GO to EC links - should they always be there?

[bmeldal]

I'm curating the VPS4 complex [e.g. Q9UN37 (VPS4A_HUMAN), O75351 (VPS4B_HUMAN)] which has

GO:0016887 ATPase activity, probably even GO:0042623 ATPase activity, coupled. (It dissociates the ESCRT-III complex at the end of membrane fission to facilitate recycling of its components)

It is also annotated to EC 3.6.4.6 Vesicle-fusing ATPase in UniProt and I agree this is correct. (Def: A large family of ATP-hydrolysing enzymes involved in the heterotypic fusion of membrane vesicles with target membranes and the homotypic fusion of various membrane compartments. They belong to the AAA-type (ATPase associated with a variety of cell activities) ATPase superfamily.)

But there is no link between the GO and the EC term. Should I create a new GO child to match EC 3.6.4.6 Vesicle-fusing ATPase?

Thanks, Birgit

[paolaroncaglia]

Aha: searching the ontology with the string '3.6.4.6' retrieves this:

[Term] id: GO:0008576 name: obsolete vesicle-fusing ATPase activity namespace: molecular_function def: "OBSOLETE. Catalysis of the reaction: ATP + H2O = ADP + phosphate. The hydrolysis of ATP by an AAA-ATPase, involved in the heterotypic fusion of membrane vesicles with target membranes and the homotypic fusion of various membrane compartments." [EC:3.6.4.6] comment: This term was made obsolete because it incorporates process information. synonym: "vesicle-fusing ATPase activity" EXACT [] is_obsolete: true replaced_by: GO:0042623 consider: GO:0016192 consider: GO:0061025

So the obsoleted term was replaced by 'ATPase activity, coupled'. The way EC:3.6.4.6 is phrased makes it out of scope for representation in GO, but I could add a RELATED synonym to 'ATPase activity, coupled' for "vesicle-fusing ATPase activity" with xref EC:3.6.4.6. How's that?

[bmeldal]

I saw the obsoleted term. A related synonym would at least tie them together, that would work.

[paolaroncaglia]

Done.