raw transcripts to gene name

Question

[amoltej]

Hello all,

I have just completed my first RNAseq analysis using EdgeR.

But since I am working on non-model organism I do not have annotation file or corresponding gene names for it. I have got 22864 transcripts by using reference transcript data of closest relative. My question is how can I assign gene names to those transcripts? Blast is one way but it is not helpful for such a big data set. I have ran interproscan but that gives me different domains. This information is not 100% helpful since several genes may have same domains. can somebody please tell me any method or software that I can use to give proper functional names to my transcripts?

Thank you in advance

Amol

[Manuel Landesfeind]

Hi, I think the easiest method for functional annotation is homology search, i.e., BLAST analysis. You can probably not use NCBI NR as database but the genes/proteins from you close relative if they are already annotated. Otherwise, you have to create your own database from a set of related organisms. But any method you choose will be computationally demanding and time consuming.

Gene/transcript/protein annotation is a complex task and you should have a look at some papers about de novo genome assembly and annotation as well. This might give you a feeling about the challenges that are waiting for you ;-)

Answer 1

[Alternative]

What you can do is to assign orthologs to your genes. The following thread describes ways and tools to do that What Is The Best Method To Find Orthologous Genes Of A Species?. Inparanaoid is one such tool that works well. Good luck.