Quality assessment of Astragali Radix based on pseudo-targeted metabolomics approach

Abstract

Astragali Radix (AR) is widely used because of its dual use in medicine and food, and its quality evaluation is of great importance. In this study, a pseudo-targeted metabolomics approach based on scheduled multiple reaction monitoring (sMRM) was developed, and a total of 114 compounds with good linearity, sensitivity and reproducibility were selected for relative quantification. With the help of multivariate and univariate analysis, 26 differential compounds between wild/semi-wild AR (AR-W) and cultivated AR (AR-C) were determined. Then 5 marker compounds were screened out by lasso regression, and further verified by systematic clustering, random forest, support vector machine, and receiver operating characteristic (ROC) analysis. In addition, malonyl-substituted flavonoids showed relative higher content in AR-W. Thus, the malonyl-substitution was the characteristic for flavonoids in AR-W. In conclusion, the application of pseudo-targeted metabolomics and various statistical methods could offer multi-dimensional information for the holistic quality evaluation of AR.

Metabolomics approach to growth‐age discrimination in mountain‐cultivated ginseng (Panax ginseng C. A. Meyer) using ultra‐high‐performance liquid chromatography coupled with quadrupole‐time‐of‐flight mass spectrometry

“…SVM is a popular discriminative method in machine learning, aiming to find an optimal hyperplane in feature space for maximum margin between positive and negative samples. SVM is a supervised learning algorithm that can handle linear problems as well as solving nonlinear problems using kernel functions [25,33]. The SVM classifier uses a linear kernel.…”