Hi Nina,

On Wed, Oct 28, 2009 at 2:45 PM, Nina Jeliazkova <nina at acad.bg> wrote:
> - Do we need to make distinction between different e.g. XLogP
> implementations (I would say yes) ?

I think so too.

> Is it possible to handle this via BO ontology, or we need an extension?

The ontology I am working on supports this.

> - What would be the best way to extend BO ontology (this is more a
> question to Egon)?

By simply adding classes and axioms. But the above is covered...

> - How would we handle quantities, defined in existing data sets (e.g.
> all LogP flavours available in EPA DSSTOX), not calculated via OpenTox,
> or an user uploaded dataset.
> - How to handle quantities, calculated via some algorithm, but with
> different parameters (e.g. eHOMO calculated with AM1 or PM3).
>
> I would prefer that the property (e.g. blueobelisk:xlogp) refer to a
> specific implementation, rather to the algorithm itself  (same concept
> as algorithm/model split we already invented).
> The implementation itself will be linked to the algorithm.

Agreed; the ontology works like that.

> We would need a way to handle dynamically defined properties and even
> ontologies.  I am particularly thinking of user-defined datasets.

One advantage of RDF is that everyone can contribute... Bioclipse
allows people to 'add' descriptors by providing a number of RDF
triples...

>>   - It can help us to solve the problem of unique IDs, by using URIs
>>
> AFAIK, that will require an RDF store for the ontology service
> (centralised one?)  - am I right?   It would be good if three is a
> distributed solution.

RDF is highly distributed... the store can take many formats. A simple
file on a web server, a full fledged RDF triple store with SPARQL
frontend, or XHTML with the RDF embedded.

Egon

-- 
Post-doc @ Uppsala University
Homepage: http://egonw.github.com/
Blog: http://chem-bla-ics.blogspot.com/
PubList: http://www.citeulike.org/user/egonw/tag/papers