Deliverable 3.11: Pilot of the genomics data science ontology interconversion

Le résumé fourni par la source

This deliverable, D3.11 – Pilot of the Genomics Data Science Ontology Interconversion, presents an initial demonstration of semantic interoperability between genomic and clinical data within the HEREDITARY project. HEREDITARY seeks to enable federated, privacy-preserving machine learning across diverse datasets for neurodegenerative and neurodevelopmental conditions, a goal that requires harmonized data structures and FAIR-compliant mechanisms for data discovery. The pilot integrates two complementary efforts. First, a Beacon v2 instance was deployed on a controlled-access local server and populated with 200 whole-genome sequences from the University of Turin (UNITO) Amyotrophic Lateral Sclerosis (ALS) cohort, demonstrating the feasibility of standardized variant-level data discovery within the project. Second, a manual semantic harmonization was carried out between the AnswerALS clinical dictionary, which contains 880 variables, and the PARALS dictionary used by UNITO for their ALS cohort. This work identified 18 exact or semantically aligned correspondences and provides an initial bridge to the HEREDITARY Ontology (HERO). A parallel mapping between PARALS and the Beacon v2 schema further supports future integration of phenotypic metadata into the Beacon environment. Together, these developments constitute the first step toward an ontology-driven, interoperable data ecosystem capable of supporting federated analytics across ALS cohorts. It establishes the technical and methodological foundation for the full demonstrator to be delivered next year.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.