芳香族化合物生物降解性的QSBR研究

分别采用线性基团贡献法和人工神经网络法对芳香族化合物的生物降解最大去除率QTOD进行QSBR研究。得到不同基团对生物降解性的贡献顺序为 :C6H5>COOH >OH >CO >CH3 >C1 >NH2>NO2 。线性基团贡献法对于训练组和测试组的预测正确率分别为 86%和 80 % ,总的预测正确率达85 % ;而人工神经网络法的预测正确率分别为 94%、80 %和 92 %。结果表明 ,线性基团贡献法和神经网络法的预测效果均很好 ,而神经网络法的预测更精确。     

Estimation of aqueous solubility for a diverse set of organic compounds based on molecular topology

An accurate and generally applicable method for estimating aqueous solubilities for a diverse set of 1297 organic compounds based on multilinear regression and artificial neural network modeling was developed. Molecular connectivity, shape, and atom-type electrotopological state (E-state) indices were used as structural parameters. The data set was divided into a training set of 884 compounds and a randomly chosen test set of 413 compounds. The structural parameters in a 30-12-1 artificial neural network included 24 atom-type E-state indices and six other topological indices, and for the test set, a predictive r2 = 0.92 and s = 0.60 were achieved. With the same parameters the statistics in the multilinear regression were r2 = 0.88 and s = 0.71, respectively.     

A neural network based virtual high throughput screening test for the prediction of CNS activity

A virtual high throughput screening test to identify potentially CNS-active drugs has been developed. Discrimination was based on the knowledge available in databases containing CNS-active (Cipsline from Prous Science) and inactive compounds (Chemical Directory from Sigma-Aldrich). Molecular structures were represented using 2D Unit y fingerprints and a feedforward neural network was trained to classify molecules regarding their CNS activity. The parameterized network was validated by reclassification of the training set elements, by the classification of a test set preselected from the Prous database, and also by the prediction of activity for known CNS drugs not used in the training set but available in the Medchem database (Daylight). These tests revealed that our neural net recognized at least 89% of CNS-active compounds and would be suitable for use in our virtual screening protocol.     

Quantitative structure-toxicity relationships (QSTRs): a comparative study of various non linear methods. General regression neural network, radial basis function neural network and support vector machine in predicting toxicity of nitro- and cyano- aromatics to Tetrahymena pyriformis

Prediction of toxicity of 203 nitro- and cyano-aromatic chemicals to Tetrahymena pyriformis was carried out by radial basis function neural network, general regression neural network and support vector machine, in non-linear response surface methodology. Toxicity was predicted from hydrophobicity parameter (log Kow) and maximum superdelocalizability (Amax). Special attention was drawn to prediction ability and robustness of the models, investigated both in a leave-one-out and 10-fold cross validation (CV) processes. The influence that the corresponding changes in the learning sets during these CV processes could have on a common external test set including 41 compounds was also examined. This allowed us to establish the stability of the models. The non linear results slightly outperform (as expected) multilinear relationships (MLR) and also favourably compete with various other non linear approaches recently proposed by Ren (J. Chem. Inf. Comput. Sci., 43 1679 (2003)).     

The predicting study for chromatographic retention index of saturated alcohols by MLR and ANN

A QSRR method was followed to relate the observed Kovats retention indexes of saturated alcohol compounds with their molecular connectivity indices by means of multilinear regression analysis and artificial neural networks technique. The alcohols included linear, branched with hydroxyl group on a primary, secondary, or tertiary carbon atom. At first, models were generated for six OV (Ohio Valley) series columns separately, with high value of R and F statistics. Then a combined model, added a polarity term of stationary phase (M), was also developed for all these columns, and the result was satisfactory. For comparison, the neural network of BP algorithm was applied, and it was found that the neural network could exceed the level of the multiple regression method. The stability and validity of both models were tested by cross-validation technique and by prediction response values for the prediction set.     

The Learned Symmetry Concept in Revealing Quantitative Structure-Activity Relationships With Artificial Neural Networks

Abstract

A method to build QSAR models based on substituent constants for congeneric sets of compounds having several topologically equivalent substituent positions was proposed. The approach is based on the application of artificial neural networks (learning to construct nonlinear structure-activity relationships taking into account necessary symmetry properties of training set structures) to a training set expanded by adding the copies of compounds with the same activity values but with permuted assignment of equivalent substituent positions. The better predictive power of these constructed models, as compared with the performances of neural network models for non-expanded sets was demonstrated for the calcium channel blockers of 1,4-dihydropyridine type and for hallucinogenic phenylalkylamines.     

Machine learning techniques for the prediction of the peptide mobility in capillary zone electrophoresis

Three machine learning techniques including back propagation artificial neural network (BP-ANN), radial basis function artificial neural network (RBF-ANN) and support vector regression (SVR) were applied to predicting the peptide mobility in capillary zone electrophoresis through the development of quantitative structure-mobility relationship (QSMR) models. A data set containing 102 peptides with a large range of size, charge and hydrophobicity was used as a typical study. The optimal modeling parameters of the models were determined by grid-searching approach using 10-fold cross-validation. The predicted results were compared with that obtained by the multiple linear regression (MLR) method. The results showed that the relative standard errors (R.S.E.) of the developed models for the test set obtained by MLR, BP-ANN, RBF-ANN and SVR were 11.21%, 7.47%, 5.79% and 5.75%, respectively, while the R.S.E.s for the external validation set were 11.18%, 7.87%, 7.54% and 7.18%, respectively. The better generalization ability of the QSMR models developed by machine learning techniques over MLR was exactly presented. It was shown that the machine learning techniques were effective for developing the accurate and relaible QSMR models.