How to cluster list of proteins with uniprot IDs by KEGG orthology groups

Question

[dlotzk]

I've got a list of proteins, complete with Uniprot IDs. I'd like to group them into orthology groups based on data from KEGG, but I have no idea how to add a column with KEGG groups. I've tried poking around in the documentation for UniProt, but all I can find is this page on KEGG - no mention of how to link to KEGG data. I've also found some online KEGG mapper tools (here and here) but those don't seem to be what I need in order to cluster my proteins into groups of KEGG orthologues.

[Ernest Bonat]

One simple good idea could be use the K-Means Machine Learning clustering algorithm.

Answer 1

[Elisabeth Gasteiger]

Unfortunately UniProt can no longer link to KEGG Orthology because there were clashes with the KO licencing scheme (cross-references to KO were removed from UniProt in December 2020, see https://legacy.uniprot.org/news/2020/12/02/release).

If you want to try with one of the other phylogenomic databases cross-referenced to UniProtKB (https://www.uniprot.org/database?facets=category_exact%3APhylogenomic%20databases&query=%2A) you can first upload your UniProtKB identifiers to the IDmapping service, https://www.uniprot.org/id-mapping, and then reconfigure your result table to add a column for your preferred phylogenomic database.

Download the resulting table and sort it by that column to see entries with the same cluster identifiers together.

Please don't hesitate to contact the UniProt helpdesk if you have any additional questions.