Or is it only possible with paired-end reads? Any previous experience with this is highly appreciated.
I don't see why not. As long as you have sufficient depth of coverage across splice junctions it should be fine. The longer the reads the easier it would be.
It is mainly the matter of coverage (similarly to described eg in http://www.ncbi.nlm.nih.gov/pubmed/21903743). Junction mappers (tophat, Star) can locate the junctions from single reads (since they are >75bp it's feasible)
Certainly you can do it using tophat pipeline. You should have to a good coverage(above 15 million reads).
A bit late to this but a colleague recently decided that single-ended reads of 100bp length or more are BETTER than paired end. The reason being that, for the same price, you get twice as much coverage. As Richard above says, high coverage and long reads lend weight to splice junctions. Paired-reads often only span one splice junction (if at all) and don't add much extra information.
As far as I know, there is still no perfect program to find differential isoforms in RNAseq. Some people recommend only comparing exon by exon and then use some other method to work out the structure of the isoforms present.
Hope this helps.
Dave
- Badges
- Science topic
- Similar topics
- Biological Science
- Molecular Biology
- Biomolecules
- Nucleic Acids
- RNA