Enriching the Gene Ontology via the Dissection of Labels Using the Ontology Pre-processor Language

In this paper we describe the process of taking an axiomatically lean ontology and enriching it through the automatic application of axioms using ontology design patterns (ODP). Our exemplar is the Gene Ontology’s Molecular Function Ontology; this describes an important part of biology and is widely used to describe data. Yet much of the knowledge within the GO’s MF is captured within the term’s that label the concepts and within the natural language definitions for those concepts. Whilst both of these are absolutely necessary for an ontology, it is also useful to have the knowledge within the textual part of the ontology exposed for computational use. In this work we use an extension to the Ontology PreProcessor Language (OPPL) to dissect terms within the ontology and add axiomatisation, through OPPL’s application of ODP, that make the knowledge explicit for computational use. We show the axiomatic enriching of the GO MF; that this can be accomplished both rapidly and consistently; that there is an audit trail for the transformation; and that the queries supported by the ontology are greatly increased in number and complexity.