Docking and 2D-Structure-activity Relationship and ADMET Studies of Acetylcholinesterase Inhibitors

In this work, a quantitative structure-activity relationship (QSAR) for some tacrine derivatives inhibitors of acetylcholinesterase was modeled using ligand-receptor interconnection interaction space. The descriptors were obtained by multivariate image analysis (MIA) of each molecule. Docking studies were performed to determine the best conformers of inhibitors. In the first step, the best pose of all the ligands was selected. Afterward, an MIA-QSAR model using ligand-receptor interconnection data was developed. The pool of descriptors was compressed by principal component analysis (PCA). Variable selection was carried out by genetic algorithm (GA) followed by model building using the support vector machine (SVM) regression method. The validation of the model's predictive ability was studied by a validation set containing 11 individual compounds. The Q2, r2 and, ∆r_m^2 test prediction values for PCA-GA-SVM model were 0.62, 0.89 and 0.145, respectively. After validating the results with all statistical data, three new molecules were designed by the MIA-QSAR model. Afterward, new molecules docked in the AChE active site. Docking studies were showed the amino acids TYR70, TYR121, TYR334, TRP279, PHE288, PHE290, TRP84, TRP334, and SER286 are active amino acids in the complex. Finally, the ADMET parameters of the new compounds were calculated and were in acceptable ranges.