Artificial Neural Network and Docking Study in Design and Synthesis of Xanthenes as Antimicrobial Agents

Abstract

The aim of the study was to investigate the efficiency of artificial neural networks and docking studies in prediction of antimicrobial activity for new compounds. For that purpose, two multilayer neural networks with feedforward architecture were developed. Also, docking studies were performed to investigate the hypothetical binding mode of the target compounds. A series of 2,2,5,5-tetramethyl-9-aryl-3,4,5,6,7,9-hexahydro-1H-xanthen-1,8(2H)-dione derivatives have been previously synthesized, characterized and evaluated for in vitro antimicrobial activity against Escherichia coli and Candida albicans strains. By comparing results of in vitro investigation, new 2,2,5,5-tetramethyl-9-(3,5-dibromophenyl)-3,4,5,6,7,9-hexahydro-1H-xanthen-1,8(2H)-dione possessed better antimicrobial activity against tested microorganisms than previously synthesized derivatives and these results also correlated well with results of docking studies.