Making New Approach Methodologies (NAMs) Computable for Mechanistic Discovery and Regulatory Decisionmaking Through NAM Ontology (NAMO)

NAMs have become a critical focus in scientific, regulatory, and toxicology communities, with various regulatory bodies explicitly calling for their use. NAMs are generally accepted as any method for modeling a disease or biological processes that facilitates the 3Rs of animal research: Replacement, Reduction, and Refinement. However, there are few standards and best practices in documenting NAMs and what they measure - which limits computable approaches. It is therefore difficult to discover relevant NAMs or infer whether there is sufficient evidence for a given regulatory decision. To tackle these issues, we introduce the New Approach Methodology Ontology (NAMO), a LinkML schema for describing and integrating NAMs. NAMO provisions not only a classification of overarching NAM types, such as in silico models, 2-D cellular models, and 3-D models (e.g. organoids), it also defines the structure for documenting the experimental factors and physiological endpoints that are being modeled and validated. This focus on rigor and reproducibility will make NAMs data more AI-ready and fully provenanced for trustworthy decision-making. Moreover, this also ensures the resulting data is biologically meaningful for understanding toxicological outcomes mechanistically across the full lifecycle of research from hypothesis exploration through to clinical trials and safety monitoring. NAMO is available at https://github.com/monarch-initiative/namo/.