The paper deals with QSAR modeling-based Monte Carlo optimization. The molecular descriptors involve the local molecular graph invariants and the SMILES notation for the molecules whose anti-MES activity is active against maximal electroshock seizure (MES). The developed QSAR model was validated with the use of various statistical parameters, such as the correlation coe cient, cross-validated correlation coe cient, standard error of estimation, mean absolute error, root-mean-square error R m 2 , MAE-based metrics, the Fischer ratio as well as the correlation ideality index. Along with the robustness of the developed QSAR model, the used statistical methods yielded an excellent predictability potential. The discovered molecular fragments utilized for the preparation of the computer-aided design of the new compounds were thought to have led to the increase and decrease of the examined activity. Molecular docking studies were referred to when making the nal assessment of the designed inhibitors. This emphasized excellent correlation with QSAR modeling results. The computation of physicochemical descriptors was conducted in order to predict ADME parameters, pharmacokinetic properties, the drug-like nature and medicinal chemistry friendliness, with the aim of supporting drug discovery. Based on the results, all the designed molecules indicate the presence of high drug-likeness.