2020 Fiscal Year Final Research Report
Collection of inter- and intramolecular interactions data and applications of database to drug discovery based on fragment molecular orbital method
| Project/Area Number |
18K06619
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Review Section |
Basic Section 47020:Pharmaceutical analytical chemistry and physicochemistry-related
|
|---|
| Research Institution | Institute of Physical and Chemical Research |
|---|
Principal Investigator |
Takaya Daisuke 国立研究開発法人理化学研究所, 生命機能科学研究センター, 研究員 (50571395)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
渡邉 千鶴 国立研究開発法人理化学研究所, 生命機能科学研究センター, 研究員 (60549187)
|
|---|
| Project Period (FY) |
2018-04-01 – 2021-03-31
|
| Keywords | フラグメント分子軌道法 / データベース / 量子化学計算 / タンパク質-リガンド相互作用 / 機械学習 / 活性予測 |
|---|
| Outline of Final Research Achievements |
Fragment Molecular Orbital (FMO) method is one of quantum chemical calculation methods for large scale molecules such as protein structures. In drug discovery research, IFIE/PIEDA calculated by the FMO method is drawing attention as new interaction descriptor. Through the research period, the FMO calculation results were added to FMODB. To demonstrate effectivity of IFIE/PIEDA, machine learning models were constructed for the drug targets including dataset of target proteins containing known inhibitors obtained from public data source such as ChEMBL. The effects of the prediction methods and procedure of construction of the models to the prediction performance were investigated. Moreover, A WebAPI interface was developed for easily accessing data of FMODB such as the interaction energies from programming languages. A research paper about construction of FMODB and the interaction analysis of the registered data has been published.
|
| Free Research Field |
SBDDを主とした理論創薬分野
|
| Academic Significance and Societal Importance of the Research Achievements |
FMODBの大規模なデータを使った相互作用解析により、古典力学ベースの分子力場と比較した相互作用エネルギー分布の差を統計的観点から示し、IFIE/PIEDAの予測モデルの記述子としての可能性を客観的に示すことができた。IFIE/PIEDAを記述子に含んだ阻害剤予測モデル構築では、予測手法、記述子選択、性能評価により良い予測精度が得られることがわかった。今後も創薬研究で応用事例が増えていくと期待される。医薬品設計研究分野におけるタンパク質-リガンド間相互作用評価研究においてフラグメント分子軌道法による精密な相互作用エネルギー計算の有効性を複数の観点から示すことができたと考えている。
|
