[Ontology-editors] [Software-group] Regulates transitivity

[Chris Mungall]

On Dec 5, 2008, at 2:35 PM, Stoddard, Alexander wrote:

> I have more questions for clarification (probably from Chris).
>
> I am looking at the geneontology wiki page:
> http://wiki.geneontology.org/index.php/Relation_composition
>
> If I am understanding things correctly then regulates is not  
> transitive over itself (unlike is_a and part_of).
>
> In the simplest case:
> A regulates B regulates C
>
> A has no currently named relationship to C.

This is correct. We considered introducing "indirectly_regulates".  
This would never be asserted in the GO, but it would be inferred.  
Someone pointed out that it may have been more intuitive to call the  
existing relations directly_{+/-}_regulates.

>  If protein p is annotated to A then a search for C should not  
> return p. (Perhaps a tool could allow a user the option to return  
> indirect regulators.)
>
>
> Looking at the page in general http://wiki.geneontology.org/index.php/Relation_composition
> there was a confusing piece of symbology; R is used as a symbol for  
> any regulation relationship but R1, R2, R3 etc. are used as symbols  
> for any generic relationship, perhaps Rel1, Rel2, Rel3 would be less  
> easily confused with previous discussion on the page.

Good point. It may be better to go the other way, and fully spell out  
the relation name

>
> Thank you and kind regards,
> Alex Stoddard
> RGD Developer
>
> _______________________________________________
> Software-group mailing list
> Software-group at geneontology.org
> http://fafner.stanford.edu/mailman/listinfo/software-group