Entering edit mode
4.8 years ago
c_u
▴
520
Hi,
As we know that fusion transcripts could be a result of genomic translocation events, or could be due to trans/cis splicing events among mRNA. Let's naively call them gene fusions and transcript fusions. Now there are a ton of tools that predict transcript fusions from RNA-seq data. Can these tools distinguish between these two types?
Like for a given fusion predicted, can they say whether it must have been due to genomic rearrangements or fusion at the mRNA level? Or is this not even possible?
First do you know if your organism does transplicing or not? Most fusion transcript software is develop for use on cancer samples, and mammals don't do trans-splicing.
If you are working in something that does do transplicing.... you could look at where the break point is? If its a transplice then it won't just fuse in the middle of an exon.
But I suspect that there is no real way to be sure without matching DNA sequencing.
Are you sure that trans-splicing does not happen in mammals? It seems to have been reported in human cancers - https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5305119/
All things are possible I guess. I know that you can get transcriptional read through and splicing of adjacent genes that have ended up on the same transcript, but this is the first I've heard of true intergentic-trans splicing. But I'd guess for any given transcript, the prior on whether something is a genomic rearragnement or mapping error compared to a trans-splice must be massive. Particularly if the break point isn't a normal splice donar/acceptor sequence.