Dynamics of protein-protein interface: Development of analysis methods based on machine learning and MD simulation

2022 Fiscal Year Final Research Report

• Project/Area Number: 18K05025
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 32010:Fundamental physical chemistry-related
• Research Institution: The University of Tokyo
• Principal Investigator: Yamashita Takefumi 東京大学, 先端科学技術研究センター, 特任准教授 (50615622)
• Project Period (FY): 2018-04-01 – 2023-03-31
• Keywords: 分子動力学 / 機械学習 / タンパク質間相互作用 / タンパク質 / 抗体

Outline of Final Research Achievements

Due to the development of machine learning technology, artificial intelligence has surpassed human intelligence in areas such as image recognition, chess, and Go. In this study, we aim to introduce such machine learning techniques into molecular science to obtain deeper insight into the characteristic dynamics of the protein systems. In this study, we first developed a machine learning technique to predict the dynamics of a simple model system. We gradually increased the complexity of the target system and eventually succeeded in developing a machine learning technique to analyze the antigen-antibody interface.

Free Research Field

計算化学

Academic Significance and Societal Importance of the Research Achievements

機械学習技術により、人間が認識できなかったものを認識することが可能になってきた。本研究では、抗体と抗体に捕えられた抗原の間の界面を、機械学習を活用して解析する技術の開発に成功した。抗原-抗体界面の複雑なダイナミクスの理解が深まると、抗原と強く結合できる抗体の設計が容易になると期待される。この研究成果は、分子と分子の複雑な関係性を紐解くと言う学術的意義だけでなく、抗体医薬品の設計にも役立つ可能性がある。