Computational toxicology has become an integral component of safety assessments for pharmaceuticals, industrial chemicals, and across environmental health. Its applications span early‑stage screening, hazard identification, and risk assessments, including support for regulatory submissions. In our recent manuscript1, my colleague Frances Hall and I highlight that the expansion of these applications has created a need for educational approaches that align scientific principles, methodology, and regulatory context within a coherent training pathway.
Context of Use as a Central Educational Principle
A key contribution of our review is the emphasis on context of use (COU) within educational frameworks. The suitability of any computational approach depends on the question being addressed, the endpoint of interest, and the evidentiary standards required. For example, applications under ICH M7 for mutagenic impurities require complementary use of statistical and rules-based models, followed by expert review. On the other hand, the appropriateness of a read-across assessment is evaluated based on the application context, which influences the amount of uncertainty that can be accepted within a specific COU framework.
Embedding real-world scenarios in educational frameworks fosters the application of computational results within defined COU expectations (including regulatory requirements) and reinforces the need for transparency, uncertainty characterization, and clear communication of scientific reasoning.
Developing Competency Through Applied Learning
In the manuscript, 1 an educational framework that progresses from foundational scientific understanding to practical application is reviewed. Foundational elements include chemical structure representation, descriptor theory, and their relationship to toxicological principles.
Applied components are incorporated to support the development of domain fluency and may include:
- examining training analogs, applicability domains, and mechanistic relevance
- integrating computational predictions into weight of evidence assessments
- documenting rationale according to regulatory expectation, or more generally, expectations defined based on COU.
Collaboration and Access to Resources
Access to public/commercial tools and datasets strengthens educational pathways as they present opportunities to engage with scientific processes that help bridge academic learning with practical application. Collaboration among academic institutions, industry partners, and additional stakeholders further enhances methodological training and exposes learners to evolving standards.
Integration of Emerging Methods
The suitability of a method for decision-making (whether traditional or AI-enabled) depends on the specific context of use. Regulatory expectations emphasize transparency, clearly defined endpoints, and validation approaches that are commensurate with the decision being supported. In our review 1, we reference emerging guidance that promotes risk-based credibility frameworks, demonstrating the importance of integrating scientific and governance principles into training programs to ensure methodological approaches are applied and developed responsibly and with appropriate uncertainty communication.
Conclusion
The future of computational toxicology includes continued capability in applying predictive methods with scientific rigor and COU awareness. Our review provides a structured, context-driven framework that integrates foundational science, computational methodologies, and practical application, supporting the development of scientific competency, which, in turn, contributes to new approach methodology assessment strategies. gies can support the entire drug discovery and development pipeline.
Read the Full Article: Bridging science and curriculum: preparing future leaders in computational toxicology
References
1. Hall, F., & Johnson, C. (2026). Bridging science and curriculum: preparing future leaders in computational toxicology. Frontiers in Toxicology, Volume 7-2025. https://www.frontiersin.org/journals/toxicology/articles/10.3389/ftox.2025.1662963
Please contact us to speak to an expert or learn more about Instem’s In Silico solutions here.
