Here’s an interesting new paper we found: Mahajan et al. have developed their own knowledgebase and tool to conduct GO over-representation analysis for a metabolite list: IDSL.GOA. Although the GO is intended to describe gene function, this is a great example of other ways GO might be used. We have not had a chance to test it out, but if you’re interested they have an online tool: the IDSL.GOA online tool is a user-friendly resource for identifying overrepresented metabolic processes in a list of metabolites. The online interface offers features including analysis, query, explore, statistic and download options. Note this is not an official GO tool and is not affiliated with GO or GOA, please contact the tool developers with any questions about IDSL.GOA.
the script to connect the GOA terms to metabolites is available IDSL.GOA/IDSL.GOA Knowledgebase Script at main · idslme/IDSL.GOA · GitHub
In short, we are relying on PubChem to get the EC to compound linking, which is sourced from Rhea, Reactome, Wikipathways, and Biocyc databases. Then, ECs are linked to Genes by NCBI Gene db, and then Genes are linked with metabolic GO terms via the GO db.
They are happy to answer any additional questions. The resource is part of Dinesh Barupal’s U24 on building a database for exposome ( RePORT ⟩ RePORTER ).