Simultaneously Optimized Support Vector Regression Combined With Genetic Algorithm for QSAR Analysis of KDR/VEGFR‐2 Inhibitors

Abstract

Considering the fact that majority of support vector regression models have not been fully optimized in the realm of quantitative structureactivity relationship, an idea of simultaneous optimization has been proposed and evaluated on a set of novel kinase insert domain receptor ⁄ vascular endothelial growth factor receptor-2 inhibitors including naphthalene and indazole-based compounds in this study. After the powerful feature searching process using genetic algorithm, the final support vector regression model was constructed on an optimal set of six descriptors, based on which simultaneous optimization was carried out. Specifically, the global optimum is grid searched in the joint parametric space defined by cost (C), c and e, where performance of support vector regression using each combination of (C, c, e) is evaluated and recorded, resulting in bulky information. Based on the data decomposition strategies provided in the main paper, the best performance was achieved for C = 1.2, c = 0.15 and e = 0.065. As a comparison, a linear model based on genetic algorithm-multiple linear regression has also been developed and compared. Performances of these models are rigorously validated using both leave-one-out cross-validation and also external validation. The significant higher R 2 (0.908, 0.837) and lower root-meansquare error (0.237, 0.311) for 45 training and 16 test samples compared to that of genetic algorithm-multiple linear regression (0.764, 0.700 and 0.402, 0.421) confirm the superior performance of genetic algorithm-support vector regression. Robustness and predictive ability of this model is further prudently evaluated. The resulting models introduced not only the idea of simultaneous optimization in support vector regression, but also an efficient strategy for estimating the vascular endothelial growth factor receptor-2 inhibitory activity of novel naphthalene and indazole-based compounds. Moreover, some insights into the structural features related to the biological activity of these compounds have also been provided, which might be of great help for further designing novel vascular endothelial growth factor receptor-2 ⁄ kinase insert domain receptor inhibitors with potent activity.

A machine learning‐based KNIME workflow to predict VEGFR‐2 inhibitors

“…The R 2 obtained for the test set for MLR and SVM‐based models were found to be 0.943 and 0.853, respectively (Nekoei et al, 2015). Sun et al (2010) performed a QSAR study using the SVM approach for the prediction of inhibitory activity of naphthalene and indazole‐based compounds against VEGFR‐2. The model was developed using 2D descriptors and the developed model had R 2 of 0.908 in leave‐one‐out cross‐validation.…”

QSAR Models for Phosphoramidate Prodrugs of 2′‐Methylcytidine as Inhibitors of Hepatitis C Virus Based on PSO Boosting

“…Quantitative structure–activity relationship (QSAR) is a good method which seeks to discover and use mathematical relationships between chemical structure and biological activity. However, the machine learning method plays a key role in constructing QSAR models, which include linear and non‐linear methods such as linear discriminant analysis, multiple linear regression (MLR), principal component analysis, hierarchical cluster analysis, K‐nearest neighbor, partial least squares, soft independent modeling of class analogy, artificial neural networks, support vector machine (SVM), multivariate adaptive regression splines, gene expression programming, genetic algorithms (GA), heuristic method, stepwise multiple linear regression (stepwise‐MLR), and particle swarm optimization (PSO) (11–16). Good QSAR models should save resources and accelerate the process of developing new molecules that are used as drugs, material, additives, and any other purpose.…”