Hi,
I have RNA-seq data for a certain tissue type for human, chimpanzee, gorilla and orangutan and macaque. I am thinking to look into differentially expressed transcripts instead of differentially expressed genes (DEGs).
For a reason, I was wondering if it is possible to do this in an unbiased manner. In order to get differentially expressed transcripts between species, I need transcriptome index (cDNA.fa) for every species. I assume that cDNA.fa file would be more through and complete for human compared to non-human primates. That’s why I was wondering if it is possible to get differentially expressed transcripts. If someone can please guide me about it. Thanks
(PS: Most of the studies I came across regarding comparison of RNA-seq data in primates, looked into DEGs and I didn’t find any analysis concluding the differentially expressed transcripts.)
Whilst I agree that the number of transcripts in human assemblies is expanded in comparison to the other genomes, it may not be an issue. You could look at whether orthologous housekeeping genes are getting similar TPMs among datasets, and if so, then you can check if that same expression is spread across more transcripts in humans.
If you do find that expression is more distributed among transcripts in humans you could try to limit the analysis to orthologous genes with similar numbers of transcripts among assemblies, but it's hard to say how much information you'd lose by doing so. Another idea would be to use a much older human assembly where there are fewer transcripts. It would look strange, but it's defendable given the context. Or just accept that expression will be more distributed in humans and weight the analysis accordingly.
Overall, I would say you just need to start the analysis to see how big the effect potentially is.