Refined Detection of NBOMe Phenethylamines based on their Most Important Topological Descriptors

Progress in information technology may respond to the current worldwide growth of the drug abuse. NBOMe is a group of toxic phenethylamines recently identified on the illicit market as a cheaper alternative of LSD. This paper presents a series of artificial neural networks (ANNs) designed to detect NBOMe hallucinogens based on their most important topological descriptors. The topological descriptors were determined for the optimized molecular structures of 160 illicit drugs and other related compounds. Several ANNs have been built with an increasingly smaller number of the most important topological descriptors. The best performing systems have been identified based on classification rates measuring their efficiency.