摘要
运用人工神经网络自组织神经树方法研究了氢醌类化合物的抑菌活性与其结构特征参数间的非线性关系，并对一个“未知样本”的活性类别作了预测，结果与实际一致，表明该方法可作为氢醌类药物构效关系研究的有效辅助手段。
The nonlinear relationship between the characteristic parameters and analgesic activities of a series of hydroquinone derivatives was calculated by selforganization neural tree model.The bioactivity kind of one unknown compound was predicted correctly on the basis of our computation.Therefore,the model could be used as an effective auxiliary technique for the molecular design of hydroquinone.
出处
《药学进展》
CAS
1999年第3期161-165,共5页
Progress in Pharmaceutical Sciences
基金
甘肃省自然科学基金
中国农业科学院院长基金
关键词
人工神经网络
氢醌类化合物
构效
自组织神经树
Hydroquinone derivativesStructureactivity relationshipArtificial neural networkSelforganization neural tree
