[Ontology-editors] CC organization question for xps

[Midori Harris]

Hi Tanya,

Thanks for chiming in ... keeps us from haring off into weirdness.

I'm actually entirely sympathetic to what you've said here ... but I can't 
shake the feeling that there's a catch somewhere.

It seems like everyone's happy with the "all cellular components are 
anatomical structures" part of the equation.

If that's true, though, then it follows that formation (= assembly) of a 
CC is a type of anatomical structure formation, doesn't it? I haven't 
figured out how to explain why we wouldn't add the relationship, based on 
the definitions we have.

But then we run into the path-to-development problem. Do we have to 
reconsider the relationship anatomical structure formation is_a 
developmental process? Perhaps not every instance of anatomical structure 
formation is an instance of a developmental process -- especially for the 
microscopic anatomical structures otherwise known as cellular components. 
New child terms for anatomical structure formation???

m

On Mon, 17 Nov 2008, Tanya Berardini wrote:

> Chiming in REALLY late here...
>
> On Mon, Nov 17, 2008 at 2:51 AM, Midori Harris <midori at ebi.ac.uk> wrote:
>
>> Hi all,
>>
>> I have a suggestion and question about the CC organization terms that are
>> relevant to XPs ...
>>
>> Several of us agreed recently that all cellular components are anatomical
>> structures.
>
>
> I can understand this.
>
>
>> For the xp set I'm looking at -- biological_process_xp_cellular_component.*
>> -- I can put in more specific genus terms for several xp defs if I add the
>> relationship:
>>
>> GO:0022607 cellular component assembly is_a GO:0048646 anatomical structure
>> formation
>
>
> This disturbs me a little bit exactly for the reason you point out below,
> the resulting path up to 'developmental process'
>
> "but some users may think it looks odd if they've been accustomed to
> thinking of 'development' having a narrower meaning."
>
> I am reluctant to redefine what 'development' means for the broader
> community by making this parentage.  Here are some specific children of
> 'cellular component assembly' that make me hesistate:
>
> 'actin cortical patch assembly'
> 'chromatin assembly'
> 'macromolecular complex assembly'
>
> There are others as well, but these should serve to illustrate cases where
> '% assembly' is not part of a progression from an initial condition to a
> later condition.  I really don't think that anyone would consider 'protein
> complex assembly' a developmental process even if GO's definition is broad
> enough to encompass this term.
>
>> Note that we already have GO:0032989 cellular structure morphogenesis is_a
> GO:0009653 anatomical structure morphogenesis.
>
> That parentage makes sense to me and I think it should stay.
>
> What happens if we don't make the link between 'cellular component assembly'
> and 'anatomical structure formation'?
>
> It still makes sense to add  cellular component formation as an exact
> synonym of cellular component assembly.
>
> I'm not so sure now whether we still would want to merge GO:0016043
> (cellular component organization) and GO:0032989 (cellular structure
> morphogenesis) and add the relationship GO:0022411 (cellular component
> disassembly) is_a GO:0060033 (anatomical structure regression)?
>
> There's my monkey wrench.  Let's discuss.
>
> Tanya
>
>
>>
>> It might also be good (though less pressing for my xps) to add a parent in
>> the anatomical structure branch of BP for GO:0016043 cellular component
>> organization. I'm not sure whether GO:0048532 anatomical structure
>> arrangement exactly fits the bill, though. The definitions don't quite match
>> up, in that GO:0048532 is restricted to "the configuration of the
>> constituent parts of an anatomical structure" and so covers only the
>> "arrangement of constituent parts" part of the GO:0016043 def, not the
>> formation or disassembly. (I also don't quite understand how GO:0048532 can
>> have 'structural organization' as an exact synonym when the more restricted
>> phrase 'anatomical structure organization' is a broad synonym.)
>>
>> GO:0022607 (and GO:0016043, if we decide on an anatomical structure process
>> parent for it) will acquire a path back to GO:0032502 developmental process.
>> The definition GO uses for developmental process is broad enough to be
>> consistent with the new paths, and it would be entirely consistent with our
>> thinking about the CC organization terms, but some users may think it looks
>> odd if they've been accustomed to thinking of 'development' having a
>> narrower meaning. I can live with that, but we should be prepared for
>> questions.
>>
>> Note that we already have GO:0032989 cellular structure morphogenesis is_a
>> GO:0009653 anatomical structure morphogenesis.
>>
>> Shall I go ahead and add GO:0022607 is_a GO:0048646?
>>
>> m
>> _______________________________________________
>> Ontology-editors mailing list
>> Ontology-editors at geneontology.org
>> http://fafner.stanford.edu/mailman/listinfo/ontology-editors
>>
>
>
>
>