Identifying Principles for the Construction of an Ontology-Based Knowledge Base: A Case Study Approach

Area of Science:

• Bioinformatics
• Medical Informatics
• Genetics

Background:

• Ontologies are crucial for the Semantic Web and information sharing.
• The Web Ontology Language (OWL) standardizes ontology representation.

Purpose of the Study:

• To build an Ontology-based Knowledge Base for Cystic Fibrosis (OntoKBCF) to supply customizable molecular and health information via electronic health record (EHR) interfaces for precision medicine.
• To detail the construction principles, approaches, applications, and challenges in developing OntoKBCF.
• To establish a reusable framework for other ontology-based domain knowledge bases.

Main Methods:

• Defined OntoKBCF scope for molecular and clinical cystic fibrosis information needs.
• Selected knowledge sources and employed top-to-bottom and bottom-up construction methods.
• Utilized Protégé-OWL for construction and HermiT reasoner for consistency checking.

Main Results:

• Successfully constructed OntoKBCF with 408 classes and 35 properties.
• OntoKBCF integrates atomic and complex concepts, enabling automated, usable molecular and health data via EHR prototypes.
• Identified challenges in representing diverse patient groups, uncertain knowledge, and complex molecular pathways.

Conclusions:

• OntoKBCF provides a foundation for extending knowledge bases to other topics, particularly human monogenetic diseases.
• The development principles are reusable for constructing similar ontology-based knowledge bases.