checkpoint response phenotypes

[fypoadmin]

Another cell cycle-related overhaul, sort of inspired by [fission-yeast-phenotype:#2155]:

Looking at GO after the latest cell cycle overhaul, I think I ought to move phenotypes affecting responses to cell cycle checkpoint signaling out from under "cell cycle checkpoint phenotype".

So, does this plan sound ok?

• create generic "abnormal response to cell cycle checkpoint signaling" (or specifically mitotic, GO:0072396 or GO:0072414 depending on specificity), and put it under "cell cycle phenotype" and "cellular response phenotype"
• move the more specific existing terms under the new "checkpoint response" ancestor as appropriate, including anything to do with "checkpoint override"

Original comment by: mah11

[fypoadmin]

I think in GO that the "response to checkpoint signalling " has a slightly different meaning to that proposed here.

An example would be in the "DNA damage checkpoint" pathway. The DNA damage checkpoint pathway response are the normal processes which occur as a result of the checkpoint signal (i.e. are positively regulated by the checkpoint). The response would be the "DNA repair pathway"

The rationale for this separation was that you wouldn't not want all of the DNA repair genes to become annotated to the checkpoint itself.

To have genes which are upstream of the checkpoint signalling, and cause a problem which eventually means that the checkpoint is over-ridden would be a different usage from the GO response to checkpoint terms.

Is there a problem with the definitions of the "response to checkpoint" terms that could make their purpose clearer?

Original comment by: ValWood

[fypoadmin]

I'm not sure I understand what you're getting at. The purpose of the changes I'm proposing here is to make the FYPO structure more consistent with the current GO structure. It's all part of the same general approach I use throughout FYPO — it should mirror the ontologies it uses in logical defs. Specifically in the area of checkpoint responses, there are a few bits in FYPO that haven't caught up with the latest GO checkpoint revisions, and that's what I want to address here.

GO defines what happens in response to checkpoint x signaling. If that response doesn't happen the same way in a mutant as it does in wild type, why should that not be described as an abnormal response?

And if a checkpoint is overridden, surely that qualifies not responding normally to the checkpoint?

My best guess about your comment is that I think this just flags up yet another situation where a mutation causing phenotype x doesn't provide justification for annotating the gene to the GO term used in the phenotype x logical definition. There's plenty of precedent for that.

Original comment by: mah11

[fypoadmin]

Yeah I think I know why it seems odd. The community uses the "presence of Mad2 at the spindle pole body" as a 'marker' to indicate that the spindle checkpoint is activated.

I am not convinced that this is a single pathway, so it might be that some branches are activated and some now. Anyway won't worry about this now. Might need to revisit all of the "spindle checkpoint activated" annotations at some point or migrate them to a different term.... can deal with that when it happens.

Something is a bit weird with the biology, but you are right the logic here is fine.

Original comment by: ValWood

[fypoadmin]

• status: open --> closed-fixed

Original comment by: mah11

[fypoadmin]

have FYPO:0002517 'abnormal response to mitotic cell cycle checkpoint signaling', so adjusted its links

decided a more generic term can wait until we need it (would be for meiotic cell cycle stuff)

adjusted a few more links involving is_a descendants of FYPO:0002517

Original comment by: mah11