Predicting the grades of Astragali radix using mass spectrometry-based metabolomics and machine learning

Abstract

Astragali radix (AR, the dried root of Astragalus ) is a popular herbal remedy in both China and the United States. The commercially available AR is commonly classified into premium graded (PG) and ungraded (UG) ones only according to the appearance. To uncover novel sensitive and specific markers for AR grading, we took the integrated mass spectrometry-based untargeted and targeted metabolomics approaches to characterize chemical features of PG and UG samples in a discovery set ( n =16 batches). A series of five differential compounds were screened out by univariate statistical analysis, including arginine, calycosin, ononin, formononetin, and astragaloside Ⅳ, most of which were observed to be accumulated in PG samples except for astragaloside Ⅳ. Then, we performed machine learning on the quantification data of five compounds and constructed a logistic regression prediction model. Finally, the external validation in an independent validation set of AR ( n =20 batches) verified that the five compounds, as well as the model, had strong capability to distinguish the two grades of AR, with the prediction accuracy > 90%. Our findings present a panel of meaningful candidate markers that would significantly catalyze the innovation in AR grading.

Pseudotargeted Metabolomics Approach Enabling the Classification-Induced Ginsenoside Characterization and Differentiation of Ginseng and Its Compound Formulation Products

“…Metabolomics is a rapidly developing “-omics” technique to investigate the composition of intracellular small-molecule metabolites (<1000 Da) and their changes related to the physiology and pathology, which mainly relies on LC–MS, GC–MS (gas chromatography–mass spectrometry), and NMR (nuclear magnetic resonance). − Metabolomics has also received wide use in the quality control of medicinal herbs, particularly to differentiate those sharing the similar chemical compositions. , Comparatively, untargeted metabolomics comprehensively analyzes all of the measurable analytes in a sample through full-scan MS (MS 1 ) or MS E , thus being particularly suitable for the discovery of biomarkers . It can enable holistic differentiation among the easily confusing species, , different geographic origins, parts, growing ages, , and processing technologies .…”

Quality assessment of Astragali Radix based on pseudo‐targeted metabolomics and chemometric approach

“…Metabolomics is increasingly playing a significant role in the quality evaluation of herbal drugs by analyzing the metabolites with a holistic view. For example, untargeted metabolomics analysis was used to predict the grade of AR [11], and the color, chemical compounds, as well as antioxidant capacity of various AR, were compared based on untargeted metabolomics and targeted quantification [12]. Compared with the untargeted metabolomics analysis, the pseudo-targeted metabolomics approach combines the advantages of HRMS and triple quadrupole MS (QQQ-MS) system [13][14][15], which shows the advantages of good linearity, higher accuracy, and no standard compounds are required [16,17].…”

A comprehensive strategy for prediction and quality evaluation of standardized planting herbs based on plant metabolomics coupled with extreme learning machine: Astragali Radix as an example

“…AR with longer and thicker roots is usually considered to be of better quality, but it is susceptible to subjective influences and uniform metrics are lacking. 14 When chemical composition is used as an indicator, such as the contents of bioactive substances, there is no direct correlation between different commercial specifications and the contents of astragaloside IV and calycosin-7-O-β-D-glucoside. 15 Therefore, it has been a challenge for analysts to scientifically and comprehensively evaluate the effect of standardized planting on plant quality and quickly determine whether unknown samples were produced through standardized planting.…”