| Year: 2003 | 
Download Free Adobe Acrobat Reader |
| LHNCBC-2003-030 |
| Biomedical Ontologies |
| Bodenreider O, Mitchell JA, McCray AT |
| Pacific Symposium on Biocomputing: World Scientific. 2003;:562-564. |
| Biomedical ontologies provide an organizational framework of the concepts involved in biological entities and processes in a system of hierarchical and associative relations that allows reasoning about biomedical knowledge. In contrast, biomedical terminologies promote a standard way of naming these concepts. Differences among various kinds of terminological systems can be briefly summarized as follows. Controlled vocabularies define a set of terms to be used for a given purpose (e.g., indexing the literature, annotating gene functions). Thesauri organize the terms in a system of relations designed to help navigate among terms as needed, for example, in information retrieval tasks. Ontologies, on the other hand, aim at representing what exists independently of any specific use; they also typically follow general theories (e.g., mereology) and carefully distinguish between the various kinds of relations among things that exist. Thesauri are often limited to tasks such as information retrieval, whereas ontologies support reasoning. Both can be shared, but ontologies lend themselves to reuse, sometimes in widely differing applications from the ones for which they were originally designed. Although more than sixty terminological systems exist in the biomedical domain, few actually qualify as an ontology. Interestingly, the most recent systems tend to be ontologies, developed either from the top down (e.g., GALEN1) or from reengineering the knowledge present in older systems (e.g., SNOMED CT2). |
| PDF |
