Computational studies on α-aminoacetamide derivatives with anticonvulsant activities

Abstract

Computational methods were used to study the structural parameters that influence that anticonvulsant activity of some α-aminoacetamides which were active in maximal electroshock seizure test. Their molecular structures were optimized with B3LYP/631G** density function theory method using Spartan 14 software. Modified-K-mediod clustering algorithm was used for data division, genetic function algorithm was used for variable selection and multiple linear regressions method was used for model construction. Developed model was statistically significant with coefficient of determination (R2) of 0.957, cross-validated R2 i.e. Q2 of 0.927, variance ratio (F4,15) of 82.94, Y-randomization R2 i.e. cR2P of 0.840 and predicted R2 (R2Pred) of 0.812. The molecular descriptors contained in the model were GATS8c (Geary autocorrelation of lag-8/weighted by atomic charges); VCH-7 (valence chain of order 7); VE3_D (Logarithmic coefficient sum of the last eigenvector from topological distance matrix) and RDF100p (radial distribution function − 100/weighted by polarizability). Molecular docking result showed that studied compounds had high binding affinity for neuronal sodium channel (PDB: 2KaV). Their binding affinity compared favorably with that of phenytoin, a validated sodium channel blocker. In addition, a linear relationship existed between anticonvulsant activity of studied compounds and their binding affinity for neuronal sodium channel. Keywords: QSAR, Molecular docking, Maximal electroshock seizure test, Genetic function algorithm, Modified-k-mediod clustering