From NAM Data to Trusted Evidence: RISK-HUNT3R Knowledge Infrastructure for Next Generation Risk Assessment

Next Generation Risk Assessment (NGRA) requires the integration of increasingly heterogeneous evidence from New Approach Methodologies (NAMs), computational models, chemical information, biological mechanisms and exposure assessment. RISK-HUNT3R WP9 has addressed a fundamental challenge for this transition: how to organise, connect and reuse diverse scientific information so that it can support transparent, reproducible and human-relevant safety assessment.

The RISK-HUNT3R Knowledge Infrastructure provides a FAIR-oriented digital environment for managing and connecting experimental data, methods, compounds, computational resources and mechanistic knowledge. Its architecture combines harmonised data and metadata templates, controlled vocabularies and ontologies, compound and methods databases, BioStudies and EdelweissData integration, knowledge graphs, computational tools, AI assistance resources and cloud-based services. Together these resources support the progression from data capture and curation to analysis, mechanistic interpretation and evidence integration. The infrastructure has been deployed across distributed project partners to support collaborative research and reuse of RISK-HUNT3R and legacy EU-ToxRisk resources.

A central principle emerging from this work is that effective NGRA infrastructure must support more than FAIR data. Methods, models, workflows, provenance, applicability and uncertainty also need to be captured so that scientific results can be interpreted within their Context of Use and assembled into evidence supporting a decision. Resources including ToxTemp, compound and methods databases, knowledge graphs, EdelweissData and ASPA-based workflows provide complementary components of this evolving evidence architecture.

The resulting infrastructure provides both an operational resource for RISK-HUNT3R and a foundation for future development. By connecting data, methods, models and mechanistic knowledge through interoperable digital resources, RISK-HUNT3R demonstrates a pathway from fragmented NAM outputs towards reusable evidence workflows supporting transparent NGRA.