[Go] seeking comment on GO:0004772 (SF 1953281)

[Midori Harris]

Dear GO,

SF 1953281 reports a problem with GO:0004772. At present the term name is 
'sterol O-acyltransferase activity', but the definition is specific for 
cholesterol, not generally applicable to any sterol:

   id: GO:0004772
   name: sterol O-acyltransferase activity
   namespace: molecular_function
   def: "Catalysis of the reaction: acyl-CoA + cholesterol = CoA +
     cholesterol ester." [EC:2.3.1.26]
   xref: EC:2.3.1.26

The definition is derived from the cross-referenced EC entry, and is also 
consistent with some, but not all, of the synonyms. It is clear that we 
will have to diverge from EC slightly, because the current term name their 
recommended name, and for GO we want the name and definition to be 
consistent.

We intend to implement this structure:

   sterol O-acyltransferase activity GO:term1
   --[i] cholesterol O-acyltransferase activity GO:term2
   --[i] ergosterol O-acyltransferase activity GO:term3
   --[i] lanosterol O-acyltransferase activity GO:term4

The question is whether GO:0004772 will become term1 or term2 in the new 
structure, i.e. whether to change its name or its definition.

- If we change the name to match the definition, SGD, CGD and GeneDB S. 
pombe will have to change their annotations.

- If we change the definition (and  remove the EC dbxref) to match the 
name, MGI, RGD, UniProt and (I think) Flybase annotations will be less 
specific, and should be changed.

- If we make the term obsolete, everyone will have to change annotations. 
This option is perhaps the most correct given the name/definition 
mismatch, but would mean the most work for annotators.

SGD and CGD curators have expressed an understandable preference for the 
first option, renaming GO:0004772. Does anyone -- especially annotators 
who would have to change existing annotations to maintain specificity -- 
object?

SF link:
https://sourceforge.net/tracker/index.php?func=detail&aid=1953281&group_id=36855&atid=440764

Thanks,
Midori