Increasing the acceptance of in silico toxicology through development of protocols and position papers

Abstract

In silico toxicology protocols are currently needed to support the acceptance and deployment of computational toxicology methods as alternative methods for health hazard identification. Such protocols combine relevant in silico results with available experimental data to derive an assessment of major toxicological endpoints supported by a confidence score reflecting the uncertainty in the assessment. The protocols also identify relevant effects and/or mechanisms which can be used to guide the assessment of a toxicological endpoint. In addition, sufficient documentation of procedures and methods used to support an assessment is essential for both internal and external decision-making. The combination of relevant data, confidence scoring, and reporting provides a hazard assessment framework intended to increase the acceptance of in silico results in a toxicologic assessment. This article describes key principles and components of such protocols, including the hazard assessment framework and recommendations demonstrating how evaluating relevance, completeness, and confidence can be performed and documented. Also discussed are criteria used to develop an in silico protocol based on the state of the science and the importance of developing position papers to outline roadmaps for future in silico protocols used to guide assessments of more complex toxicological endpoints, such as cancer or neurotoxicity. The current status of providing such protocols is summarized for specific in silico protocols that are already published, in development, or planned.