KEGG mapping from de novo transcriptome annotation

Question

[pablo61991]

Hi Biostars forum

I have a de novo transcriptome annotated mainly against SwissProt, so I have contigs ==> UniProt accession codes. Using the UniProt conversion tool I can obtain GO and also KEGG which corresponds to the UniProt accession, but obviusly I have obtained matches against different organisms. My desired next step consist in to map my KEGG codes in to pathways, but the KEGG mapping tool ask about "org" code and I can't obtain a "global" result if I have to select my matching organisms one by one. I was looking for a solution googling and using the forum search tab but I can't find anything.

As an example:

bta:540248
hsa:51202
mcc:574295

I can search 1 by 1 if I use the organisms code, but I cant map them together. Is that possible?? or I need to find orthologous in just one species??

In some papers focused around non-model organism I have read in M&M section the use of KOBAS to perform KEGG pathway analysis. However, when I have tried this resource I found the same problem, I have to specify one species when my annotation file contains SwissProt IDs coming from different species. I know if I do not specify one species, KOBAS takes human by default. Maybe these authors miss used KOBAS and have analysed their data just for the transcripts with hits against human?? I have missed something about KOBAS usage??

Please give me some wisdom tips, thank you

Answer 1

[Damian Kao]

You'll probably need to write a script to do this or use the KEGGREST R package.

You can find the KEGG ID for the gene using the KEGG API. For example:

http://rest.kegg.jp/get/hsa:51202

From that you can parse out the KEGG ID from the orthology line:

ORTHOLOGY   K14777  ATP-dependent RNA helicase DDX47/RRP3 [EC:3.6.4.13]

Then you can get the entry for that KEGG ID:

http://rest.kegg.jp/get/K14777

This shows all the orthologus genes in the various organisms in the KEGG database. You can then parse the ids out for your organism of interest.

[pablo61991]

Thank you. As I supposed when can not find anything in my search is not an easy task. I will give a try to use the R package.

Answer 2

[pablo61991]

I have one question related with the use of KOBAS to perform the KEGG pathway analysis which I mentioned (data annotated against different organisms). In these papers they annotate against NR or SwissProt, so I suppose they also have their transcripts annotated against several organisms. My question is, if KOBAS ask to introduce the species under study, how its possible to do that??

Maybe in these papers they "fail" and the KEGG pathway analysis is done only for one species of the multiple species contained in the annotation??

[GenoMax]

Please ask this a a new question.

[pablo61991]

Could I edit my original post instead open one new?

[GenoMax]

Yes you can but I thought you were specifically asking about KOBAS and that is why I suggested a new post.

[pablo61991]

If you are fine with my edit (its already done) I ll avoid to open more treats. In other case I will fix it as you consider the best way.