Kovalishyn Vasyl V, Kholodovych Vladyslav, Tetko Igor V, Welsh William J
Institute of Bioorganic Chemistry and Petrochemistry, Kyiv, Murmanska 1, 02660, Ukraine.
J Mol Graph Model. 2007 Sep;26(2):591-4. doi: 10.1016/j.jmgm.2007.03.005. Epub 2007 Mar 19.
Volume learning algorithm (VLA) artificial neural network and partial least squares (PLS) methods were compared using the leave-one-out cross-validation procedure for prediction of relative potency of xenoestrogenic compounds to the estrogen receptor. Using Wilcoxon signed rank test we showed that VLA outperformed PLS by producing models with statistically superior results for a structurally diverse set of compounds comprising eight chemical families. Thus, CoMFA/VLA models are successful in prediction of the endocrine disrupting potential of environmental pollutants and can be effectively applied for testing of prospective chemicals prior their exposure to the environment.
使用留一法交叉验证程序比较了体积学习算法（VLA）人工神经网络和偏最小二乘法（PLS），以预测异雌激素化合物对雌激素受体的相对效力。通过威尔科克森符号秩检验，我们表明，对于包含八个化学家族的结构多样的化合物集，VLA通过产生具有统计学上更优结果的模型，其表现优于PLS。因此，比较分子场分析/体积学习算法（CoMFA/VLA）模型在预测环境污染物的内分泌干扰潜力方面是成功的，并且可以有效地应用于在潜在化学品暴露于环境之前对其进行测试。
