Ontology methods (was: Re: Embedding specimen (and other) annotations in NeXML)
As far as I'm aware the group with the greatest experience in ontology building - admittedly a young field - is OBO Foundry. The organizers have thought hard about good practice and community process and have practical experience with what does and doesn't work. The group says they're about biomedical ontologies, but there's nothing about their practices that wouldn't work in taxonomy-related domains.
http://www.obofoundry.org/about.shtml http://www.nature.com/nbt/journal/v25/n11/full/nbt1346.html
They are concerned with taxonomy, of course, but mostly as it relates to model organisms. Discussions about how to treat taxa have come up recently and I have told them they should look to the community that has fought these wars already...
Personally I'd like to see a lot more cross-fertilization between taxonomy (biodiversity, collections, ...) and biomedical domains. For example, collections issues come up in contexts such as pathology and biological materials (plasmids etc.). Anatomy is probably another point of overlap. I'm not saying you should join, but I urge you to see if there's anything in what they do you might learn from - and vice versa.
Best Jonathan Rees Science Commons
I should maybe add that Barry Smith (one of the OBO Foundry PIs and lead author of the Nat Biotech paper) has suggested to us (us being several multi-species anatomy ontology development efforts, such as the Teleost Anatomy, Amphibian Anatomy, Spider Anatomy, and Hymenopteran Anatomy) several times to create a separate foundry for the domain of evolution as that would better foster community.
We haven't been overly excited about this yet, but broadening this scope to biodiversity and ecology and maybe embedding such a foundry within the TDWG infrastructure and organization could quite possibly make a compelling proposal.
-hilmar
On Feb 23, 2009, at 5:27 PM, Jonathan Rees wrote:
As far as I'm aware the group with the greatest experience in ontology building - admittedly a young field - is OBO Foundry. The organizers have thought hard about good practice and community process and have practical experience with what does and doesn't work. The group says they're about biomedical ontologies, but there's nothing about their practices that wouldn't work in taxonomy-related domains.
http://www.obofoundry.org/about.shtml http://www.nature.com/nbt/journal/v25/n11/full/nbt1346.html
They are concerned with taxonomy, of course, but mostly as it relates to model organisms. Discussions about how to treat taxa have come up recently and I have told them they should look to the community that has fought these wars already...
Personally I'd like to see a lot more cross-fertilization between taxonomy (biodiversity, collections, ...) and biomedical domains. For example, collections issues come up in contexts such as pathology and biological materials (plasmids etc.). Anatomy is probably another point of overlap. I'm not saying you should join, but I urge you to see if there's anything in what they do you might learn from - and vice versa.
Best Jonathan Rees Science Commons _______________________________________________ tdwg-tag mailing list tdwg-tag@lists.tdwg.org http://lists.tdwg.org/mailman/listinfo/tdwg-tag
Thanks for this Jonathan,
I'd not read the OBO Foundry paper before. I think it would certainly be useful to think in terms of TDWG products being in a form that we could take to the foundry or followed foundry principles.
I would stress again though that we need to think in small chunks i.e. not in terms of one ontology for biodiversity informatics.
Roger
On 23 Feb 2009, at 22:27, Jonathan Rees wrote:
As far as I'm aware the group with the greatest experience in ontology building - admittedly a young field - is OBO Foundry. The organizers have thought hard about good practice and community process and have practical experience with what does and doesn't work. The group says they're about biomedical ontologies, but there's nothing about their practices that wouldn't work in taxonomy-related domains.
http://www.obofoundry.org/about.shtml http://www.nature.com/nbt/journal/v25/n11/full/nbt1346.html
They are concerned with taxonomy, of course, but mostly as it relates to model organisms. Discussions about how to treat taxa have come up recently and I have told them they should look to the community that has fought these wars already...
Personally I'd like to see a lot more cross-fertilization between taxonomy (biodiversity, collections, ...) and biomedical domains. For example, collections issues come up in contexts such as pathology and biological materials (plasmids etc.). Anatomy is probably another point of overlap. I'm not saying you should join, but I urge you to see if there's anything in what they do you might learn from - and vice versa.
Best Jonathan Rees Science Commons _______________________________________________ tdwg-tag mailing list tdwg-tag@lists.tdwg.org http://lists.tdwg.org/mailman/listinfo/tdwg-tag
On Feb 24, 2009, at 7:07 AM, Roger Hyam wrote:
I would stress again though that we need to think in small chunks i.e. not in terms of one ontology for biodiversity informatics.
I don't think that's different from the recommended approach within the OBO Foundry. There isn't one ontology for bio-informatics or genome informatics or medical informatics either :-)
What the OBO Foundry principles do say is that ontologies shall aim to be orthogonal to each other.
-hilmar
participants (3)
-
Hilmar Lapp -
Jonathan Rees -
Roger Hyam