Transcriptome Database

Question

[Bdv]

We have sequenced a novel transcriptome and we want to map it to some close organism. Would it only be correct to map transcriptome to transcriptome or may I map my transcriptome to the genome of the known organism? Where can I find database of the known transcriptomes if I still need to map my transcriptome to the known one? Many thanks!

Answer 1

[Chris Miller]

You can try to map your transcriptome to the other genome, but you'll end up discarding a lot of useful information if you do it naively. Many reads won't be from a single exon, but will span splice junctions. Unless you use a splice-aware mapper, you'll end up discarding those reads as unmappable. Something like Tophat/Cufflinks may give you reasonable results if the organisms are closely related.

For the second part, I don't know of any good transcriptome databases off of the top of my head.

Answer 2

[Ketil]

You don't say why you want to map your reads to the other organism, which may have some impact on the approach you should use.

In general, I think you are better off mapping your reads against related proteomes than against genome or transcriptome data, since amino acid sequences are usually better conserved. I don't quite see that "modern" aligners would help you much, they are usually optimized for highly similar sequences, and unless your related species is very close, I don't think they'll give you good results. I'd use blastp or tblastn, or similar.

Sequences can be downloaded from NCBI, uniprot.org, Ensembl, etc.