All gene product specific secretion related terms, block for direct annotation

[ValWood]

gene product specific, no annotations, suggest obsoletion?

GO:0038055 BMP secretion

GO:0110010 basolateral protein secretion
(3 annotation FlyBase) suggest
"secretion" occurs_at GO:0097575? or similar @hattrill ?

[RLovering]

This is not gene product specific, there are many paralogs in mammals

[ValWood]

...but since it has no annotation isn't this a case where you would know which molecule was being secreted. Anyway BMP secretion isn't really a process, the secreted molecule is the target of the secretion process.

I don't really understand how this term got in? Unless it has its own specific secretory pathway?

It's like having a term for say, say "BMP glycosylation" isn't it? I don't get it?

[hattrill]

I am looking at many signaling pathways at the moment - haven't got to BMP yet. But the biosynthesis and secretion pathways for these ligands is often highly specific e.g. TGFalpha-like EGFR ligands (e.g. rhomboid proteases, Star chaperone), wnt (porcupine, wntless). They are almost a part of the pathway and critical to regulation. Once I have finished my review of pathways (months!), I think want to propose that we embrace some of these as separate processes. For users, we need them to be able to access these with specific terms without stuffing them under the pathway term. Relations just don't cut it here.

[RLovering]

I think the term is there because it is needed for the regulation of BMP secretion. It is important to understand the genes regulating secretion of these growth factors. The regulation terms have annotations

[ValWood]

OK I'll close this one, but it sounds as though you are referring to the signalling component rather than the actual secretory pathway, which is proposed to be highly conserved from yeast to man.

[ValWood]

In this case should GO:0038055 BMP secretion be blocked dor direct annotations?

[RLovering]

probably ;)

[ValWood]

Action, gene product specific secretion terms. Block all for direct annotation. (regulation terms will be available).

Is that OK?

[RLovering]

Hi Val I think this is in line with a recent GOC meeting discussion (about biosynthesis I think), as you say the proteins involved in the secretory process itself are not specific for the secreted product. This does mean that we will be saying that 'regulation of secretion' ends with the 'cargo' being targeted to a secretory vesicle. The process of secretion starts with the cargo associated with the secretory vesicle. Does that sound right? or should the transport to the secretory vesicle be included in secretion? Can the parent term definitions be made more specific? Currently the parent (secretion and regulation of secretion) definitions have no start and end points.

[ValWood]

I think you can annotate a signalling pathway regulating a process to "regulation of that process". You can regulate a process (i.e frequency) via signalling. I don't think this means that regulation of a process ends with cargo selection.

However, the secretion branch is problematic. I think the meaning depends if you are referring to secretion by cell (which is covered by exocytosis), or secretion by tissue. I know nothing about the latter, but exocytosis from the def:

"Exocytosis begins with steps that prepare vesicles for fusion with the membrane (tethering and docking) and ends when molecules are secreted from the cell. PMID:22323285"

I think there are problems here because the "signalling" for cargo selection isn't regulating these steps is it?.....it's regulating something way-way upstream....

[RLovering]

good point. Sorry I don't have more time for this now.

[ValWood]

Will be addressed by

https://github.com/geneontology/go-ontology/issues/19115