A number of ontologies are starting to use "shortcut" relations. These are expanded to more complex class expressions prior to OWL reasoning:
http://precedings.nature.com/documents/5292/version/1 http://berkeleybop.org/~cjm/obo2owl/obo-syntax.html#7 On May 4, 2011, at 9:00 AM, Mark Wilkinson wrote: > Heya, > > My personal take on this is that it becomes a trade-off. More granular > predicates generally means that you are creating less descriptive Classes > (i.e. that the Class does not have a lot of class-defining properties). So > while more descriptive predicates are good for SPARQL querying, they are less > good for DL reasoning (class reasoning is ~more powerful than predicate > reasoning) > > That's a superficial view, but it's something that we have also been > struggling with. We've tended to try bridging the two approaches by defining > elaborate classes with "basic" predicates, and then minting new, more > descriptive predicates that join these classes as well. > > Don't know if I am explaining myself clearly :-/ > > M > > > > On Wed, 04 May 2011 08:42:21 -0700, James Malone <mal...@ebi.ac.uk> wrote: > >> Hi Scott, All, >> >> I was wondering what the general take is on predicates in RDF >> representations used by the HCLS group. I've been looking at our RDF model >> for Gene Expression Atlas at EBI and presently I'm using the same "is_about" >> relation for a lot of the predicates as this is the lowest level of >> constraint from the OBO Foundry folks for some of these information >> relations. Alan Ruttenberg tells me that empirical evidence suggests that >> using a larger number of relationships correlates to poorer ontologies. >> However, I've also been told from various RDF advocates that having more >> granular level predicates is useful for querying. Are there any thoughts >> from the group on this? I have no preconceptions here (I have no reason to >> disbelieve Alan or the RDF folks) so open to thoughts and suggestions. >> >> Cheers, >> >> James >> > > > -- > Dr. Mark Wilkinson > Assistant Professor, Medical Genetics > PI Bioinformatics, Institute for Heart and Lung Health > St. Paul's Hospital/UBC > Vancouver, BC, Canada >
