,
Feb 8, 2025, 3:35:16 PM2/8/25
to Trinity_CTAT_users
Hi Brian,

Thank you for explaining, this is very helpful! And I would imagine in this case if there is no known CDS associated with the transcripts then no fusion CDS region is reported? I see some of my results have amino acid sequences in this column in the results table and some have nothing reported.

Best,
Asher

Brian Haas

unread,
Feb 9, 2025, 7:48:59 AM2/9/25
to Asher Preska Steinberg, Trinity_CTAT_users
Right - it requires that the breakpoint happen in coding regions of both fusion gene pairs.

Asher Preska Steinberg

unread,
Feb 9, 2025, 9:03:47 PM2/9/25
to Trinity_CTAT_users
Okay great, thanks for your help!

Best,
Asher
Reply all
Reply to author
Forward
0 new messages