Fusion Isoform Quantification

Question

[Nicolas Rosewick]

Hi,

I performed an RNA-Seq experiment with a very heterogenous tumor sample. In this sample, I found a fusion gene transcript ( genes on different chromosomes), and also the WT gene transcripts. How can I check the quantification of this "fusion isoform" compared to the WT transcript ( see figure for more explanation ). I think using a count-based approach like DEXSeq. I don't know how other isoform quantification approach (RSEM, cufflinks,...) support fusion isoform (so isoform on different chromosomes). In final I just want to know the proportion of WT transcript 1, WT transcript 2 and the fusion transcript, and maybe compute a ratio.

What do you think ?

[Biomonika (Noolean)]

I recently came across this paper http://www.biomedcentral.com/1471-2105/15/81 but have no idea how suitable it would be for your problem.

[Nicolas Rosewick]

No that's more for splicing events. I want to have an idea of the expression of these fusion transcript compared to the WT transcript.

Answer 1

[Manvendra Singh]

I recently have done it. I used TopHat-Fusion, STAR and MapSplice2.

I found STAR and Mapsplice2 quite promising.

you can visualize and see that how the fusions look like, and moreover, the number of reads mapped on splice junction would also be there in output.