Mosier Philip D, Jurs Peter C
Department of Chemistry, The Pennsylvania State University, 152 Davey Laboratory, University Park, Pennsylvania 16802, USA.
J Chem Inf Comput Sci. 2002 Nov-Dec;42(6):1460-70. doi: 10.1021/ci020039i.
The Probabilistic Neural Network (PNN) and its close relative, the Generalized Regression Neural Network (GRNN), are presented as simple yet powerful neural network techniques for use in Quantitative Structure-Activity Relationship (QSAR) and Quantitative Structure-Property Relationship (QSPR) studies. The PNN methodology is applicable to classification problems, and the GRNN is applicable to continuous function mapping problems. The basic underlying theory behind these probability-based methods is presented along with two applications of the PNN/GRNN methodology. The PNN model presented identifies molecules as potential soluble epoxide hydrolase inhibitors using a binary classification scheme. The GRNN model presented predicts the aqueous solubility of nitrogen- and oxygen-containing small organic molecules. For each application, the network inputs consist of a small set of descriptors that encode structural features at the molecular level. Each of these studies has also been previously addressed in this research group using more traditional techniques such as k-nearest neighbor classification, multiple linear regression, and multilayer feed-forward neural networks. In each case, the predictive power of the PNN and GRNN models was found to be comparable to that of the more traditional techniques but requiring significantly fewer input descriptors.
概率神经网络（PNN）及其近亲广义回归神经网络（GRNN），作为简单而强大的神经网络技术被引入，用于定量构效关系（QSAR）和定量构性关系（QSPR）研究。PNN方法适用于分类问题，而GRNN适用于连续函数映射问题。本文介绍了这些基于概率的方法背后的基本理论，以及PNN/GRNN方法的两个应用。所提出的PNN模型使用二元分类方案将分子识别为潜在的可溶性环氧化物水解酶抑制剂。所提出的GRNN模型预测含氮和含氧的小有机分子的水溶性。对于每个应用，网络输入由一小组描述符组成，这些描述符在分子水平上编码结构特征。该研究小组此前也曾使用k近邻分类、多元线性回归和多层前馈神经网络等更传统的技术来处理这些研究中的每一个。在每种情况下，发现PNN和GRNN模型的预测能力与更传统的技术相当，但所需的输入描述符要少得多。
