Hi,

I am using Limma/edgeR packages for analysing my RNA-seq data from two bacteria. I am going to do Gene ontology analysis on the data. There is no built-in package for the two bacteria that I work on, so I made organism packages using makeorgpackagefromncbia(). The problem is, the packages made this way do not contain 'egGO2ALLEGS', so it is impossible to do gene otology analysis using gonna() function from edgeR. Does anybody know if there is an alternative way to do so in R environment? or I need to export data and do the analysis on some websites?

Cheers, Babak

If there are websites that exist for doing GO analyses on bacteria, that might be the easiest way to go. Last I checked there wasn't a lot of GO information for bacteria at NCBI in the gene2go table, or UniProt, which is the current fallback if there's nothing at NCBI. It wouldn't be that big of a hassle to download GO data from a different source and insert into the SQLite database that you are creating, but if you can do it online without any fuss, why bother?

limma can do the GO analysis for a custom species if you provide the GeneID to GOTERM associations as a data.frame with two columns. To use that facility, use the kegga function instead of goana and give the data.frame as the gene.pathway argument. But I agree that using a bacteria annotation website would be the easiest if such exists.