Can the GO Consortium meeting logistics be added to the GO wiki?
Judy will finalise the details with Claire and Amy and put instructions for accommodation booking onto the wiki within two weeks. She will also contact the SAB members.
Scope of GO
We have had a series of term requests from someone who works for the Ontology Adverse Events (OAE) who is using GO to annotate the action of specific chemicals (drugs), see . These processes are probably (but not necessarily) also carried out by proteins so we could probably add the terms on that basis, but the question is more general: to what degree do we as a project support this type of 'non-standard' annotation? And if not in GO, where else does it live? [Jane & David]
This was agreed OK in principle, it could be quite useful for us in future. However, it was determined that the OAE group had not responded to Tanya's request for more information, one of the ontology editors may make an attempt to contact the group again.
Recent paper about GO annotation from HTP papers
It was discussed that we should be more transparent about where certain datasets come from, this could be done be using ECO codes in the GPAD format, then users can pick up or ignore sets they don't want.
Rachael will make new GO_REFs for the Human Protein Atlas and LIFEdb annotations. These currently reference a paper describing the experimental method, the actual experiments are not published but located on the HPA or LIFEdb websites so are not referencable. A GO_REF was deemed a more appropriate reference for these annotations. The paper for HPA was included in the HTP analysis paper but it was felt this paper does not contain the same data as the other HTP papers, so referencing these annotations using a GO_REF should avoid confusion in future. A further possible mis-interpretation by other analysts of moving these annotations to a GO_REF could be that they assume these annotations were incorrect and so removed, but it was felt that as long as the change was announced and the GO_REF included both the PMID and the URL to the experiments then these annotations can be traced back correctly.