Enabling online studies of conceptual relationships between medical terms: developing an efficient web platform

Area of Science:

• Biomedical Informatics
• Ontology Engineering
• Medical Terminology

Background:

• The Unified Medical Language System (UMLS) is crucial for understanding semantic relations between biomedical concepts.
• Existing tools for extracting transitive relationships from the UMLS have limitations in relation length and data source scope.
• Efficiently studying complex conceptual relationships within the UMLS is essential for biomedical research.

Purpose of the Study:

• To develop an efficient online platform, onGrid, for studying conceptual relationships between medical terms.
• To overcome limitations of existing methods for transitive relationship extraction from the UMLS.
• To enable comprehensive analysis of biomedical concept interdependencies.

Main Methods:

• Developed the onGrid Web platform for efficient transitive queries and conceptual relationship studies using the UMLS.
• Implemented natural language query conversion to UMLS concepts.
• Enabled visualization of result paths and dynamic generation of relationship matrices for term sets.

Main Results:

• The onGrid platform supports studies on relationships within and between any two sets of biomedical concepts.
• Demonstrated effectiveness by analyzing disease-disease relationships in the Online Mendelian Inheritance in Man (OMIM) database.
• Validated onGrid's efficacy through cross-validation with the Comparative Toxicogenomics Database (CTD).

Conclusions:

• onGrid serves as an efficient tool for querying the UMLS for transitive relations.
• Facilitates in-depth study of relationships between medical terms.
• Aids in the generation of novel research hypotheses based on semantic connections.