Near infrared spectroscopy combined with chemometrics for growth stage classification of cannabis cultivated in a greenhouse from seized seeds
Bruna Tassi Borille a,⁎, Marcelo Caetano Alexandre Marcelo b, Rafael Scorsatto Ortiz c
,Kristiane de Cássia Mariotti a, Marco Flôres Ferrão b, Renata Pereira Limberger a
a Laboratory of Analysis and Research in Toxicology, Department of Pharmacy, Federal University of Rio Grande do Sul, Porto Alegre, Rio Grande do Sul, Brazil b Chemistry Institute, Federal University of Rio Grande do Sul, Porto Alegre, Rio Grande do Sul, Brazil.
c Rio Grande do Sul Technical and Scientifical Division, Brazilian Federal Police, Porto Alegre, Rio Grande do Sul, Brazil.
article info abstract
Cannabis sativa L. (cannabis, Cannabaceae), popularly called marijuana, is one of the oldest plants known to man and it is the illicit drug most used worldwide. It also has been the subject of increasing discussions from the scientific and political points of view due to its medicinal properties. In recent years in Brazil, the form of cannabis drug trafficking has been changing and the Brazilian Federal Police has exponentially increased the number of
seizures of cannabis seeds sent by the mail. This new form of trafficking encouraged the study of cannabis seeds seized germinated in a greenhouse through NIR spectroscopy combined with chemometrics. The plants were cultivated in a homemade greenhouse under controlled conditions. In three different growth periods (5.5 weeks, 7.5 weeks and 10 weeks), they were harvested, dried, ground and directly analyzed. The iPCA was used
to select the best NIR spectral range (4000–4375 cm−1) in order to develop unsupervised and supervised methods. The PCA and HCA showed a good separation between the three groups of cannabis samples at different growth stages. The PLS-DA and SVM-DA classified the samples with good results in terms of sensitivity and specificity.
The sensitivity and specificity for SVM-DA classification were equal to unity. This separation may be due to the correlation of cannabinoids and volatile compounds concentration during the growth of the cannabis plant. Therefore, the growth stage of cannabis can be predicted by NIR spectroscopy and chemometric tools in the early stages of indoor cannabis cultivation.
Bruna Tassi Borille a,⁎, Marcelo Caetano Alexandre Marcelo b, Rafael Scorsatto Ortiz c
,Kristiane de Cássia Mariotti a, Marco Flôres Ferrão b, Renata Pereira Limberger a
a Laboratory of Analysis and Research in Toxicology, Department of Pharmacy, Federal University of Rio Grande do Sul, Porto Alegre, Rio Grande do Sul, Brazil b Chemistry Institute, Federal University of Rio Grande do Sul, Porto Alegre, Rio Grande do Sul, Brazil.
c Rio Grande do Sul Technical and Scientifical Division, Brazilian Federal Police, Porto Alegre, Rio Grande do Sul, Brazil.
article info abstract
Cannabis sativa L. (cannabis, Cannabaceae), popularly called marijuana, is one of the oldest plants known to man and it is the illicit drug most used worldwide. It also has been the subject of increasing discussions from the scientific and political points of view due to its medicinal properties. In recent years in Brazil, the form of cannabis drug trafficking has been changing and the Brazilian Federal Police has exponentially increased the number of
seizures of cannabis seeds sent by the mail. This new form of trafficking encouraged the study of cannabis seeds seized germinated in a greenhouse through NIR spectroscopy combined with chemometrics. The plants were cultivated in a homemade greenhouse under controlled conditions. In three different growth periods (5.5 weeks, 7.5 weeks and 10 weeks), they were harvested, dried, ground and directly analyzed. The iPCA was used
to select the best NIR spectral range (4000–4375 cm−1) in order to develop unsupervised and supervised methods. The PCA and HCA showed a good separation between the three groups of cannabis samples at different growth stages. The PLS-DA and SVM-DA classified the samples with good results in terms of sensitivity and specificity.
The sensitivity and specificity for SVM-DA classification were equal to unity. This separation may be due to the correlation of cannabinoids and volatile compounds concentration during the growth of the cannabis plant. Therefore, the growth stage of cannabis can be predicted by NIR spectroscopy and chemometric tools in the early stages of indoor cannabis cultivation.