I have a list of metabolites quantified using targeted metabolomics. I then have KEGG KOs inferred from metagenomic data.
I would like to see whether there is any KEGG pathway among the one detected that has several metabolites in my data.
In other words, I want to see what pathway is enriched in the metabolites I detected. I am aware of Bioconductor packages as pathview, but there I need to plot each pathway individually. Instead I am more interested in have a sort of automatic screening that could tell me that a path in my metagenomic datasets has also many metabolite matches in my metabolomic dataset.
I could also use another database, I am not limited to KEGG, this was just a initial choice. Any idea whether this is feasible?
Thank you
I have been doing some similar work. Currently, I am using the MetaboAnalyst for this purpose where I can not only perform KEGG pathway analysis but also separate Metabolite enrichment analysis. There is also the programmatic version available here. Check it out and try if it works at your end.
Also if you come across any other tool, do share as well here. I am keen to learn what all are available out there.
That looks interesting indeed. Thank you! However, it needs to have a reference organism selected and for metagenomic data this might be a problem. I will see how it performs.
I'm also interested in these kinds of tools. They do not seem to be so numerous or "straightforward" as the ones using genes.
I have tried the Bioconductor FELLA package. As far as used it, it allowed me to input a list of metabolites and a custom background of metabolites (with KEGG IDs such as
C00121
), and has a variety of tests (including a modified-hypergeometric). It works with KEGG pathways, and can return tables of significant results, although the analysis it does is a bit different to the typical hypergeometric enrichment. Additionally, in the introduction of the vignette it also discusses other tools.