FEGA Metadata Technical Report: EGA v2 model

The Federated EGA (FEGA) Metadata Working Group (MWG) has designed a new, process-oriented metadata model intended to replace the EGA v1 model, used by Central EGA (CEGA) and FEGA nodes using Local EGA. It prepares the FEGA network for FAIR, linked-data interoperability across human omics. The model is under active development, and this report accompanies that work in progress. What exists today is an abstract specification and a first set of JSON Schema drafts with embedded JSON-LD contexts; detailed serialisations, production deployments and full validator roll-out are future milestones, not completed deliverables. Core entities include Biomaterial, Protocol, Process, Datafile, Dataset, Policy, Data Access Committee (DAC), Study, Cohort, Project, Protocol Collection and DCAT-style Catalog objects. Validation through the open-source ELIXIR Biovalidator ensures both syntactic and selected semantic checks (e.g., ontology term validation). Use-case workshops in genomics, microarrays, proteomics, and microbiomes confirmed the model's flexibility without needing schema rewrites. A transparent GitHub repository contains the schemas, documentation, versioning, automated workflows, and a change process aligned to FEGA's network governance. As the first stepping stones for future migration of the current EGA v1 model to the EGA v2 model, we propose completing the set of model schemas, an initial v1-to-v2 model mapper, and a test implementation at CEGA. Finally, we present a phased adoption by FEGA nodes and other stakeholders. This approach would culminate when the maturity and efficacy of the model have been proved end-to-end and the model shift can happen in the production environments.