Abstract
Aim and Objective: Quantitative Structure- Property Relationship (QSPR) has been widely developed to derive a correlation between chemical structures of molecules to their known properties. In this study, QSPR models have been developed for modeling and predicting thermodynamic properties of 76 camptothecin derivatives using molecular descriptors.
Materials and Methods: Thermodynamic properties of camptothecin such as the thermal energy, entropy and heat capacity were calculated at Hartree–Fock level of theory and 3-21G basis sets by Gaussian 09.
Results: The appropriate descriptors for the studied properties are computed and optimized by the genetic algorithms (GA) and multiple linear regressions (MLR) method among the descriptors derived from the Dragon software. Leave-One-Out Cross-Validation (LOOCV) is used to evaluate predictive models by partitioning the total sample into training and test sets.
Conclusion: The predictive ability of the models was found to be satisfactory and could be used for predicting thermodynamic properties of camptothecin derivatives.
Keywords: Camptothecin (CPT) derivatives, QSPR, thermodynamic properties, genetic algorithm, MLR, molecular descriptors, Leave-One-Out cross-validation.
Related Journals
Current Computer-Aided Drug Design
25
3