Abstract
New Approach Methods (NAMs) that employ computational molecular modeling approaches including artificial intelligence for predicting adverse effects of chemicals have generated optimistic expectations as alternatives to animal testing. However, there are key challenges in developing robust and predictive in silico models that relate to the impact of the quality of the input data on the model accuracy. In this chapter, we review recent trends toward greater use of computational models in regulatory assessment of chemical toxicity and the impact of data curation on the confidence in model predictions. We discuss and contrast key computational protocols for chemical toxicity prediction using case studies on acute toxicity and address the issue of model interpretation and toxicity mechanism elucidation. We conclude by emphasizing that broader acceptance of in silico approaches in regulatory toxicology is imminent, but reliability of data and rigor in model development and validation must remain central as we develop novel in silico NAMs.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call