2021 Fiscal Year Final Research Report
Deep analysis of chemical communication space using artificial intelligence technology
| Project Area | Frontier research of chemical communications |
|---|
| Project/Area Number |
17H06410
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research on Innovative Areas (Research in a proposed research area)
|
| Allocation Type | Single-year Grants |
|---|
| Review Section |
Science and Engineering
|
|---|
| Research Institution | Keio University |
|---|
Principal Investigator |
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
佐藤 健吾 慶應義塾大学, 理工学部(矢上), 講師 (20365472)
齋藤 裕 国立研究開発法人産業技術総合研究所, 情報・人間工学領域, 主任研究員 (60721496)
|
|---|
| Project Period (FY) |
2017-06-30 – 2022-03-31
|
| Keywords | ケミカルスペース / 深層学習 / マルチオミックス / バーチャルスクリーニング / 人工知能 |
|---|
| Outline of Final Research Achievements |
The purpose of this research is to develop a model that represents a wide variety of chemical communication in a unified manner. We have developed the next-generation COPICAT, which is a virtual screening system that comprehensively and highly accurately predicts protein-compound interactions, and achieved higher accuracy than the state-of-the-art existing methods. We have developed a variational auto-encoder (NP-VAE) for handling natural compounds and succeeded in acquiring a chemical latent space that encodes natural macromolecular structures. A latent space of natural compounds and macromolecular structures was constructed using 1,900 types of compound data provided from the members of this research project. We succeeded in discovering a large number of new PKC ligand candidates through machine learning and expert domain knowledge feedback strategies.
|
| Free Research Field |
バイオインフォマティクス
|
| Academic Significance and Societal Importance of the Research Achievements |
これらの研究成果は，化学コミュニケーションの理解と制御に学術的に貢献するとともに，医薬品や農薬などの開発にも寄与することが期待される．とくに，世界で初めて構築した天然物・巨大分子構造の潜在空間は本領域でしか成しえない成果である．それに基づいてAIプラットフォームを完成することにより，化合物を介したあらゆる化学コミュニケーションを統一的に理解し，医薬農薬の創薬や共生などの生命現象の解明に資することになる．
|
