Query: receptor activator activity

[pgaudet]

Following up on https://github.com/geneontology/go-ontology/issues/14220- @thomaspd mentioned that 'receptor activator activity' should be synonymous to receptor ligand activity, but the term is not clearly defined and the usage has been quite varied. The good news is that there are only 5 direct EXP annotations:

MGI:MGI:96434 insulin-like growth factor 2 GO:0030546 BHF-UCL NCBITaxon:10090 IMP MGI:MGI:4438697|PMID:20032056 20110420 -> I think this was used a 'receptor ligand activity'

UniProtKB:Q86YN6 Peroxisome proliferator-activated receptor gamma coactivator 1-beta GO:0030546 UniProt NCBITaxon:9606 IDA PMID:11854298 20061127 -> nuclear factor coactivator

UniProtKB:P07355 Annexin A2 contributes_to GO:0030546 BHF-UCL NCBITaxon:9606 IDA PMID:22848640 20150302 -> not a receptor activator (it inhibits receptor degrdation)

IntAct:EBI-10814663 ANXA2-PCSK9 complex GO:0030546 BHF-UCL NCBITaxon:9606 IDA PMID:22848640 20161019 -> not a receptor activator (it inhibits receptor degrdation)

MGI:MGI:97599 protein kinase C, epsilon GO:0030546 MGI NCBITaxon:10090 IDA MGI:MGI:3762436|PMID:17875639 20110325 -> in this paper they describe a modulation in receptor sensitivity by Prkce

@ukemi @bmeldal @RLovering is 'receptor activator activity' really what you want to capture in these annotations? If we forget existing GO classes, what function would you like to capture ?

Thanks, Pascale

[bmeldal]

IntAct:EBI-10814663 ANXA2-PCSK9 complex GO:0030546 BHF-UCL NCBITaxon:9606 IDA PMID:22848640 20161019

@NancyCampbell 's complex (and I reviewed it...). I agree, probably not the right term. Could you please suggest an alternative, Nancy? You can make the edits directly in the editor.

[RLovering]

MGI:MGI:96434 changed to receptor ligand activity UniProtKB:P07355 changed to molecular function regulator

I would suggest EBI-10814663 revised to molecular function regulator

Ruth

[bmeldal]

I would suggest EBI-10814663 revised to molecular function regulator

A little top-level (doesn't really say much...), what about grandchild term GO:0030546 receptor activator activity?

[ukemi]

I think originally this term was meant to cover the things that can activate a receptor other than just ligands. That would make sense for the Prkce. But I can remove that annotation if the term changes.

[pgaudet]

Hi @ukemi
The problem is that it seems some people interpret this to also include ligands. We should clarify the term and the definition if we want to capture things that can activate a receptor other than just ligands.

Generally - do we want to have 'gene product class x' regulator activity and regulation? (this is really parallel to the binding question).

[hattrill]

Hi,

Only just seen this. This has cropped up on another ticket (receptor agonist activity ; GO:0048018) #12272, which @vanaukenk found as part of questions wrt to wnt signaling/receptor ligand activity, etc.

Here's a link to a spreadsheet of the annotation to the term 'receptor activator activity' and child terms and quick analysis I did of them: https://docs.google.com/spreadsheets/d/15HcC6LFgOhA-BdQHvKMRjk3e-uIaGvitD5xH8Qzdkpk/edit#gid=233101787

It's a bit of a mess. As @pgaudet said, there are clearly those that should be annotated as 'receptor activator activity'. There are some things that directly regulate receptor activity (allosteric, PTMs) and things that regulate the expression of the receptor and there things that act at the transcriptional level with steriod hormone receptors (this applies to receptor activator activity and its children).

[NancyCampbell]

@bmeldal

I would suggest EBI-10814663 revised to molecular function regulator

A little top-level (doesn't really say much...), what about grandchild term GO:0030546 receptor activator activity?

EBI-10814663 complex is currently annotated to receptor activator activity and it seems this is where the problem is - at least as I understand it from Pascal's @pgaudet original comment

Following up on geneontology/go-ontology#14220 @thomaspd mentioned that 'receptor activator activity' should be synonymous to receptor ligand activity,

The complex inhibits LDL receptor degradation which is definitely not synonymous to receptor ligand activity defined as:"Binding to a signaling receptor as part of signaling and modulating the effector activity of that receptor to initiate a change in cell activity .....". i.e. my complex does not bind to the receptor.

