This is a test version of Biostars. For the public version, visit https://www.biostars.org.
kegg IDs to pathways in R

Hi all,

I have never done this before but i suspect there must be an easy way to convert KEGG ids to pathways in R. I have a dataframe containing info from 1000 sth bacterial genomes; their taxonomy, their orfs and the kegg ids (which i identified with eggNOG functional annotation) and I would like using these ids to produce figures of pathways. Ideally I d like to have the different taxa with different colors (e.g. for carbon metabolism Firmicutes enzymes presence could be represented with red arrows, Actinobacteria enzymes presence with brown arrows etc etc) Does anyone know or have any recommendations for an R software that would easily do that? I saw KEGGprofile and I dont think it can do exactly what I am looking for - or maybe should i give it another go?

Thanks for any recommendations in advance P

r kegg list pathways

2 answers

I've used clusterprofiler before to annotate pathways and find enrichment. I think it can do what you are asking, if I'm understanding you correctly.

https://bioconductor.org/packages/release/bioc/vignettes/clusterProfiler/inst/doc/clusterProfiler.html

thanks, pathview and clusterprofiler did the trick. I can now use a list of KEGG IDs as querry and render them to a metabolic path and automatically save the image. That speeds things up a lot. However, useful as it is, it allows to color different compounds and genes based on their expression levels (transcriptomic data) but does not allow to plot multiple species on the same pathway map (which you could present with different colors)... would be nice if they integrate sth like that i the future

I used to use the KEGGREST package for this purpose. Using (e.g.) the keggLink function, you can map KEGG IDs to pathways. I wrote a script to split input queries into batches so that the server doesn't time out.

did not manage to find a way to get keggLink do what i want! it is helpful but not what i was looking for! thanks

Log in to answer this question.