Classification of cannabinoids using mass spectral data to assist in the identification of novel synthetic cannabinoids
- WCU Author/Contributor (non-WCU co-authors, if there are any, appear on document)
- Kristopher Charles Evans-Newman (Creator)
- Institution
- Western Carolina University (WCU )
- Web Site: http://library.wcu.edu/
- Advisor
- Nuwan Perera
Abstract: Detection and characterization of newly synthesized cannabinoids (NSCs) is challenging due to the lack of availability of reference standards and chemical data. Identifying these substances usually involves the use of intelligence gathering or prior knowledge on new psychoactive substances (NPSs). The current methods for the structural elucidation of NPSs are using expensive and time-consuming analysis methods such as high-resolution mass spectrometry (HRMS), and nuclear magnetic resonance spectroscopy (NMR). The focus of this study is to develop a solution to identify NSCs. A classification system was developed using existing mass spectral data of synthetic and classical cannabinoids to determine the presence of previously unknown cannabinoid-related substances encountered in laboratories. Research used computer learning software from Eigenvector, called PLS -Toolbox, which is an addon in a MATLAB programming language. Principle component analysis (PCA) and partial least square discriminant analysis (PLS-DA) were used to develop a binary classification system using mass spectra obtained from a freely available spectral database maintained by the Scientific Working Group for the Analysis of Seized Drugs (SWGDRUG). Genetic algorithm (GA) was used to select the most discriminatory mass to charge (m/z) ratios of the mass spectral data. First, a binary classification model was developed to discriminate cannabinoids and cannabinoid-related compounds from other drug classes such as opioids, tryptamines, fentanyl and fentanyl derivatives. Then, a classification model was developed to discriminate classical and synthetic cannabinoids. Finally, sub-models were developed to discriminate the presence and absence of functional groups found on commonly encountered synthetic cannabinoids. Hierarchical cluster analysis (HCA) in conjunction with PCA was used to determine the possible drug classes and HCA along with chemical structure similarities resulted in the different unique groups seen with synthetic cannabinoids that were classified. These groups include benzopyrrole (indole), isoindazole (indazole), naphthalene (Naphthyl), 4-Fluorobenzyl (FUB), and 1-amino-3,3-dimethyl-1-oxobutan-2-yl (BUT). Classification models were developed for the determination of the presence or absence of functional groups in an unknown cannabinoid. Current results show that these models are highly accurate (>95%) and applicable in determining the presence of cannabinoid-related substances and different functional groups. Limitations encountered include issues with mass spectral fragmentation patterns for similar structural compounds such as indazoles and azaindoles. Future research motivation includes building on to the binary classification system. Some of these groups include but are not limited to 3-dimethylbutanoate (MDMB) and N-[1-(Aminocarbonyl)-2,2-dimethylpropyl]-1-butyl (ADB).
Classification of cannabinoids using mass spectral data to assist in the identification of novel synthetic cannabinoids
PDF (Portable Document Format)
6809 KB
Created on 4/1/2024
Views: 167
Additional Information
- Publication
- Thesis
- Language: English
- Date: 2024
- Keywords
- Computer Learning, Mass Spectrometry, Novel Psychoactive Substances, Synthetic Cannabinoids
- Subjects
- Mass spectrometry
- Machine learning
- Cannabinoids
- Synthetic marijuana