#11774 Relationship edits in Cardiac conduction ontology

[Ruth]

Hi

looking at writing cardiac paper (still). I think that the following terms need a part_of relationship removed because the parent term is also a child of the other parent term.

GO:0086056 voltage-gated calcium channel activity involved in AV node cell action potential
GO:0086060 voltage-gated sodium channel activity involved in AV node cell action potential
GO:0086086 voltage-gated potassium channel activity involved in AV node cell action potential repolarization
GO:0086087 voltage-gated potassium channel activity involved in bundle of His cell action potential repolarization
GO:0086088 voltage-gated potassium channel activity involved in Purkinje myocyte action potential repolarization
GO:0086058 voltage-gated calcium channel activity involved in Purkinje myocyte cell action potential
GO:0086057 voltage-gated calcium channel activity involved in bundle of His cell action potential
GO:0086061 voltage-gated sodium channel activity involved in bundle of His cell action potential
GO:0086062 voltage-gated sodium channel activity involved in Purkinje myocyte action potential
GO:0086063 voltage-gated sodium channel activity involved in SA node cell action potential
GO:0086059 voltage-gated calcium channel activity involved SA node cell action potential

Also noticed the same also possibly needs to be done for the following (although here there are slightly different relationships providing the double 'path' to the grandparent term):

GO:0008324 cation transmembrane transporter activity
GO:0005267 potassium channel activity
GO:0086019 cell-cell signaling involved in cardiac conduction
GO:0005262 calcium channel activity
GO:0005272 sodium channel activity

I went through all the voltage-gated channel activities involved in AP terms and not all of them have this double path. However, if you think that the double path is correct then the voltage-gated channel activities involved in AP not listed above need to be edited to add the grandparent relationships directly so that there is some consistency for all of these.

Please let me know which way this should be, so that I can write about this now rather than waiting for the edits to be done, as I do appreciate this won't be a high priority.

Thanks

Ruth

[David Osumi-Sutherland]

These redundancies should be being removed during the release process. The broad issue was discussed on a recent editor's call where I was very surprised to find out that this wasn't happening. There is apparently a plan to impelment this redundancy stripping in the release pipeline soon using robot (the replacment for OWL tools).

Why we should be careful about removing this by hand: Once we can remove instantiated inference from the ontology, then much apparent redundancy in the editor's file will simply disappear. Trying to remove this earlier risks the ontology autophagy that was happening a year or so ago.

[David Osumi-Sutherland]

UPDATE: looks like redundant classification stripping in public GO just got turned on last night. Should be visible in QuickGO in the next day or two. Worth checking if this has fixed the problem. May not have though as the correct way to do this is probably to remove part_of relationships and I suspect the updated pipeline only removes is_a links in cases where there are two is_a paths.