Developing a hybrid screening method based on machine learning with Ligand database

2017 Fiscal Year Final Research Report

• Project/Area Number: 15K00408
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Life / Health / Medical informatics
• Research Institution: Tokyo University of Science
• Principal Investigator: Hayato Ohwada 東京理科大学, 理工学部経営工学科, 教授 (30203954)
• Co-Investigator(Kenkyū-buntansha): 青木 伸 東京理科大学, 薬学部生命創薬科学科, 教授 (00222472) ; 西山 裕之 東京理科大学, 理工学部経営工学科, 教授 (80328567)
• Project Period (FY): 2015-04-01 – 2018-03-31
• Keywords: 機械学習 / 化合物スクリーニング / リガンドデータベース

Outline of Final Research Achievements

This research focuses on a hybrid machine learning method to predict chemical properties of drug candidates using ligand databases. In-silico screening is a promising selection method for drug discovery, we have combined support vector machines with inductive logic programming, yieding a new method for improving the predictive accuracy for drug candidate selection. Moreover, p53 targeting radio protective compounds are predicted to decrease the side effect of radio based therapy. The outcomes are presented at a journal and international conference proceedings.

Free Research Field

知能情報学