Density Functional Theory Based Quantitative Structure-Activity Relationship Study of Cycloguanil Derivatives Acting as Plasmodium falciparum.

Abstract

This work presents a study of quantitative structure-activity relationship (QSAR) on the cycloguanil derivatives which are reported as growth inhibitors of clone of Plasmodium falciparum (T9/94 RC17) which houses A16V+S108T mutant dihydrofolate reductase (DHFR) enzyme. A set of 24 molecule-derived cycloguanil was modeled using the Gauss View software (03) using DFT B3LYP 6,6-31G-31G (d) as a base function. The obtained descriptions are purely electronic. The set constitute the inhibitory activity and the calculated electronic descriptors were statistically processed with principal component analysis (PCA), multiple linear regression (MLR), multiple nonlinear regressions (MNLR) and artificial neural network (ANN). The results obtained by the artificial neural network (ANN) show that the expected activities are in good agreement with the experimental results, with equal correlation coefficient R = 0, 912.To determine the architecture of this network, we varied the number of hidden layers, the number of neurons in the hidden layers, the transfer functions and the pairs of transfer functions. The best results were obtained with a network architecture [3-3-1], activation functions (Tansig-Purelin) and a learning algorithm of Levenberg-Marquardt.