Development of a computational peptide-protein interaction prediction method based on their tertiary structures

2017 Fiscal Year Final Research Report

• Project/Area Number: 15K16081
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Life / Health / Medical informatics
• Research Institution: Tokyo Institute of Technology
• Principal Investigator: Ohue Masahito 東京工業大学, 情報理工学院, 助教 (50743209)
• Project Period (FY): 2015-04-01 – 2018-03-31
• Keywords: ペプチド創薬 / タンパク質間相互作用 / バイオインフォマティクス / 分子ドッキング / 標的予測 / カーネル法 / ランク学習

Outline of Final Research Achievements

The purpose of this study is to efficiently search for peptide targets, which is considered important in middle molecule drug discovery, and to develop a prediction method reveals binding target.
I have performed large-scale parallelization to predict protein-protein interactions based on three-dimensional structure information. In addition, an all-to-all prediction was performed on representative human protein structures, and an web-interface was constructed to find the prediction results via a web browser. In addition, I have developed computational methods that conducting binding analysis of peptide molecules and proteins, LIK method which is a molecular target prediction method based on bipartite graph prediction by machine learning, and another molecular target prediction method named PKRank based on learning-to-rank.

Free Research Field

バイオインフォマティクス