Principal component and hierarchical cluster analysis for discrimination and classification of mulberry tree varieties and cultivation zones

The contents of natural active compounds extracted from mulberry leaves have been used as chemotaxonomic markers to construct chemometric models in order to discriminate and classify 61 varieties of mulberry (Morus alba L.) trees from different geoghaphical origins. Discrimination between samples as a function of the tree varieties and cultivation zone was interpreted by principal component analysis (PCA) and hierarchical cluster analysis (HCA) by contents of active compounds. According to multivariate statistics models, six principal component variables can be considered important to discriminate varieties of mulberry trees, samples of 61 samples were characterized into four groups by HCA on the basis of the PCA pattern. In conclusion, classification of varieties of mulberry trees by contents of active compounds is closely related to geographic latitude.