Menu
▾
▴
Re: [Chebi-ontology] Oxygen-18 to isotope: instance or subclass?
|
From: Alan R. <ala...@gm...> - 2014-09-01 22:18:08
|
On Mon, Sep 1, 2014 at 11:45 AM, Emw <emw...@gm...> wrote: > Barry, Alan, > > Thank you for clarifying that the statement > > oxygen-18 > *subclass of* oxygen > *instance of* isotope > > is not compatible with BFO. > > I find Barry's statement that "it must be incorrect to assert that > oxygen-18 is an instance of anything" intriguing. Per Barry's rationale, > given entities Foo and Bar, if > > a) Foo has instances > b) Every instance of Foo is also an instance of Bar > > then it is not compatible with BFO to state "Foo *instance of *Bar". > That is what BFO says (approximately). I don't see how you come to the conclusion that it is not compatible with BFO to say so. To be more precise, however, BFO would phrase this in terms of universals, particulars and the instance of at t relation that BFO defines. > Does that mean BFO prohibits class-instance metamodeling, i.e. punning? > BFO doesn't prohibit anything. It is a framework in which some kinds of things are defined and others not. The attempt is to be reasonably careful about defining, where possible, what is there. In the framework of BFO, universals are not instances of anything. The the instanceOf relationship is, by definition, between universals and particulars. Metamodeling and punning are not within the scope of BFO, which talks about entities in the world, not about *how* to represent them. I understand metamodeling, as you use it, to be a way of modeling in OWL. Punning is an OWL-centric concept. In the view of BFO as an ontology, representation systems such as OWL serve the need to create computational systems that support the ontology and valid inferences one can make. The goal of the ontology is not to make use of OWL - that is backwards. > If so, then that would preserve the foundational definition of "instance" > in Smith and Rosse 2004 [1]: "Instances are individuals (particulars, > tokens) of special sorts... bound to a specific... location in space and > time." However, it would also prevent BFO ontologies from using a major > feature in OWL 2 DL. > See above. It isn't a goal of BFO to use major features of OWL. However, as a matter of fact, there are ways in which OWL representations of BFO-based ontologies might make use punning. For example there are other relations that relate particulars to universals. An example are the "lacks" relations. If something doesn't have a part, there is no instance of the part to relate as missing. So one representation of a "lacks" relation is to relate the particular to the type of the thing that is lacking, and for this one might use OWL's punning. > If not, then that entails broadening the definition of "instance" in BFO > to account for classes being instances of metaclasses. > If such a think is desirable, it would be motivated by ontological issues, not OWL. There *is* motivation in some cases, although it isn't clear to me that isotope as metaclass is one of those. To determine so would be to have both a theory of what it means for a universal to be an instance of something, and justification that isotope fits within that theory. We don't have the former (though there is work in this direction) and I don't see a clear basis for the latter - certainly there is not enough information that specifies what it would mean. > It would clarify that statements like those in the canonical OWL 2 DL > examples and UniProt are compatible with BFO. > Much of Uniprot's RDF representation is not compatible with BFO. In the framework of BFO, instances of proteins are physical molecules. Certainly when Uniprot uses an OWL instance that is not what they mean. You are free to use BFO or not. A prerequisite would be having a clear understanding of BFO, and I don't see that you do. I think if you did you would better understand why certain things are said. To the extent that I can, I've responded in ways that try to clarify that. I will continue to do so. I think there are untapped abilities for OWL to help represent BFO-based ontologies. I do some work on that myself, and I hope that work to be part of BFO 2. I don't think what I'm working on, however, will solve the issue of representation of isotopes. For that I think you first need to have a solid definition of what an isotope is. I offered a start of one from the perspective of BFO in my last email message. I think it would be good for you to offer an alternative. Simply saying the isotope is a metaclass of Oxygen-18 isn't an adequate definition. -Alan > My message from 2014-08-04 [2] discusses this in more detail. > > Thanks, > Eric > > [1] Barry Smith, Cornelius Rosse (2004). The Role of Foundational > Relations in the Alignment of Biomedical Ontologies. > http://ontology.buffalo.edu/medo/isa.pdf > [2] http://sourceforge.net/p/chebi/mailman/message/32679967/ > > > On Fri, Aug 29, 2014 at 4:45 PM, Barry Smith <phi...@bu...> wrote: > >> Given what Alan says, I infer that it must be incorrect to assert that >> oxygen-18 is an instance of anything. Certainly there are instances of >> oxygen-18, but oxygen-18 itself is for that very reason not itself an >> instance >> BS >> >> >> On Fri, Aug 29, 2014 at 4:36 PM, Alan Ruttenberg < >> ala...@gm...> wrote: >> >>> >>> >>> >>> On Fri, Aug 29, 2014 at 10:44 AM, Barry Smith <phi...@bu...> >>> wrote: >>> >>>> Re: the issue raised by: >>>> >>>> oxygen-18 >>>> * subclass of* oxygen >>>> *instance of* isotope >>>> >>>> can I ask that people help me get clear about the background: >>>> >>>> (a) is it the case that all of these expressions refer to general >>>> entities which have many many particular instances >>>> >>> >>> yes >>> >>> (b) is it the case that every instance of oxygen-18 is also an >>>> instance of isotope >>>> >>> >>> to be an isotope is something like to be a mutation. To be a mutation is >>> to be slightly different from the canonical, but of equal status except for >>> perhaps the number of class members. To be an instance of an isotope of >>> oxygen (atom) is to be a an atom with a different number of nucleons >>> (protons and neutrons) than the most abundant form (oxygen-16, with 8 >>> protons and 8 neutrons in its nucleus). >>> >>> One might argue that the number of nucleons is a quality of anything >>> oxygen, of which I list a variety of entities that ChEBI represents in an >>> earlier message. >>> >>> >>>> (c) if yes, then why not just assert 'subclass of' throughout? >>>> >>> >>> You can, just as you can say that red balls are a subclass of colored >>> things ;-) >>> >>> -Alan >>> >>> >>>> >>>> >>>> BS >>>> >>>> >>>> >>>> ------------------------------------------------------------------------------ >>>> Slashdot TV. >>>> Video for Nerds. Stuff that matters. >>>> http://tv.slashdot.org/ >>>> _______________________________________________ >>>> Chebi-ontology mailing list >>>> Che...@li... >>>> https://lists.sourceforge.net/lists/listinfo/chebi-ontology >>>> >>>> >>> >> >> >> ------------------------------------------------------------------------------ >> Slashdot TV. >> Video for Nerds. Stuff that matters. >> http://tv.slashdot.org/ >> _______________________________________________ >> Chebi-ontology mailing list >> Che...@li... >> https://lists.sourceforge.net/lists/listinfo/chebi-ontology >> >> > |
