Manager 16October2013

• In Protein2GO we do not allow ND annotations to the root node 'Molecular_Function' to co-exist with protein binding annotations. It has been requested that we allow an ND annotation for molecular function to be added to an entry when the only other function annotations are to protein binding. Curators want to indicate that there is no known activity of the protein, even though interaction partners are known.

We would like a consensus decision on whether this should be allowed.

Discussion

Present: Suzi, Judy(for 10 min), PaulT, Huaiyu, Rama, Pascale, Jane, Kimberley, Rachael, David, Chris

• Kimberley- Making a ND annotation to the root node along with a protein binding annotation allows us to say that we don't know the function but we know the protein binds to something.
• Judy- Isn't it confusing to say that we don't anything and at the same time saying it is interacting with something? Isn't that knowing something?
• PaulT- The question of is protein binding a 'Function' came up. There are 2 classes of Molecular functions:
  • biochemical activities (enzymes) that causes something to happen (influences something). These are physical changes which affects something. These are changes in physical Activity. Example: phosphatase regulator
  • actions where the cause is not known (influence is not known). These are changes in just the physical Entity. Example: protein binding
• Pascale: should we then allow GTP binding etc ?
• Chris: We should consider this proposal
• The ontology group will discuss this either during an ontology call or annotation call.