- From: Alan Ruttenberg <alanruttenberg@gmail.com>
- Date: Fri, 2 Mar 2007 15:37:00 -0500
- To: Kei Cheung <kei.cheung@yale.edu>
- Cc: HCLS <public-semweb-lifesci@w3.org>
On Mar 2, 2007, at 1:59 PM, Kei Cheung wrote: > > On that page, it describes the following OWL axiom example: > > *EquivalentClass*(CerebellarPurkinjeCell *intersectionOf* > (PurkinjeNeuron, *Restriction*(locatedInBrainRegion *allValuesFrom* > (Cerebellum))) > > We can use a similar axiom to define the localization of a certain > type of neuron cell type (CA1 pyramidal neuron) in a particular > brain region (e.g., hippocampus). Due to my limited knowledge of > OWL, can this type of axioms/rules be parameterized (e.g., cell > type and brain region can be treated as variables)? Not sure what you mean by parameterized here. Certainly we could, from a spreadsheet encoding this information, generate the appropriate classes. Also, in LSW or (dig) one can easily construct one of these classes on the fly and ask what the equivalent, subclasses, and superclasses are. To find all cells located in, e.g. Cerebellum you would ask for subclasses of intersectionOf*(Cell, *Restriction*(locatedInBrainRegion *allValuesFrom*(Cerebellum)) Don't think I'd use locatedInBrainRegion though, probably just locatedIn, or partOf, or perhaps derivedFrom some PartOf if the cells were no longer in the brain (i.e. in a dish). -Alan
Received on Friday, 2 March 2007 20:37:51 UTC