Support Vector Machine-Based Quantitative Structure-Activity Relationship Study of Cholesteryl Ester Transfer Protein Inhibitors

2009 ◽  
Vol 73 (5) ◽  
pp. 558-571 ◽  
Author(s):  
Siavash Riahi ◽  
Eslam Pourbasheer ◽  
Mohammad Reza Ganjali ◽  
Parviz Norouzi
Keyword(s):  
Support Vector Machine ◽  
Cholesteryl Ester ◽  
Cholesteryl Ester Transfer Protein ◽  
Quantitative Structure Activity Relationship ◽  
Activity Relationship ◽  
Support Vector ◽  
Quantitative Structure ◽  
Transfer Protein ◽  
Structure Activity ◽  
Relationship Study

Quantum chemical and molecular dynamics studies of imidazoline derivatives as corrosion inhibitor and quantitative structure–activity relationship (QSAR) analysis using the support vector machine (SVM) method

2014 ◽  
Vol 13 (02) ◽  
pp. 1450012 ◽  
Author(s):  
Lei Du ◽  
Hongxia Zhao ◽  
Haixiang Hu ◽  
Xiuhui Zhang ◽  
Lin Ji ◽  
...  
Keyword(s):  
Molecular Dynamics ◽  
Support Vector Machine ◽  
Quantum Chemical ◽  
Qsar Model ◽  
Quantitative Structure Activity Relationship ◽  
Activity Relationship ◽  
Support Vector ◽  
Quantitative Structure ◽  
Hydrophilic Group ◽  
Structure Activity

The inhibition performance of 10 imidazoline molecules with number of carbon from 15 to 21 of hydrocarbon straight-chain was studied by weight-loss method and theoretical approaches. The main purpose was to build a quantitative structure–activity relationship (QSAR) between the structural properties and the inhibition efficiencies, and then to predict efficiencies of new corrosion inhibitors. The quantum chemical calculation suggested that the active region of imidazoline molecules was located on the imidazoline ring and hydrophilic group, and active sites were concentrated on the nitrogen atoms of the molecules and carbon atoms of hydrophilic group. A model in accordance with the real experimental solution was built in the molecular dynamics, and the equilibrium configuration indicated that the imidazoline molecules were adsorbed on Fe (110) surface in parallel manner. Descriptors for QSAR model building were selected by principal component analysis (PCA) and the model was built by the support vector machine (SVM) approach, which shows good performance since the value of correlation coefficient (R) was 0.99 and the root mean square error (RMSE) was 0.94. Additionally, six new imidazoline molecules were theoretically designed and the inhibition efficiencies of three molecules were predicted to be more than 86% by the established QSAR model.

Sign in / Sign up

Export Citation Format

Share Document