Innovative perspective on the geographical origin and quality of Peucedanum praeruptorum Dunn through the integration of inorganic and organic substance profiles

The clinical efficacy of traditional Chinese medicines (TCMs) is closely linked to their genuine quality. Identifying the geographical origin and genuine characteristics of medicinal materials is pivotal for enhancing quality control and efficacy. Qianhu (Peucedanum praeruptorum Dunn), a commonly used TCM, still lacks a clear understanding of the specific connection between its quality and geographical origin. To bridge this gap, we innovatively propose a new recognition model that integrates inorganic and organic substances, using Qianhu as the model TCM. By incorporating the concepts of metallomics and metabolomics, we merge elemental fingerprint profiles with chemical component contour maps to pinpoint its geographical origin. The research findings suggested that, compared to chemometrics, machine learning with data oversampling could precisely identify Qianhu from areas like Anhui, Zhejiang, Guizhou, and Chongqing of China, especially distinguishing genuine ones. Upon this groundwork, we further introduced an innovative ensemble model that deeply integrated the optimal models for elements and chemical components, thereby substantially enhancing classification accuracy. Additionally, through variable importance analysis, we provided professional and in-depth interpretations for the elements and chemical components within the model. In summary, this study, for the first time, revealed the scientific basis of Qianhu's producing area and genuine quality through machine learning, integrating inorganic and organic factors. It provides a solid foundation for scientific and reasonable quality control and clinical application of TCMs.