Closed Freymaurer closed 3 years ago
I'd argue that the gene ontology is"only" important for derived datasets (e.g. RNASeq / gene expression data) and not so much to annotate the pipeline towards the data. And I'd be surprised if anyone ever was to use this (manually) in an excel sheet rather than annotating with tools (interproscan) or importing from existing resources.
So i checked the number of ontologies from go and quite frankly we do not loose much by adding it. To summarize, our initial ontology set had references to some "directly linked" ontologies and we thought it might be a good start to add those. Our database now contains 2 316 953 terms and 9 629 are from the go ontology. So i'd argue we just keep it. At the moment most time needed to search for terms comes from the huge ncbitaxon (2 241 111) and chebi (136 671).
Ontology is available at: http://www.obofoundry.org/ontology/go.html
Reasons for this addition
USed by ontologies in db.