Nonlinear SVM Approaches to QSPR/QSAR Studies and Drug Design

Doucet, Jean‐Pierre; Barbault, Florent; Xia, Hairong; Panaye, Annick; Fan, Botao

doi:10.2174/157340907782799372

Cited by 81 publications

(48 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Detailed description of SVM theory has been discussed in reviews. [29][30][31] Here, the theory of SVM for regression was introduced simply in the following. [32] In SVM regression, the basic idea is to map the input data into a higher-dimensional feature space F via a nonlinear mapping and subsequently to carry out linear regression in this space.…”

Section: Descriptor Selection and Model Development Mlrmentioning

confidence: 99%

QSPR study of Setschenow constants of organic compounds using MLR, ANN, and SVM analyses

Wang

et al. 2011

J Comput Chem

View full text Add to dashboard Cite

A quantitative structure-property relationship (QSPR) study was performed for the prediction of the Setschenow constants (K(salt)) by sodium chloride of organic compounds. The entire set of 101 compounds was randomly divided into a training set of 71 compounds and a test set of 30 compounds. Multiple linear regression, artificial neural network (ANN), and support vector machine (SVM) were utilized to build the linear and nonlinear QSPR models, respectively. The obtained models with four descriptors involved show good predictive ability. The linear model fits the training set with R(2) = 0.8680, while ANN and SVM higher values of R(2) = 0.8898 and 0.9302, respectively. The validation results through the test set indicate that the proposed models are robust and satisfactory. The QSPR study suggests that the molecular lipophilicity is closely related to the Setschenow constants.

show abstract

Section: Descriptor Selection and Model Development Mlrmentioning

confidence: 99%

QSPR study of Setschenow constants of organic compounds using MLR, ANN, and SVM analyses

Wang

et al. 2011

J Comput Chem

View full text Add to dashboard Cite

show abstract

“…Support vector machines (SVM) [14][15][16][17][18][19][20][21][22][23][24][25] are a powerful state-of-the-art data mining algorithm for nonlinear input-output knowledge discovery. In SVM, the idea is to map the input data into a high dimensional feature space and subsequently carry out the linear regression in the feature space.…”

Section: Support Vector Machinementioning

confidence: 99%

Support Vector Machine Regression for Reactivity Parameters of Vinyl Monomers

Yu¹,

Wang

Chen³

2011

J. Chil. Chem. Soc.

View full text Add to dashboard Cite

Recently, the support vector machine (SVM), as a novel type of learning machine, has been introduced to solve chemical problems. In this study, ε-support vector regression (ε-SVR) and ν-support vector regression (ν-SVR) were, respectively, used to construct quantitative structure-property relationship (QSPR) models of Q and e parameters in the Q-e scheme, which is remarkably useful in the interpretation of the reactivity of a monomer in free-radical copolymerizations. The quantum chemical descriptors used to developed the SVR models were calculated from styrene and radicals with structures CH 3 CH 2 C 1The optimum ε-SVR model of lnQ (C= 9, ε =0.05 and γ =0.2) and the optimum ν-SVR model of e (C=100, ν = 0.5 and γ =0.4) produced low root mean square (rms) errors for prediction sets: 0.318 and 0.266, respectively. Thus, applying SVR to predict parameters Q and e is successful.

show abstract

“…Three activity landscapes and the representative molecules 3.2 SVR models SVR was used to develop the predictive model for each dopamine isoenzyme using ECFP_6 descriptors. SVR is a regression type of support vector machine (SVM) by the introduction of a loss function comprising of the squared deviations [8][9][10]. The general principle of SVM is to perform a classification by constructing an n-dimensional hyper plane that optimally separates the data set into two categories.…”

Section: Activity Landscapesmentioning

confidence: 99%

“…Since many structure-activity data sets are inherently nonlinear in nature, it is desirable to have a flexible method, which can model any nonlinear relationships. Recently, there has been a considerable interest in machine learning methods (ML) such as Bayesian approach [6,7] and support vector regression (SVR) [8,9] for nonlinear modeling. In general, since ML employs a sort of mathematical transformations of chemical descriptors, they have drawback that any correlations between the biological activity and the original descriptors should be lost.…”

Section: Introductionmentioning

confidence: 99%