论文部分内容阅读
运用模式识别部分结合人工神经网络对４６种紫杉烷的衍生物进行药理肿瘤生物活性的分类，成功率达１００％。研究结果表明，模式识别结合神经网络用于药化活性分类筛选可行。期望在新药设计研究中探索出一套有效的化学计量优先方法。
Forty-six taxane derivatives were identified by pharmacological tumor bioassay using pattern recognition and artificial neural network. The success rate was 100%. The results show that the combination of pattern recognition and neural network is feasible for the active chemical classification. Expect to explore an effective set of chemometric priority methods in drug design research.