relationship: immediately_preceded_by (or preceded_by)

[ANiknejad]

Hi Chris,

I still do not understand the way we have to use these 2 different relationships in developmental ontologies: is there related to the structure of the ontology (same level or allowed between levels?) or only based on the temporal relation? I tried to check here http://semanticscience.org/resource/SIO_000251 http://semanticscience.org/resource/SIO_000249

Please let me know

Thank you,

Anne

PS: If you have time...see below as well ;-)

What I understand so far is that if stages are 'temporally precise' and directly one after the other, so we use 'immediately_preceded_by'. If the stage is a grouping stage or having a large temporal definition (with or without children) then we can not use 'immediately_preceded_by' to link to any other stage (but 'preceded_by' is fine). Just as a precise stage can not be related by 'immediately_preceded_by' to a grouping and large defined stage (but 'preceded_by' is fine)...hum... Here below a schematic example, could you check and comment, please

-A
--a
--b
-B
--c
--d
-C
-D

a, b (respectively c, d) are children of A (respectively B), occurring successively so: a is the first children of A, NO 'preceded_by' neither 'immediately_preceded_by' b, immediately_preceded_by a c, immediately_preceded_by b (not the same level) d, immediately_preceded_by c C, immediately_preceded_by d (not the same level) AND preceded_by B (B and C are at the same level) D, immediately_preceded_by C, OR preceded_by C

[cmungall]

almost. it's not about levels, it's about whether the starts and ends coincide.

preceded by is the transitive form. So d is preceded by a, but not immediately (they do not abut)

ideally we would only use the precise form, but sometimes we need vagueness.

For example, we can say that a fully formed is necessarily preceded by an embryo stage, but we can't say for sure whether this is immediate in all species whether this holds. Yes in humans, no in organisms with larval, pupal etc forms.

[cmungall]

See also http://purl.obolibrary.org/obo/RO_0002087

[ANiknejad]

Thank you Chris, much more clear, yes I will check the starts and ends to put the precise form when possible.

[ANiknejad]

another question, sorry: possible to have more than one sentence like this one in the same file? relationship: immediately_preceded_by BtauDv:0000003 ! immature stage

cause here we have:

-immature (no child but precise start/end) -mature --adulthood ....

so mature and adulthood, both start at end of immature (end_ypb "1.5") (mature does not have a end time, while adulthood has). Could I put the relation/sentence to both mature and adulthood?

[cmungall]

Yes, this is valid

Note that one of statements becomes redundant once you start using more precise parthood relations - e.g. starts_with and ends_with. E.g if adulthood starts_with mature stage, and mature stage is immediately preceded by immature, then by a property chain we have adulthood IPB immature.

But don't worry about that for now, no harm in accidentally stating a redundancy.

The main drawback is if you're doing some kind of graph visualization, having statements at 'different levels' could be unaesthetic. But there's ways around this.

[cmungall]

But as an aside we should figure out a strategy for reusing existing classes rather than restating the generic stages in every Dv ontology...