How to Develop and Use an OBO Foundry Ontology
Abstract
Biological research is being revolutionized by an explosion of data. Ontologies are increasingly being used to bring order to this plethora of data and to promote reasoning for information extraction from these data and thereby infer new information and inform new hypotheses. The Open Biomedical Ontologies (OBO) Foundry has played a pivotal role in coordinating and standardizing ontology development in the biomedical domain. In order for ontologies to successfully serve their intended purposes and bring together data to address new hypotheses, the ontologies must be constructed consistently so that both researchers and software can understand and use them in an interoperable manner.
This tutorial aims to bring together consumers and developers of ontologies with the aim of forging community partnership. We will review the list of OBO Foundry core principles, and provide guidance on their usage. This will help both existing and new OBO consortium ontology developers conform to these principles, as well as contribute new requirements. We will address the present and future use of the OBO Format and interoperability with OWL(http://purl.obolibrary.org/obo/oboformat/spec.html) and the use of new tools to facilitate format conversion.
Rationale Biomedical ontologies are widely used to index biomedical data and as such it is important to develop and use them in an orthogonal and interoperable way in support of maximal data use, reuse, and hypothesis generation. The OBO Foundry was created in support of this goal and provides a set of guiding principles that are used in the evaluation of ontologies. Ontologies that meet these criteria are members of the OBO Foundry. As the OBO consortium has grown, there have been a diversity of design patterns and variable adherence to these community standards. The goal of this tutorial is to provide guidance towards the application of the OBO principles to facilitate the consistency and orthogonality, and the promotion of OBO Library ontologies into the Foundry. As Foundry ontology development relies on parties working collaboratively, we will emphasize interoperability between different ontologies and community participation in the review process. Further, there is a need to promote interoperability between OBO format and OWL, and we provide guidance on the use of tools and metadata conversion for this purpose.
We hope to draw from the community ontology success stories and needs analysis, towards the end of developing requirements for ontology editing tools and ontology design principles and standards that can further their interoperability within the Foundry framework.
Presenters
The workshop proposers are coordinating members of the OBO Foundry and are all actively involved in the biomedical ontologies community, and will rely on existing dissemination resources to circulate the workshop, such as mailing lists. They have been and will continue to directly contact colleagues that should be interested in participating, encouraging them to prepare relevant presentations. Outreach will also be made to consumers of ontologies who will want to have this chance to directly engage the Foundry and influence future work. Finally, a website will be set up with details of the workshop, that can be linked from and will link to the main ICBO webpage, increasing visibility. Proposed format This half-day ICBO introductory tutorial aims to bring together consumers and developers of OBO ontologies with the aim of forging community partnership in the application of standards. The first section of the workshop will consist providing guidance to current and future submitters as to how they can ensure that they conform to OBO principles. Attendees will participate in a short version of applying the OBO review process. The second section of the workshop will address the current use of the OBO Format, its future, and interoperability with OWL. A short third section will be used to gather requirements for the OBO ontology portal to best suit the needs of OBO ontology consumers in support of interoperability and orthogonality in development practices. Requirements The tutorial requires little background and is suitable for creators or clients of ontologies. The following links provide information for those who would like to read about the Foundry and related tools and resources before attending the workshop: http://obofoundry.org/ http://www.obofoundry.org/wiki/index.php/OBO_Foundry_Principles http://www.ontobee.org/ http://code.google.com/p/oboformat/ http://www.obofoundry.org/id-policy.shtml http://obi-ontology.org/page/MIREOT http://code.google.com/p/owltools/wiki/OortIntro Deliverables Participants who attend the workshop will leave with an understanding of the OBO Foundry process including: - How to reserve a namespace and make your ontology available in the OBO Library - What the OBO Foundry principles are, how they came to be, and how they change - The difference between OBO Library and OBO Foundry ontologies - Assessing whether to submit your ontology to the OBO Foundry - How to initiate and participate in ontology review - Ontology formats (OBO and OWL) and how to convert between them - Release tools and process - How to use Foundry tools to make your ontology available as Linked Open Data - What ontologies are currently available as part of the OBO Foundry - How to participate in the future development of the Foundry Proposers Melissa Haendel Oregon Health & Science University Assistant professor, OHSU Library and Dept. of Medical Informatics 3181 Southwest Sam Jackson Park Road Portland, OR 97239-3079 email haendel@ohsu.edu Dr. Haendel has co-lead several previous ontology workshops and has participated in development of a number of other Open Bio-Ontology (OBO) standards for anatomy (CARO, Uberon, VAO, ZFA), phenotype (PATO), and biomedical resource ontologies (OBI). She is currently working on the ontology driven eagle-i research resource discovery platform. Her interest is in using ontologies for translational research to link human diseases to model organism data, and to infer experimental details relating to the use of organismally derived research resources. Chris Mungall Genomics Division, Lawrence Berkeley National Laboratory Building 64, 1 Cyclotron Road Berkeley, CA 94720 email cjmungall@lbl.gov Chris Mungall is a computer scientist at LBNL with an interest in applications of biological ontologies. He is part of the Gene Ontology Consortium, where he leads the software group and works on the maintenance of the GO and the cell type ontology. He is also involved in a number of other Open Bio-Ontology (OBO) efforts, and with the organization of the OBO Foundry. His research is in applying phenotype ontology based algorithms for linking animal models to human diseases using a diversity of OBO ontologies. One aspect of this work was integrating multiple vertebrate anatomy ontologies through a bridging ontology (Uberon). Alan Ruttenberg University at Buffalo School of Dental Medicine Squire Hall, Buffalo, NY 14215 email: alanruttenberg@gmail.com Alan Ruttenberg’s research interest is in structuring and using biological and clinical knowledge to enable computational interpretation of clinical and experimental data, and the scope of his current work spans technical, medical, and organizational aspects of improving access to essential knowledge. In that context, he has been an active member in a number of open biomedical ontology efforts, including: the Ontology for Biomedical Investigations (OBI), the Basic Formal Ontology (BFO) that forms the upper level ontology for the OBO Foundry, of which he is a coordinating editor, the Infectious Disease Ontology (IDO), the Program on Ontologies of Neural Structures (PONS), and the Information Artifact Ontology (IAO). He has been an active participant in W3C Semantic Web activities, and was chair of the OWL working group, developing the Neurocommons project as a prototype of how to deploy biomedical knowledge using Semantic Web technologies. Barry Smith University at Buffalo Department of Philosophy 126 Park Hall, Buffalo NY 14260 email:phismith@buffalo.edu Barry Smith is Professor of Philosophy, Neurology and Computer Science and Director of the National Center for Ontological Research at the University at Buffalo. He is also one of the principal scientists of the National Center for Biomedical Ontology (NCBO), an NIH Roadmap National Center for Biomedical Computing, a member oft he Scientific Advisory Board of the Gene Ontology Consortium, and a PI of the Protein Ontology and Infectious Disease Ontology projects. He is a Coordinating Editor of the OBO Foundry initiative, and plays a guiding role in Basic Formal Ontology and the Ontology for General Medical Science, the Environment Ontology, the Vaccine Ontology, and the Plant Ontology initiatives. Suggested reviewers Colin Batchelor, Royal Society of Chemistry, UK Judith Blake, The Jackson Laboratory, US Lindsay Cowell, University of Texas
