Toxicity Prediction System for Chemical Substances Based on Toxicity Expression Mechanisms—AI-SHIPS

Traditionally, the evaluation of the safety of chemical substances through long-term exposure has been carried out by animal experiments, such as repeated dose toxicity tests. However, due to the high cost and time required for animal experiments and from the perspective of animal welfare, alternative methods such as (quantitative) structure–activity relationships ((Q)SAR), which predict toxicity from the structure of chemical substances, are being developed, mainly in developed countries. On the other hand, these methods have issues such as the lack of clear relevance to the mechanism of toxicity. For this reason, the Ministry of Economy, Trade and Industry (METI) in Japan launched the project for the development of a toxicity prediction system for general chemicals based on toxicity expression mechanisms (AI-SHIPS project) as a five-year plan starting in 2017 and is studying the combination of chemical structure, pharmacokinetics, in vitro test data, and in vivo effects. The AI-SHIPS system was developed to provide information on the mechanism of toxicity expression using the combination of chemical structure, pharmacokinetics, in vitro test data, and in vivo effects as learning data.