QSAR modeling on benzo[c]phenanthridine analogues as topoisomerase I inhibitors and anti-cancer agents

Benzo[c]phenanthridine (BCP) derivatives were identified as topoisomerase I (TOP-I) targeting agents with pronounced antitumor activity. In this study, hologram-QSAR, 2D-QSAR and 3D-QSAR models were developed for BCPs on topoisomerase I inbibitory activity and cytotoxicity against seven tumor cell lines including RPMI8402, CPT-K5, P388, CPT45, KB3-1, KBV-1and KBH5.0. The hologram, 2D, and 3D-QSAR models were obtained with the square of correlation coefficient R2 = 0.58-0.77, the square of the crossvalidation coefficient q2 = 0.41-0.60 as well as the external set's square of predictive correlation coefficient r2 = 0.5-0.80. Moreover, the assessment method based on reliability test with confidence level of 95% was used to validate the predictive power of QSAR models and to prevent over-fitting phenomenon of classical QSAR models. Our QSAR model could be applied to design new analogues of BCPs with higher antitumor and topoisomerase I inhibitory activity.     

基于支持向量机的药物诱导磷脂质病预测模型

本文应用一种组合遗传算法和共轭梯度法的支持向量机(GA-CG-SVM)方法建立了药物诱导磷脂质病分类预测模型.首先对描述符进行了优化,选出了19个描述符用于模型的构建,所建模型对训练集的预测准确率为81.6%,对测试集的预测精度为87.5%,说明所建SVM分类模型不仅能正确预测训练集药物诱导的磷脂质病,也对其他化合物具...     

Predictive activity profiling of drugs by topological-fragment-spectra-based support vector machines

Aiming at the prediction of pleiotropic effects of drugs, we have investigated the multilabel classification of drugs that have one or more of 100 different kinds of activity labels. Structural feature representation of each drug molecule was based on the topological fragment spectra method, which was proposed in our previous work. Support vector machine (SVM) was used for the classification and the prediction of their activity classes. Multilabel classification was carried out by a set of the SVM classifiers. The collective SVM classifiers were trained with a training set of 59,180 compounds and validated by another set (validation set) of 29,590 compounds. For a test set that consists of 9,864 compounds, the classifiers correctly classified 80.8% of the drugs into their own active classes. The SVM classifiers also successfully performed predictions of the activity spectra for multilabel compounds.     

机器学习方法用于二氢叶酸还原酶抑制剂的活性预测

分别采用支持向量学习机、人工神经网络、调节性逻辑回归和K-最临近等机器学习方法对761个二氢叶酸还原酶抑制剂建立了其活性分类预测模型. 采用组成描述符和拓扑描述符表征抑制剂的分子结构及物理化学性质, 使用Kennard-Stone方法进行训练集的设计, 并用Metropolis Monte Carlo模拟退火方法作变量选择. 结果表明, 支持向量学习机优于其它机器学习方法, 所得到的最优模型具有较好的预测结果, 其预测正确率为91.62%. 说明通过合适的训练集设计及变量选择, 支持向量学习机方法可以很好地用于二氢叶酸还原酶抑制剂的活性分类预测.     

CoMFA Model and Molecular Design of Anti-excitatory Activity for Benzodiazepinooxazole Derivatives against Mice

A 3D-QSAR study was conducted to analyze the anti-excitatory activity(p E) of benzodiazepinooxazole derivatives to mice by the comparative molecular field analysis(CoMFA) method. Among the 54 active molecules, a training set of 46 compounds was randomly selected to construct the CoMFA model; the remaining compounds, together with template molecule(No. 54) and two newly designed molecules constitute a test set of 17 compounds to validate the model. The obtained cross-validation coefficient(R_(cv)~2), the non-cross validation coefficient(R~2), and the test value F of the CoMFA model for training set are 0.516, 0.899, and 57.57,respectively. The model was used to predict the activities of all compounds in the training and testing sets, and the results indicated that the model had good correlation, strong stability and good predictability. Based on the 3D contour maps, eight novel benzodiazepinooxazole derivatives with higher anti-excitatory activity were designed.However, the effectiveness of these novel benzodiazepinooxazole derivatives is still needed to be verified by the experimental results.     

基于三维分子描述符的埃坡霉素类衍生物抗癌活性预测

为预测埃坡霉素类衍生物的抗癌活性, 定义了一套表征分子形状的描述符, 即K阶形状参数, 并计算了67个表征分子的电子、拓扑和几何结构的分子描述符. 描述符经遗传算法筛选, 用于建立基于支持向量学习机(SVM)的抗癌活性分类模型; 用留一法和5重交叉验证法对SVM模型参数进行了优化. 结果表明模型具有较高的预测性且两种方法得到相近结果, 交叉验证的预测正确率达80.6%; 经筛选后的描述符有30个, 其中含有5个K阶形状参数, 这些描述符对埃坡霉素类衍生物的抗癌活性的模型建立具有比较重要的作用.