[hcls] Mappings of neuroscience ontologies and their advantages

I have started (yet another) wiki page describing the mappings of neuroscience ontologies / data sources and their potential benefits:

http://esw.w3.org/topic/HCLS/HCLSIG_NeuroscienceOntologyMappings

If you are working on a mapping yourself, please add a description to this wiki page. 
The possible targets for further mappings are plentiful, which, unfortunately, also means that many of the existing ontologies we have been working on so far have made little effort of reusing existing ontological resources. This is probably caused by the fact that most of the ontologies were derived from existing databases via an automated export. 
While doing mappings through OWL constructs such as owl:equivalentClass solves the problem theoretically, many of our existing datastores do not really support such reasoning (and if they do, this is probably associated with a loss in performance). It would be great if we would have a tool that can REPLACE the URIs in different ontologies with a canonical URI, based on our mappings anda priorization of different namespaces (e.g. if something is the equivalentClass of a class from an OBO ontology, replace the URI with the URI from OBO and remove the equivalentClass statement afterwards). This would enable us to make the mappings understandable to applications that have no extensive reasoning capabilites. Does anyone know if such a tool exists?


cheers,
Matthias Samwald





.
-- 
"Feel free" - 10 GB Mailbox, 100 FreeSMS/Monat ...
Jetzt GMX TopMail testen: http://www.gmx.net/de/go/topmail

Received on Monday, 9 April 2007 14:51:49 UTC