Choosing relevant functional groups for optimizing Artificial Neural Networks detecting NBOMe hallucinogens
In the early 2010s, a new group of illicit psychedelic phenethylamines was reported by the law enforcement agencies, namely the NBOMe hallucinogens. The latter seem to be sold on the black market as an alternative to LSD, due to their powerful psychoactive effects. The goal of this study was to develop an optimized Artificial Neural Network (ANN) able to classify NBOMe hallucinogens based on their functional groups. These chosen molecular descriptors (functional groups) have been computed, by using the Dragon 5.5 program, for the molecular structures of the main NBOMe hallucinogens, which have been first optimized by using the Hyperchem program. The ANN system was built with the Easy NN plus program. Then, the importance of each functional group has been assessed. A new input database has been built with the functional groups found to be the most important.The performance of the new ANN system has been characterized based on several classification accuracy criteria. The impact of the variable selection on the ANN performances is discussed in detail.