Abstract
Alkenyl pheromones are a class of insect sex pheromones that are characterized by the presence of one or more double bonds, which can be either in the E(trans) or Z(cis) configuration. This structural variation is essential in mating, as it influences reproductive behavior and provides a potential method for insect control. As a base for rapid and in-situ screening of synthetic pheromones or pheromone-based products, this study explores the potential of Raman spectroscopy to differentiate between the two geometrical isomers, E(trans) and Z(cis), of the alkenyl pheromones. As a case study, four types of pheromones were analyzed: 5-decen-1-ol, 8-dodecyl acetate, 9-dodecyl acetate, and 10-dodecyl acetate; in the latter case, the E(trans) isomer was particularly investigated. In this regard, a detailed analysis of their experimental Raman spectra has been realized along with a DFT-based study of the investigated compounds. Moreover, to find the best machine learning (ML) model that can efficiently identify the E(trans) or Z(cis) isomers of alkenyl pheromones, several algorithms and two different designs of datasets were tested. The results indicate that the ML models could identify patterns and accurately predict the class even if the training dataset contains both experimental and theoretical data.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call