[Ontology-editors] haspart documentation (meeting with Chris (fwd))

[Alexander Diehl]

I agree with your restatement (hadn't thought deeply about has_part).  
We shouldn't let use of the terms 'child' and 'parent', which worked 
well for the original GO with its is_a and part_of relationships burden 
our thinking about other relationship types, particularly as the GO is 
rebuilt with logical definitions based on relationships to other terms 
both within and external to the GO.


Chris Mungall wrote:
>
> On Jun 16, 2009, at 10:05 AM, Alexander Diehl wrote:
>
>> I have to agree with Chris.  Graphical conventions should always be 
>> readable in the same direction with term1 (the child) having a 
>> relationship to term2 (the parent), so that the combination can be 
>> turned directly into an English sentence.
>>
>> T cell is_a lymphocyte
>> T cell --is_a--> lymphocyte
>>
>> I think the problem here is that the whole idea of child and parent 
>> does not apply to non-transitive relations.
>
> I would say "does not apply to many relations, including 
> non-transitive relations"
>
> (has_part is transitive)
>
>>  After all, many non-transitive relationships will be between 
>> different ontologies:
>>
>> T cell differentiation has_participant T cell.
>> T cell differentiation ---has_participant-->  T cell
>>
>> T cell has_part T cell receptor
>> T cell ---has_part--> T cell receptor
>>
>> Thanks,
>>
>> Alex
>>
>>
>> Jennifer Deegan (nee Clark) wrote:
>>> Is there a known correct use of parent/child for has_part? It seems 
>>> to me that has_part is bound to be a bit tricky, but that if there 
>>> is a correct usage then it would be best if we find out what it is 
>>> and try to stick with it from day 1.
>>>
>>> Jen
>>>
>>> Amelia Ireland wrote:
>>>>
>>>> On Jun 16, 2009, at 7:08 AM, Chris Mungall wrote:
>>>>
>>>>>
>>>>> Unfortunately, the whole 'parent/child' terminology becomes 
>>>>> confusing with has_part
>>>>> http://www.ebi.ac.uk/~aji/go/GO.ontology.relationships.shtml#haspart
>>>>>
>>>>> If we take 'child' to mean the subject of the assertion, and 
>>>>> 'parent' to mean the target/object of the assertion, then in
>>>>>    chromosome has_part chromatin
>>>>>
>>>>> chromosome is the child and chromatin is the parent
>>>>>
>>>>> This usage is consistent with our graphical metaphors, where we 
>>>>> always have child ---> parent
>>>>>
>>>>> Of course, this conflicts with the intuition we have drummed into 
>>>>> people after 10 years, where 'child' is the smaller and 'parent' 
>>>>> is the larger.
>>>>
>>>>
>>>> I would argue that your average everyday mortal (e.g. me) considers 
>>>> the parent to be the broader term (closer to the root node) and the 
>>>> child to be the more specific term. If I've used the terms 'child' 
>>>> and 'parent' in the opposite way somewhere, it's accidental, and 
>>>> due to a copy/paste error!
>>>>
>>>> Perhaps I'll add something at the top about the nomenclature 
>>>> conventions used in these docs so it's not so confusing.
>>>>
>>>> -- 
>>>> Amelia Ireland
>>>> GO Editorial Office
>>>> http://www.berkeleybop.org || http://www.ebi.ac.uk
>>>> BBOP Plant Project: http://bbopgarden.blogspot.com
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> Ontology-editors mailing list
>>>> Ontology-editors at geneontology.org
>>>> http://fafner.stanford.edu/mailman/listinfo/ontology-editors
>>> _______________________________________________
>>> Ontology-editors mailing list
>>> Ontology-editors at geneontology.org
>>> http://fafner.stanford.edu/mailman/listinfo/ontology-editors
>>
>>
>> -- 
>> Alexander D. Diehl, Ph.D.
>> Senior Scientific Curator
>> Mouse Genome Informatics
>> The Jackson Laboratory
>> 600 Main Street
>> Bar Harbor, ME  04609
>>
>> email:  adiehl at informatics.jax.org
>> work:  +1 (207) 288-6427
>> fax:  +1 (207) 288-6131
>>
>>
>


-- 
Alexander D. Diehl, Ph.D.
Senior Scientific Curator
Mouse Genome Informatics
The Jackson Laboratory
600 Main Street
Bar Harbor, ME  04609

email:  adiehl at informatics.jax.org
work:  +1 (207) 288-6427
fax:  +1 (207) 288-6131