Gaussian Processes: A Method for Automatic QSAR Modeling of ADME Properties

详细信息    查看全文

• 作者：Olga Obrezanova ; Gá ; bor Csá ; nyi ; Joelle M. R. Gola ; Matthew D. Segall
• 刊名：Journal of Chemical Information and Modeling
• 出版年：2007
• 出版时间：September 2007
• 年：2007
• 卷：47
• 期：5
• 页码：1847 - 1857
• 全文大小：218K
• 年卷期：v.47,no.5(September 2007)
• ISSN：1549-960X

文摘

In this article, we discuss the application of the Gaussian Process method for the prediction of absorption,distribution, metabolism, and excretion (ADME) properties. On the basis of a Bayesian probabilistic approach,the method is widely used in the field of machine learning but has rarely been applied in quantitativestructure-activity relationship and ADME modeling. The method is suitable for modeling nonlinearrelationships, does not require subjective determination of the model parameters, works for a large numberof descriptors, and is inherently resistant to overtraining. The performance of Gaussian Processes compareswell with and often exceeds that of artificial neural networks. Due to these features, the Gaussian Processestechnique is eminently suitable for automatic model generation-one of the demands of modern drug discovery.Here, we describe the basic concept of the method in the context of regression problems and illustrate itsapplication to the modeling of several ADME properties: blood-brain barrier, hERG inhibition, and aqueoussolubility at pH 7.4. We also compare Gaussian Processes with other modeling techniques.