| 摘要: |
| 采用误差反传前向人工神经网络（artificial neural network, ANN）建立了21种2-(4-取代-苯基)-3-异噻唑啉酮类化合物的结构与其抗菌活性之间的定量关系模型（ANN模型），以21种3-异噻唑啉酮类化合物的量子化学参数和拓扑指数作为输入、抗菌活性作为输出，所构建网络模型的交叉检验相关系数为0.9916、标准偏差为0.0801、残差绝对值≤0.221，应用于外部预测集，预测集相关系数为0.9731；而多元线性回归（multiple linear regression, MLR）法模型的相关系数为0.8418、标准偏差为0.3039、残差绝对值≤0.636。结果表明，ANN模型获得了比MLR模型更好的拟合效果。 |
| 关键词: 异噻唑啉酮 定量结构-活性关系 人工神经网络 抗菌活性 大肠杆菌 |
| DOI： |
|
| 基金项目:河南省教育厅自然科学研究计划项目（2009B150023）资助。 |
|
| QSAR of 3-isothiazolinone compounds using artificial neural network |
| HE Qin,HUANG Bao-jun,LI Gong -chun |
| (College of Chemistry and Chemical Engineering, Xuchang University, Xuchang 461000;Institute of Surface Micro and Nano Materials, Xuchang University, Xuchang 461000) |
| Abstract: |
| The study of the quantitative structure-activity relationship (QSAR) on 21 kinds of 2-（4- substituted-phenyl）-3-isothiazolinones was established by the artificial neural network based on the back propagation algorithm. For the artificial neural network method, the quantum chemical parameters about structure and the molecular topological index were used as the inputs of the neural network, and the antibacterial activities as the outputs of the neural network. As a result, the leave-one-out cross-validation regression coefficient was 0.9916; the standard error was 0.0801; the correlation coefficient of the test set was 0.9731 and the absolute values of residual were less than 0.221. For comparison, the QSAR model was set up by multiple linear regressions (MLR) method. For the model built by MLR, the correlation coefficient was 0.8418; the standard error was 0.3039 and the absolute values of residual were less than 0.636. The results showed that the performance of neural network method is better than that of MLR method. |
| Key words: 3-isothiazolinone quantitative structure-activity relationship artificial neural network antibacterial activity Escherichia coli |