I agree with David @ukemi. Yes the term's definition: ''The function of interacting (directly or indirectly) with receptors such that the proportion of receptors in the active form is increased" seems to

cover the things that can activate a receptor other than just ligands.

However, i admit that when I did the annotation I seem to have concentrated on second half of the definition i.e. "such that the proportion of receptors in the active form is increased" as my complex inhibits LDL receptor degradation, and took the 'very' wide meaning of "indirectly" in the first half. Now, even though I can justify the former, I really can't justify the latter.

@pgaudet I am happy to change the annotation.

So for the time being I went with Ruth's suggestion and changed to molecular function regulator - or better suggestion if anyone disagrees.

Hope that is ok.

Thanks.

[bmeldal]

Thanks @NancyCampbell - I seemed to have gone round in a circle there and made the same assumption as you did when you first annotated it! So yes, I agree we have to lift the annotation up a couple of levels.

Will you make the change in the Editor, Nancy?

[NancyCampbell]

@bmeldal Thanks Birgit. Done for both mouse and human.

[pgaudet]

Hi @bmeldal what this points out is GO is really missing a concept that described 'downregulation of signaling by receptor degradation" (it is also possible that the concept exists but I don't know what it's called). Any suggestions @ukemi @vanaukenk @RLovering @bmeldal ?

Thanks, Pascale

[NancyCampbell]

@bmeldal @ukemi @vanaukenk @RLovering

Hi @pgaudet

Thank you for the suggestion that 'downregulation of signaling by receptor degradation' might be missing.

However, in this particular case i.e. the PCSK9-ANX2 complex and the LDL receptor, this is not a signalling pathway. So 'negative regulation of receptor degradation' is appropriate though I can't find this in QuickGO. So maybe I should request this term? (or better suggestion).

Thanks Nancy

[ValWood]

Isn't this a little bit similar to the existing term:

"regulation of transcription by transcription factor catabolism"

there is talk of obsoleting this term. So, it seems odd to add new similar ones in another context. We should probably decide how to represent this type of process regulation more generally before the new terms are created.

[bmeldal]

@ValWood are we just comparing principle here, now? Nancy's complex is a regulator of a receptor, not a TF. (although receptors can become TFs...) It acts on the synthesised protein, not at the transcription or translation level.

[ValWood]

Ontology principle. Regulation of some process by protein catabolism.

[bmeldal]

Ok, I see what you mean, Val.

Does that mean it needs an editors' discussion?

[pgaudet]

Hi @ValWood @bmeldal I don't yet know what the best solution is; @ValWood I think the terms of type 'x involved in y' were being discussed for obsoletion (in #14138 but there are probably other issues). Examples of these terms are 'cyclin-dependent protein serine/threonine kinase activity involved in negative regulation of conjugation with cellular fusion' where we don't know how the MF fits in the BP.

This case here is a mechanism of regulation: perhaps it would be useful to have:

• regulation of MF by degradation of the target
• regulation of MF by sequestration of the target
• regulation of MF by post translational modification of the target
• regulation of MF by oligomerization/complex dissociation (we have this in the GPCRs and @pmasson55 showed me an example yesterday where the activate form of a receptor was a dimer that produced after ligand binding). ... etc

This seems to come up again and again as somthing we are unable to express, and even in GO-CAM models we cannot really do it.

Thoughts ? @thomaspd @ukemi and others I'd like your input as well :)

Thanks, Pascale

[bmeldal]

regulation of MF by oligomerization/complex dissociation (we have this in the GPCRs and @pmasson55 showed me an example yesterday where the activate form of a receptor was a dimer that produced after ligand binding).

It is really common for receptors to not oligomerise until the ligand binds. Lots of examples in the CP now, e.g. PDGFs (and more to come). But I would annotate directly with the MF of the receptor activity and the BP of the signaling pathway.

[pgaudet]

@sylvainpoux dealt with the Uniprot annotation

@hdrabkin can you please have a look at MGI:MGI:97599 protein kinase C, epsilon?

Thanks, Pascale

[hdrabkin]

I don't see anything wrong with this annotation. The pkce is activating or increasing the activity of the gaba receptor.

[suzialeksander]

ontology was refactored, UniProt and MGI annotations were reviewed. Further action should be a new ticket