[Ontology-editors] has_part - time frame for implementation?

Chris Mungall cjm at berkeleybop.org
Thu May 21 08:49:16 PDT 2009


On May 15, 2009, at 7:52 AM, Jane Lomax wrote:

> I don't see why we couldn't start using them now if we restricted  
> them to the _write.obo and _ext.obo files.
>
> has_part is defined in RO, so that's not a problem. We could give a  
> couple of weeks announcement if we thought we needed to, but the  
> whole point of the editors files is to be able to add stuff like this?

We should probably have some kind of process to go through

* The relation should be well defined some, and the specific impact  
and rationale for GO should be documented. Everyone on this list  
should be aware of the meaning of the relation, esp. eg. why has_part  
is not a simple type level inverse of part_of

* Should be tested in OBO-Edit just to make sure there is nothing  
unusual introduced. Introducing has_part will likely introduce cycles  
over the unlabeled graph. Ontologically this is fine but parts of OE  
may be built with the assumption that the unlabeled graph is acyclic.  
Of course, the labeled graph has no has_part cycles.

For has_part I did this a while ago:

http://wiki.geneontology.org/index.php/Has_part

but this is out of date as it is more likely we will use this relation  
in CC or in BP rather than BP->MF

>
> Jane
>
>
>
> Midori Harris wrote:
>> Hi Chris et al,
>>
>> A recent SourceForge item on spliceosomes has once again  
>> highlighted how useful the has_part relationship will be in the CC  
>> ontology:
>>
>> https://sourceforge.net/tracker/?func=detail&aid=2784316&group_id=36855&atid=440764
>>
>> There are plenty of similar situations in CC, and I've found some  
>> things in BP that could use has_part. I see some usage in a few of  
>> the xp files too.
>>
>> So, do we have any idea when we might be able to use has_part in  
>> GO? Would it help to strip has_part links from everything except  
>> _write.obo and _ext.obo? Karen and I would be delighted to be able  
>> to do the spliceosome terms without having to add a bunch of  
>> awkward subcomplex-when-it's-part-of-this-bigger-complex terms.
>>
>> Thanks,
>> midori
>> _______________________________________________
>> Ontology-editors mailing list
>> Ontology-editors at geneontology.org
>> http://fafner.stanford.edu/mailman/listinfo/ontology-editors
>
>
> -- 
> Dr Jane Lomax
> GO Editorial Office
> EMBL-EBI
> Wellcome Trust Genome Campus
> Hinxton
> Cambridgeshire, UK
> CB10 1SD
>
> p: +44 1223 492516
> f: +44 1223 494468
>
>



More information about the Ontology-editors mailing list