Difference between revisions of "BioPortal FAQ"
Line 39: | Line 39: | ||
===How do I go about building an ontology?=== | ===How do I go about building an ontology?=== | ||
− | ===What is the OBO | + | ===What is the OBO Foundry?=== |
+ | |||
+ | A collaborative experiment, involving a group of ontology developers who have agreed in advance to the adoption of a growing set of principles specifying best practices in ontology development. These principles are designed to foster interoperability of ontologies within the broader OBO framework, and also to ensure a gradual improvement of quality and formal rigor in ontologies, in ways designed to meet the increasing needs of data and information integration in the biomedical domain. | ||
+ | |||
+ | ===Where is the OBO Foundry?=== | ||
+ | |||
+ | See http://obofoundry.org/ | ||
==Current Ontologies | ==Current Ontologies |
Revision as of 12:28, 26 July 2006
Frequently Asked Questions within cBio
This FAQ is currently in alpha - please be patient...
General
NCBO
What is NCBO?
National Center for Biomedical Ontology
OBO
What is OBO?
Open Biomedical Ontologies
Where are the ontologies?
Currently on http://obo.sourceforge.net
Soon to be moving to the NCBO site
How do I submit an ontology?
The current protocol is:
Send an email to the obo-discuss mailing list, even if the ontology is in planning stages
If the ontology satisfies basic criteria, we will add it to the OBO metadata and it will appear on the OBO site
Soon we will have an ontology submission service on the NCBO website
Ontology Content
Building Ontologies
How do I go about building an ontology?
What is the OBO Foundry?
A collaborative experiment, involving a group of ontology developers who have agreed in advance to the adoption of a growing set of principles specifying best practices in ontology development. These principles are designed to foster interoperability of ontologies within the broader OBO framework, and also to ensure a gradual improvement of quality and formal rigor in ontologies, in ways designed to meet the increasing needs of data and information integration in the biomedical domain.
Where is the OBO Foundry?
==Current Ontologies
incomplete is-a paths
Many OBO ontologies are currently incomplete with respect to all classes/terms having an is-a parent. Many tools (including OBO-Edit and AmiGO) deliberately allow the conflation of relations when presenting the ontology as a DAG or a tree to the user. However, other tools (including Protege and SWOOP) do not. This means that some OBO ontologies can look unusual when viewed with these other tools (lots of classes appear at the root level).
This is currently being addressed by the maintainers of OBO ontologies. An effort is underway to complete the is-a parentage in the GO cellular component ontology (biological process will be more difficult). The Plant Ontology Consortium are completing the is-a parentage in the plant anatomical structure ontology.
Until this work is complete, some OBO ontologies may look unusual when viewed in conventional ontology browsing and editing tools. With some of these tools, there may be some configurations which allow other relations to be conflated with the is-a relation when presenting the ontology as a DAG or tree view.
One option is to automatically create an is-a parent to the ontology root for every non-root class/term which lacks any is-a parents. Note that this may create inconsistencies in the ontology if the genuine is-a parent is somewhere deeper in the ontology hierarchy; nevertheless this step may be necessary before presenting some OBO ontologies to tools such as reasoners.
Technical
Questions for cores 1 and 2
OBO Format
Questions about the OBO text and xml formats
What is OBO Format?
Is OBO Format different from the old GO format?
Yes. The old dag-edit format (i.e. the one in which indentation was used to denote the GO hierarchy) is STRONGLY deprecated.
Where can I find the spec?
You can find details on the 1.0 spec on the GO home page http://www.geneontology.org/GO.format.shtml#oboflat
This spec is somewhat underspecified
The 1.2 spec is in preparation, and will be released to coincide with the first production OBO-Edit release sometime early in 2006
You can find a preliminary version of the spec here: HTML -- http://www.godatabase.org/dev/doc/obo_format_spec.html PDF -- http://www.godatabase.org/dev/doc/obo_format_spec.pdf Ascii text -- http://www.godatabase.org/dev/doc/obo_format_spec.txt
Why not use XML?
There is also an obo-xml format. See
http://www.godatabase.org/dev/xml
Currently only available as DTD. Relax-NG and XSD coming soon.
Is there a UML model?
Yes, there is a UML translation of the obo format in progress, produced by Harold Solbrig of Mayo clinic; see:
http://informatics.mayo.edu/schema/OBO/UMLModel/HTML/
What tools support Obo-Format?
The following are *ontology* tools that support the OBO format. There are many other tools for the analysis of gene ontology data which also happen to read the OBO format - see http://www.geneontology.org/GO.tools.shtml for a list
- OBO-Edit
- (naturally!). You should use this rather than DAG-Edit (DAG-Edit still supports obo-1.0 format), as DAG-Edit is deprecated as of 2006.
- COBRA
- developed by Stuart Aitken as part of http://www.xspan.org. COBRA is a java ontology editor/viewer for aligning ontologies
Using Protege with OBO ontologies
You have two options here - use the Protege OBO plugin, or convert to a format which Protege understands.
The OBO plugin was developed at the University of Washington, and is no longer supported.
You can convert any obo-format file to OWL and use Protege-OWL (but not Protege-classic); see the next entry.
Note that when you look at many of the OBO ontologies in Protege (classic or Protege-OWL) you will see that there are large amounts of classes apparently at the root level. This is because OBO ontologies have historically not been complete with respect to all classes having a superclass (is-a parent). This is currently being addressed. See the FAQ entry on this subject.
Can I convert Obo-format to OWL?
Yes. At this present time, the best way of foing this is to use the xslts included in go-perl. Download and install: http://search.cpan.org/~cmungall/go-perl/ Then run the script
go2owl myontology.obo
This functionality will soon be available from oboedit
For more details on the mapping, see:
http://www.fruitfly.org/~cjm/obol/doc/mapping-obo-to-owl.html
OK, but can I just download OBO ontologies as OWL without doing the conversion myself?
You can, at least for some OBO ontologies. This is an experimental service, and will eventually be subsumed into the services provided by cBio:
http://www.fruitfly.org/~cjm/obo-download
Can I convert Obo-format files to Protege-classic?
Not yet. In theory it should not be hard to provide an XSL for this. The reverse transformation may be more difficult.
Ontology Metadata
OBD
What is OBD?
OBD is a database for storing data typed using OBO ontologies
Where is it?
In development!
Is there a demo?
See http://www.fruitfly.org/~cjm/obd
Datasets
See the above URL for now