Host: The Japanese Society of Toxicology
Name : The 50th Annual Meeting of the Japanese Society of Toxicology
Date : June 19, 2023 - June 21, 2023
Many Molecular Initializing Events of toxicity manifestation are due to binding to biomacromolecules within the body. One evaluation method is an experimental binding assay using an in vitro expression system, but it lacks comprehensiveness. On the other hand, there is molecular docking using protein tertiary structure information as an in silico evaluation method. However, these tertiary structure analyses require large-scale equipment and the number of proteins with analyzed structures is far less than the total number of proteins. In contrast, several homology modeling methods have been proposed to predict protein tertiary structures from amino acid sequences using known structures as templates. However, the challenge has been that a template with high sequence homology is required for high-precision prediction. The situation changed dramatically in 2021 with the development of AlphaFold2 by DeepMind. AlphaFold2 is a deep learning-based high-precision protein tertiary structure prediction algorithm that achieved a high score comparable to experimental observations in the protein tertiary structure prediction contest CASP14. Furthermore, it has explosively advanced its use by comprehensively predicting the tertiary structures of more than 200 million amino acid sequences listed in UniProt and releasing them as AlphaFold Data Base. In this presentation, we would like to discuss the possibility of toxicity evaluation by combining high-precision prediction of protein tertiary structures since 2020, including AlphaFold2, with existing molecular docking and other methods.
