| 研究課題/領域番号 |
23K16936
|
|---|
| 研究種目 |
若手研究
|
| 配分区分 | 基金 |
|---|
| 審査区分 |
小区分61030:知能情報学関連
|
|---|
| 研究機関 | 北海道大学 |
|---|
研究代表者 |
SIDOROV PAVEL 北海道大学, 化学反応創成研究拠点, 准教授 (30867619)
|
|---|
| 研究期間 (年度) |
2023-04-01 – 2025-03-31
|
| 研究課題ステータス |
交付 (2023年度)
|
| 配分額 *注記 |
4,680千円 (直接経費: 3,600千円、間接経費: 1,080千円)
2024年度: 2,340千円 (直接経費: 1,800千円、間接経費: 540千円)
2023年度: 2,340千円 (直接経費: 1,800千円、間接経費: 540千円)
|
|---|
| キーワード | machine learning / chemoinfrormatics / molecular descriptors / Machine learning / Organic chemistry / Chemical reactions |
|---|
| 研究開始時の研究の概要 |
1) Machine learning requires relevant data, and the data collection will be performed at first from scientific literature and in-house experimentation.
2) Development of novel methodology for machine learning models
3) Validation of models experimentally to search for novel potent catalysts
|
| 研究実績の概要 |
The project is dedicated to the development of novel stereo-sensitive molecular descriptors and their application in modeling the selectivity of catalysts in chemical reactions. During last year, the initial developments of such representations have been undertaken, based on the established chemical libraries. By adding the stereochemical marking into the structure representation, we were able to computationally distinguish compounds of different stereochemistry, which is currently unavailable via other informatics-based libraries. The developments and some applications were published in a journal (Chemistry - A European Journal) and presented in domestic and international conferences (8th Nara School on Chemoinformatics, Nov 2023).
|
| 現在までの達成度 (区分) |
現在までの達成度 (区分)
2: おおむね順調に進展している
理由
Currently, the novel stereo-sensitive representations are undergoing the benchmark on in-house data provided by the collaborators in WPI-ICReDD. Further developments for the refinement of codebase and addition of new features are ongoing.
|
| 今後の研究の推進方策 |
This year, the application of the new representation to modeling the selectivity of catalysts will be the main focus of the project. The in-house data obtained from the collaborators in WPI-ICReDD will be used as the center of the study. Mukaiyama reaction is the focus of the dataset and represents a challenge for modeling due to the presence of several stereo centers, which cannot be treated with a simple approach. Furthermore, the type of catalysts that is used in this study requires additional care, as it possesses a special type of stereocenter (axial chirality) that cannot be managed with past approaches. With our new representation, we aim to design a new potent catalyst for this reaction and validate it experimentally as the final goal.
|
