Development of prediction methods for toxicity of chemical substances based on quantitative activity-activity relationship approaches

2015 Fiscal Year Final Research Report

• Project/Area Number: 25871184
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Environmental policy and social systems; Environmental impact assessment
• Research Institution: National Institute of Advanced Industrial Science and Technology
• Principal Investigator: Jun-ichi Takeshita 国立研究開発法人産業技術総合研究所, 安全科学研究部門, 主任研究員 (60574390)
• Project Period (FY): 2013-04-01 – 2016-03-31
• Keywords: 化学物質 / ヒト健康 / 有害性推論 / QAAR（定量的活性活性相関） / 統計解析

Outline of Final Research Achievements

We have developed methodologies to predict toxicity level of chemical substances to conduct a risk assessment. First, we assembled existing toxicity data in rats and mice from “Initial Risk Assessment Report,” released by Japan’s National Institute of Technology and Evaluation. We used the data to construct a training dataset, which led to the following two achievements: a) development of a statistical model to predict toxicity level of an endpoint based on the toxicity data of other endpoints, using the structural equation modeling, and b) quantification of assessment factors for each endpoint, which can be used to extrapolate toxicity level of long-term toxicity tests from that of short-term toxicity tests.
Furthermore, other than the study based on the dataset, we have developed a statistical analysis method that allows a risk assessment of combined exposure to multiple chemicals to be expressed in quantified figures.

Free Research Field

応用数学