Chemometric QSAR of Organic Compounds
| Project/Area Number |
04640548
|
|---|
| Research Category |
Grant-in-Aid for General Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Research Field |
分析・地球化学
|
|---|
| Research Institution | Toyohashi University of Technology |
|---|
Principal Investigator |
MIYASHITA Yoshikatsu Toyohashi Univ.of Tech., Assoc.Prof., 工学部, 助教授 (70113884)
|
|---|
| Project Period (FY) |
1992 – 1993
|
| Project Status |
Completed (Fiscal Year 1993)
|
| Budget Amount *help |
¥2,000,000 (Direct Cost: ¥2,000,000)
Fiscal Year 1993: ¥800,000 (Direct Cost: ¥800,000)
Fiscal Year 1992: ¥1,200,000 (Direct Cost: ¥1,200,000)
|
|---|
| Keywords | Chemometrics / QSAR / PLS method / Neural Networks / 構造一活性相関 / 化学パターン認識 / 化学データモデリング |
|---|
| Research Abstract |
A Partial Least Squares (PLS) Method and Artificial Neural Networks (ANN) are chemometric tools for data modelling. PLS and ANN were employed for modelling of structure-activity relationships of organic compounds. In this study, Structure-activity relationships of fungicidal activity of azoxy compounds were investigated by using PLS and ANN.The chemical structure was represented by Z scales. The seven component PLS model was obtained but the model was not predictive. The ANN model trained by the back propagation algorithm was more predictive and give some insights into QSAR.Nonlinear modelling of C-13 NMR chemical shift data of halomethanes was investigated. ANN was used to correlate the counts of halogen atoms and chemical shift of 35 halomethanes. The ANN model successfully predicted chemical shifts of 6 commercially available halomethanes. It is found that PLS and ANN are quite useful tools for chemical data modelling.
|
Report
(3 results)
Research Products
(23 results)
