a combined multivariate classification and calibration approach for gas chromatographic fingerprint analysis and antioxidant activity modelling of secondary metabolites of saffron

Abstract: chromatographic fingerprinting is a common method for authentication and quality control of natural complex samples such as plat extracts. in this regard, gas chromatography (gc) is the best option for fingerprinting and identification of chemical constituents of such samples, and it can also provide reliable quantitative and quantitative information about these samples. on the other hand, due to complexity of natural sample matrices and lack of selectivity in analytical instruments, multivariate chemometric methods have been extensively used to exploit useful information from chromatographic fingerprints. saffron, called as red gold, has some bioactive compounds including crocetin, crocin, safranal and picrocrocin and it can have antioxidant effects, which can be measured by the 2,2-diphenyl-1-picrylhydrazyl (dpph) radical scavenging activity of the saffron samples. in the present contribution, a chemometrics based-strategy is proposed for gc fingerprints analysis of saffron in order to control its quality and its correlation with antioxidant activity of saffron. on this matter, an optimized ultrasonic-assisted extraction-dispersive liquid-liquid microextraction (uae-dllme) was used for extraction of chemical components of thirty-eight saffron samples. the gc fingerprints of saffron samples were obtained in optimum extraction conditions and they were arranged in a data matrix and this data matrix was mean centered and pareto scaled before multivariate classification. the data was then analyzed using principal component analysis (pca), hierarchical cluster analysis (hca) to find similarities and dissimilarities among samples. in general, two clear-cut clusters were determined using pca score plot and hca dendrogram. then, partial least squares-discriminant analysis (pls-da) was used for supervised classification of the two classes and it properly could classify samples with 94.7% sensitivity and 89.5% specificity. finally, the dpph radical scavenging activity of the saffron samples was measured by the fact that the absorption of visible light by dpph declines when dpph is reduced by an antioxidant. it was attempted to correlate the gc fingerprints of saffron samples to their dpph radical scavenging activity using partial least squares regression (plsr). it is concluded that the proposed strategy in this work can be successfully applied for comprehensive analysis of chromatographic fingerprints of complex natural samples.