Desiderata for domain reference ontologies in biomedicine

Area of Science:

• Biomedical Informatics
• Ontology Engineering
• Knowledge Representation

Background:

• Domain reference ontologies provide objective, theory-based knowledge of a domain.
• The Foundational Model of Anatomy (FMA) and Chemical Entities of Biological Interest (ChEBI) are key examples in biomedical informatics.
• Existing research focuses on general ontology design and controlled medical vocabularies.

Purpose of the Study:

• To explore the application of FMA and ChEBI for aligning terminologies.
• To investigate their use in inferring knowledge for complex entities.
• To assess their utility in managing and reasoning about individual data.

Main Methods:

• Analysis of domain reference ontologies, specifically FMA and ChEBI.
• Examination of their application in terminology alignment.
• Evaluation of their role in knowledge inference and data management.

Main Results:

• Demonstration of how FMA and ChEBI can be used for specific applications.
• Identification of practical uses for reference ontologies in biomedicine.
• Analysis of the strengths and limitations of these ontologies for applied purposes.

Conclusions:

• Reference ontologies are crucial for advancing biomedical informatics.
• Five desirable characteristics for biomedical reference ontologies are proposed: lexical coverage, relation coverage, standards compatibility, modularity, and representation of variation.
• These characteristics are essential for effective application-driven ontology development.