Numerical profiling of chemical structure and prediction of environmental toxicity of chemical substances using partial least squares method

2017 Fiscal Year Final Research Report

• Project/Area Number: 15K00045
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Statistical science
• Research Institution: Toyohashi University of Technology
• Principal Investigator: Takahashi Yoshimasa 豊橋技術科学大学, 工学(系)研究科(研究院), 教授 (00144212)
• Project Period (FY): 2015-04-01 – 2018-03-31
• Keywords: 環境毒性予測 / 偏最小二乗法 / 環境毒性 / PLS法 / トポロジカルフラグメントスペクトル / TFS / QSAR

Outline of Final Research Achievements

In this research, for ecological toxicity prediction of chemical substances, we proposed a parameter-free QSAR modeling method (TFS-PLS method) using PLS latent variables derived from topological fragment spectra (TFS). We verified its usefulness with real data. We also showed that a more accurate prediction result can be obtained by combining the method with active sampling of training set based on structural similarity.

Free Research Field

分子情報学