Assessing Geographical Origin of Gentiana Rigescens Using Untargeted Chromatographic Fingerprint, Data Fusion and Chemometrics

摘要: Gentiana rigescens Franchet, which is famous for its bitter properties, a traditional drug of chronic hepatitis and important raw materials the pharmaceutical industry in China. In study, high-performance liquid chromatography (HPLC), coupled with diode array detector (DAD) chemometrics, were used to investigate chemical geographical variation G. classify medicinal materials, according their grown latitudes. The chromatographic fingerprints 280 individuals 840 samples from rhizomes, stems, leaves four different latitude areas recorded analyzed tracing origin materials. At first, HPLC underground aerial parts generated while using reversed-phase chromatography. After preliminary data exploration, two supervised pattern recognition techniques, random forest (RF) orthogonal partial least-squares discriminant analysis (OPLS-DA), applied three fingerprint sets leaves, respectively. Furthermore, separately processed joined fusion strategies (“low-level” “mid-level”). results showed that classification models are based OPLS-DA more efficient than RF models. low-level method built considerably good prediction abilities (the accuracy higher 99% sensibility, specificity, Matthews correlation coefficient, efficiency range 0.95 1.00). Low-level strategy combined could provide best discrimination result. summary, this study explored phytochemical developed reliable accurate identification at latitudes on untargeted fingerprint, fusion, chemometrics. meaningful authentication quality control Chinese