Re: OWL vs RDF

[William Bug]

I completely agree with Peter's point. As he describes, the only way we'll be able to stay commensurate with the multiple neuroanatomical frameworks in use in BIRN (not to mention in the community at-large) is to start by deconvolving the sometimes contradictory ontological "senses" these resources have pre-coordinated into one, and then work to use various OWL mechanisms for axiom construction to construct these more complex graphs via a post-coordinated mechanism. The only addition I'd make to Peter's point is the community is still working out what the best formal means for specifying "equivalence relationships" will be.  In addition to the standard class axioms (subClassOf, Complement, Union, Intersection, etc.) and sameAs at the instance level, it's clear some of the various ontology merging requirements will need to use mechanisms beyond this set of options.  More info on this effort can be obtained from Chris Mungall and/or Barry Smith from NCBO - or the Wikis NCBO is maintaining on these issues.  There has been a lot of discussion on this topic both during and in the wake of the recent Common Anatomy Reference Ontology meeting in Seattle, as well as in the context of the major refactoring of the OBI (nee FuGO) community ontology and the BIRN BIRNLex ontology. Cheers, Bill On Oct 27, 2006, at 5:18 AM, Mork, Peter D.S. wrote: I think that even if you are talking about ontology construction, in most cases you need OWL.  It is certain that multiple overlapping ontological structures will be developed.  For example, in BIRN, there is a core ontology describing neuron-anatomical structures.  However, each brain atlas introduces its own set of structures (based, in part, on the available resolution).  One could try to merge these into a single ontology, but that would be poor engineering.  Instead, a new set of axioms can be constructed that define atlas structures in terms of the core BIRN ontology (and vice versa).  This articulation requires class constructors and equivalence relationships.  Moreover, if the source structures are expressed using DL definitions, the inference engine will often be able to identify additional articulation conclusions.   Peter Mork   As Phil implies below, if you really just need a formalism for nodes and edges, you don't NEED OWL - in fact, for many types of graphs, you don't need RDF.  On the other hand, if you are clearly going to require a formalism with considerable ontological expressivity, you probably want to give the OWL dialects (and their underlying DLs and the toolsets such as ProtegeOWL  and the Pellet, FACT++, and other reasoners, etc.) serious and in depth consideration.   Just to be clear, I'm talking here about ontology construction which I consider just a portion of the required task of semantically formal data representation.  For much of the semantic representation we need to do in large scale biological data repositories, RDF alone will clearly be a sufficient first step, so long as we continue to develop effective means of expressing the triplets in the context of the ontologies and extending the ontologies via analysis (as automatic as is feasible) of the triplet repositories.   Cheers, Bill     Bill Bug Senior Research Analyst/Ontological Engineer Laboratory for Bioimaging  & Anatomical Informatics www.neuroterrain.org Department of Neurobiology & Anatomy Drexel University College of Medicine 2900 Queen Lane Philadelphia, PA    19129 215 991 8430 (ph) 610 457 0443 (mobile) 215 843 9367 (fax) Please Note: I now have a new email - [EMAIL PROTECTED]