results_in_division_of - definition

[rachhuntley]

Needs formal definition.

Domain imported twice

[dosumis]

MD doc version under discussion

Notes

• Limited use in annotation extension. All by MGI or SWIS. All with cell as object.
• Not used in ontology. A similar sounding relation results_in_fission_of is used, but only for 3 classes:
  • mitochondrial fission
  • nuclear division
  • organelle fission
• cell division classes in the ontology follow the pattern: cell division that has_input some 'X cell'. If this is sufficient, we should stick with it for AE as well. If not, we should improve the definition of this relation and use it in the ontology.

@ukemi Please comment.

[ukemi]

I think this is basically ok, but the doc is a little misleading with the fibroblast example. Does the division necessarily result in two identical cells? What about an asymmetric division?

I also think we need to be careful in the cell proliferation realm. Since proliferation is almost always measured as the number of cells increasing, there are two factors in play; creation of new cells and death of old cells. A population of cells can be said to proliferate by keeping cell division at a constant and decreasing the amount of apoptosis.

[dosumis]

Hi David,

Agree that the current definition is not great (no idea who wrote it). I think the key question is whether we should have this relation at all. Is has_input sufficient? Or do we want to allow for other inputs to cell division? If has_input is not sufficient, should we use the more general 'results_in_fission_of'

Note - 'type' is slippery here. The two outputs of any cell division share a type in the they are both cells. I think the same applies to fission more generally (mitochondrial fission produces two mitochondria).

[dosumis]

Note - divergence in location of master for relations (see #46) is an issue here.

[dosumis]

One more wrinkle: at least some of the usage (including the examples on in the doc) use a regulates class. It would be more consistent with other cases if a regulates_o relation were used here.