Using Multiple Linear Regression and Artificial Neural Network Techniques for Predicting CCR5 Binding Affinity of Substituted 1-(3, 3-Diphenylpropyl)-Piperidinyl Amides and Ureas

• 作者: Khalid El Akri Mohamed Zahouily Na?ma Hanafi Rokaya Mouhibi
• 来源期刊: 药物化学期刊(英文)
• 学科: 化学
• 关键词: Artificial Neural Network DESCRIPTORS CCR5 Multiple Linear Regression STRUCTURE-ACTIVITY Relationship
• 年，卷（期）: 2013,（1）
• 页码范围: 7-15
• 页数: 9页
• 分类号: O6

摘要

Quantitative structure–activity relationship (QSAR) models were developed to predict for CCR5 binding affinity of substituted 1-(3, 3-diphenylpropyl)-piperidinyl amides and ureas using multiple linear regression (MLR) and artificial neural network (ANN) techniques. A model with four descriptors, including Hydrogen-bonding donors HBD(R7), the partition coefficient between n-octanol and water logP and logP(R1) and Molecular weight MW(R7), showed good statistics both in the regression and artificial neural network with a configuration of (4-3-1) by using Bayesian and Leven-berg-Marquardt Methods. Comparison of the descriptor’s contribution obtained in MLR and ANN analysis shows that the contribution of some of the descriptors to activity may be non-linear.