期刊
JOURNAL OF MOLECULAR GRAPHICS & MODELLING
卷 20, 期 4, 页码 269-276出版社
ELSEVIER SCIENCE INC
DOI: 10.1016/S1093-3263(01)00123-1
关键词
QSAR modeling; LOO cross-validation; training and test sets; kNN QSAR
Validation is a crucial aspect of any quantitative structure-activity relationship (QSAR) modeling. This paper examines one of the most popular validation criteria, leave-one-out cross-validated R-2 (LOO q(2)). Often, a high value of this statistical characteristic (q(2) > 0.5) is considered as a proof of the high predictive ability of the model. In this paper, we show that this assumption is generally incorrect. In the case of 3D QSAR, the lack of the correlation between the high LOO q(2) and the high predictive ability of a QSAR model has been established earlier [Pharm. Acta Heiv. 70 (1995) 149; J. Chemomet. 10 (1996) 95; J. Med. Chem. 41 (1998) 2553]. In this paper, we use two-dimensional (2D) molecular descriptors and k nearest neighbors (kNN) QSAR method for the analysis of several datasets. No correlation between the values of q(2) for the training set and predictive ability for the test set was found for any of the datasets. Thus, the high value of LOO q(2) appears to be the necessary but not the sufficient condition for the model to have a high predictive power. We argue that this is the general property of QSAR models developed using LOO cross-validation. We emphasize that the external validation is the only way to establish a reliable QSAR model. We formulate a set of criteria for evaluation of predictive ability of QSAR models. (C) 2002 Elsevier Science Inc. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据