PODD: An Ontology Driven Architecture for Extensible Phenomics Data Management - IBC 2011

From Plant Ontology Wiki
Jump to navigationJump to search

Gavin Kennedy (1,2), Yuan-Fang Li (2), Faith Davies (2), Jane Hunter (2), Bob Furbank (1)

1. The Australian Plant Phenomics Facility: High Resolution Plant Phenomics Centre 2. School of ITEE, The University of Queensland

Address all correspondence to: Gavin Kennedy, g.kennedy1@uq.edu.au

Ontologies have found increasing favour in the plant sciences because they can deliver a set of terminologies and understandings about biological concepts that are agreed between researchers. Typically ontologies are used to annotate data on the web, but the notion of a common vocabulary with formally defined semantics makes ontologies the vehicles for representing data and knowledge in the Semantic Web. Ontologies provide unambiguous classifiers and descriptors that are made available in a format other computers can autonomously discover and interrogate, and thus may be linked across disparate databases and repositories.

In the Phenomics Ontology Driven Data repository (PODD) we have taken the notion of classification of experimental concepts using ontologies one step further by using an ontology, the PODD ontology, as the schema of our data management system. We utilise the Semantic Web ontology languages OWL and RDFS to do this because they provide the extensibility and the semantic rigour required. In this ontology-driven architecture the behaviours of domain concepts and objects are captured entirely by ontological entities, around which all data management tasks are carried out.

An ideal domain for applying these principles is plant phenomics, the systematic study of the phenotypes of model and crop plants that are a consequence of the individual plant’s genome and environment. Phenomics research generates high volumes of heterogeneous data through the use of emerging imaging and measurement technologies and processes. This data is combined with metadata to form complex digital objects and then further associated with provenance metadata on the experimental process. In this context, we describe the development of a phenomics experimental process ontology, and how we have applied the principles of ontology-driven architecture in the development of PODD, a data management system for phenomics based research.

Keywords: Phenomics, Semantic Web, Ontologies, OWL, data management, PODD

Back to IBC 2011 Bio-Ontologies Symposium