Discriminant Analysis of Gentiana Headspace Gas Chromatography - Mass Spectrometry Extract Data from In Vitro Cultures

A set of Gentiana L. species was successfully grown in vitro under the same condi-tions, and 72 samples from various cultures of these species (root, shoot, cotyledon callus, hypocotyl callus, and root callus) were obtained. The investigated species were G. affinis, G. andrewsii, G. bhutanica, G. burseri, G. cachemirica, G. capi-tata, G. crassicaulis, G. dahurica, G. decumbens, G. freyniana, G. frigida, G. gel-ida, G. grossheimii, G. kurroo, G. macrophylla, G. paradoxa, G. robusta, G. scabra, G. septemfida, G. siphonantha, and G. tianschanica. The obtained samples were extracted with a methanol-acetone-water (3:1:1) mixture, evaporated to dryness, and subjected to thin layer chromatography (TLC) on silica gel in sandwich mode with ethyl acetate-methanol-water (8:2:2) as the mobile phase. The resulting dry extracts were subjected to gas chromatography - mass spectrometry (GC-MS) fingerprinting of the headspace volatile fraction. Total ion count and average mass spectrum vectors were collected as two blocks and scaled independently to form a complex dataset. The major direction separating root or shoot samples from callus samples was found not to be fully associated with the highest variance as this information was placed in the first and fourth principal components of the principal component analysis (PCA). Therefore, linear discriminant analysis was performed on the first four (only the informative) components to reveal features responsible for the separation of culture types in the multivariate space.