Artificial neural network designed to identify NBOMe hallucinogens based on molecular descriptors

Abstract

In the last few years there has been a rapid increase in the availability and recreational use of synthetic hallucinogens. One novel group of toxic phenethylamine derivatives, referred to as NBOMe, has recently gained prominence. The goal of this study was to develop an Artificial Neural Network (ANN) able to classify NBOMe hallucinogens based on their molecular descriptors. The database consists of 161 compounds representing drugs of abuse (NBOMe hallucinogens, sympathomimetic amines, narcotics and other potent analgesics), precursors, or derivatized counterparts. The molecular structures of all the compounds included in the database have been first optimized and then the molecular descriptors have been determined by using the Dragon 5.5. software. The validation has been performed by using all the available samples and the leave-one-out algorithm. The efficiency with which the ANN system identifies the class identity of an unknown sample was evaluated by calculating several figures of merit.