Establishment of a "non-invasive" toxicity testing method for wildlife using AI
Environmental pollutants cause a variety of chemical hazards to wildlife.
The toxicity of chemical substances has been evaluated by toxicity tests using laboratory animals, but there is a large difference in animal species between these and wild animals, and it is not possible to evaluate the toxicity of chemical substances based solely on data obtained from laboratory animals. However, there are large species differences between these animals and wild animals, and it is not possible to evaluate them only from data obtained from laboratory animals. In addition, it is naturally contraindicated to administer chemical substances to rare wild animals, and a non-invasive evaluation method is required.
Therefore, in cooperation with the School of Information Science and Engineering, Tokyo Institute of Technology, our laboratory is trying to develop a new toxicity evaluation method that predicts the differences in animal species sensitivity to chemical substances using computer simulations such as molecular docking simulation and molecular dynamics simulation, and AI technologies such as machine learning and deep learning.