| Project/Area Number |
26870848
|
|---|
| Research Category |
Grant-in-Aid for Young Scientists (B)
|
| Allocation Type | Multi-year Fund |
|---|
| Research Field |
Physical pharmacy
Life / Health / Medical informatics
|
|---|
| 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, ライフサイエンス技術基盤研究センター, 研究員 (50571395)
|
|---|
| Project Period (FY) |
2014-04-01 – 2016-03-31
|
| Project Status |
Completed (Fiscal Year 2015)
|
| Budget Amount *help |
¥3,900,000 (Direct Cost: ¥3,000,000、Indirect Cost: ¥900,000)
Fiscal Year 2015: ¥650,000 (Direct Cost: ¥500,000、Indirect Cost: ¥150,000)
Fiscal Year 2014: ¥3,250,000 (Direct Cost: ¥2,500,000、Indirect Cost: ¥750,000)
|
|---|
| Keywords | タンパク質ーリガンド間相互作用 / 評価関数 / 機械学習 / ドッキング / タンパク質ーリガンド間相互作用解 / 相互作用記述子 / タンパク質 / リガンド |
|---|
| Outline of Final Research Achievements |
In SBDD, protein-ligand docking is a powerful method to identify its inhibitors. We focused on residues involving in key interaction(s) such as H-bonds, ionic bonds, tight hydrophobic contact with ligands. If a prediction method for importance measure of the binding site residues is given beforehand, we could use the rank for the do rescoring and filtering. In this study, we newly set a purpose to develop prediction methods to quantify the importance of residues with the key interaction(s) around the pockets.
As a result of this study, interaction energies of probe molecules and information of amino acid were selected as descriptors and these descriptors were used for constructing prediction models. Moreover, two kinds of validation using data set based on experimental protein-ligand structures in PDB were performed to evaluate the prediction model using performance indicators such as ROC score and correlation coefficient.
|
