Deep Sage for Plants with no genome-Good approach?

[KevinLam]

Given that there's alot of plant genomes out there.
Is it a good idea to use Deep SAGE to do gene expression profiling for a plant with no reference genome?

Would the bioinformatics be vastly different if I just used the refseq filtered from plants to annotate the tags that were counted?

[krobison]

You could try simulating the results by picking some species with a good cDNA database, simulate Deep SAGE, and then see how many of the results you can interpret using other genomes.

My guess is that unless there is a closely related species already sequenced, this is going to be troublesome -- SAGE tags are from the 3'UTR and that may not be very conserved. You run the risk of having a lot of tags which are unannotatable.

Of course, one approach around this would be to also run a library of longer RNA-Seq reads. The long reads would serve as stepping stones to annotate the SAGE reads.

[KevinLam]

Thanks for the prompt reply!
Yes that is the best scenario 454 denovo or at least hiseq mate pair as skeleton. However, the cost is still prohibitive and it is likely to be a one off sequencing project to churn out expressed tags or pointers to likely markers for environmental change to be used in a custom microarray.


hmmm keep the suggestions and feedback coming!