Re: [Obi-coord] [Obi-devel] [OBI-devel] [OBI-coord] submission DDI to OBI?

[James M. <ma...@eb...>]

Hi Larissa,

Here is my tuppence :)

I think if you are willing, aligning you work so that it can logically 
import into OBI would be extremely useful. That would likely necessitate 
using some OBI classes and dropping some of your own, like you say some 
conflicts would need resolving. I don't think all of DDI should be 
turned over to OBI. OBI already struggles with person-power and with the 
very wide scope it has self-imposed upon itself and this would likely 
produce a large ontology with parts that are not relevant to a lot of 
people. I would personally rather see the subclasses that are DDI 
specific stay with you and your collaborators, with the above 
import-consistency caveat holding. Otherwise, any type of biomedical 
investigation 'belongs' to OBI by this thesis and I think this is not 
scalable. I always thought OBI should be a reasonably lightweight 
ontology from which others extend to build their application needs; we 
have design patterns which others then reuse.

Cheers,

James




On 29/11/2010 13:55, ls...@ab... wrote:
> Dear OBI coordinators and developers,
>
> We have recently submitted an ontology for drug discovery investigations
> (DDI) to the OBO Foundry as a candidate ontology. Some description of DDI
> and links are at the end of this message.
>
> DDi is currently under review. One suggestion from the reviewers was to
> make DDI a part of OBI. It is the OBO Foundry policy - do not create a new
> namespace unless it is completely justified. I believe that we could
> justify a necessity of having a separate ontology for drug discovery. But
> after careful consideration of the suggestion and consultations with our
> collaborators and potential users, we have agreed that donating DDI to OBI
> would be the best thing to do.
>
> DDI has been developed as an application of OBI to drug discovery. We
> believe that we can go by a similar route as the MGED community: donate
> native DDI classes to OBI and work on DD-view of OBI.
>
> There are some differences (i.e. 'hypothesis' and not 'hypothesis textual
> entity'), but we hope to overcome them.
>
> I would love to make a short presentation about DDI at one of devel or
> coord calls. But I am currently in Slovenia for two months and do not have
> proper facilities for teleconferences. Such a presentation has to wait
> till mid January.
>
> Meanwhile, I hope to have your opinion about integration of OBI and DDI.
>
> Thank you,
> Larisa Soldatova
> Aberystwyth University, UK
> behalf the DDI developers
>
> -------------------------------------
> DDI has been developed in partnership between the Computational Biology
> group, Aberystwyth University, Wales, UK and the Biological Chemistry
> and Drug Discovery group, the University of Dundee, UK. The project was
> funded
> by the BBSRC. The main outputs of the project, including DDI, are
> described in the paper listed below. The authors of the paper are the main
> developers of DDI:
>
> Qi, D., King, R.D., Hopkins, A., Bickerton, R., Soldatova, L.N. (2010)
> An Ontology of Description of Drug Design Investigations.  J. of
> Integrative Bioinformatics 7(3):126.
>
> DDI follows the principles set by the OBO Foundry. DDI is freely available
> at our project site:
>
> http://users.aber.ac.uk/ddq/ddi/wiki/index.php?title=Main_Page
>
> DDI uses BFO for the set of upper classes and  RO for the relations. DDI
>   is orthogonal to other OBO-ontologies. We re-use where possible classes
> defined in other ontologies, i.e. ChEBI, PATO, EFO, OBI, instead of
> duplicating them. DDI also imports classes following the MIREOT policy
> relevant to the area of drug discovery from ontologies outside of OBO,
> i.e. BODO for the description of compound descriptors. The references to
> the sources of all the classes are provided within the code.
>
> DDI was built as an application of OBI to the area of drug design. DDI
> defines 225 native DDI-classes. All classes have textual definitions. DDI
> is encoded in OWL-DL and uses a version control system.
>
>
>
>
> ------------------------------------------------------------------------------
> Increase Visibility of Your 3D Game App&  Earn a Chance To Win $500!
> Tap into the largest installed PC base&  get more eyes on your game by
> optimizing for Intel(R) Graphics Technology. Get started today with the
> Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs.
> http://p.sf.net/sfu/intelisp-dev2dev
> _______________________________________________
> Obi-devel mailing list
> Obi...@li...
> https://lists.sourceforge.net/lists/listinfo/obi-devel


-- 
European Bioinformatics Institute,
Wellcome Trust Genome Campus,
Cambridge, CB10 1SD,
United Kingdom
Tel: + 44 (0) 1223 494 676
Fax: + 44 (0) 1223 494 468