Semantic modelling of common data elements for rare disease registries, and a prototype workflow for their deployment...

Area of Science:

• Biomedical Informatics
• Data Science
• Rare Diseases

Background:

• Fragmentation of European rare disease (RD) patient data across numerous registries hinders research and collaboration.
• The European Platform on Rare Disease Registration (EU RD Platform) established 16 Common Data Elements (CDEs) to standardize RD data.
• Achieving interoperability requires standardized data models, formats, and semantics beyond just data elements.

Purpose of the Study:

• To further the goals of the EU RD Platform by creating reusable RD semantic model templates.
• To ensure these templates adhere to the FAIR Data Principles (Findable, Accessible, Interoperable, Reusable).
• To facilitate the integration and interoperability of diverse rare disease patient data.

Main Methods:

• A team-based iterative approach was used to develop semantically grounded models for each CDE.
• The SemanticScience Integrated Ontology served as the core framework for representing entities and relationships.
• Domain ontologies (Orphanet RD Ontology, HPO, NCI Thesaurus) were mapped to CDE concepts and values.
• A reusable Extract, Transform, Load (ETL) pipeline was created to assist data repositories in creating model-compliant FAIR data.

Main Results:

• Semantically grounded models representing CDEs were successfully created using a core ontology framework.
• Mappings were established between CDE concepts and established domain ontologies.
• An exemplar ETL pipeline was developed to facilitate the deployment of model-compliant FAIR data without requiring site-specific coding or Linked Data expertise.

Conclusions:

• Reusable, expert-designed semantic model templates significantly reduce the need for specialized knowledge in OWL semantics for domain experts and data hosts.
• These templates empower biomedical experts and rare disease data hosts to publish highly expressive FAIR data using familiar tools and approaches.
• The developed approach enhances data sharing and interoperability for rare disease research within the European Joint Programme on Rare Diseases (EJP RD).