Rapid Identification of Potential Depressant Compounds in Jasmine Leaves Using UHPLC‐Q‐TOF‐MS/MS Coupled with Machine Learning Techniques**

Abstract

AbstractThe following is an abstract of the study. The objective of this study was to identify the chemical constituents of Jasmine Leaves and predict their potential sedative effects. A rapid chemical analysis identified the compounds present in Jasmine Leaves. A database was created to establish “target‐disease” and “compound‐target” networks. Four machine learning models were developed and evaluated to predict the sedative potential of various compounds. Molecular docking was performed on the two most promising compounds identified through network analysis and their respective targets. A total of 34 compounds were identified in Jasmine Leaves. Eight compounds, including citric acid and Pogostone, were identified as being associated with genes involved in the process of sedation. The Random Forest (RF) model showed the best performance, with an accuracy of 0.75, an F1 score of 0.76, an AUC of 0.83, a sensitivity of 0.70, a specificity of 0.72, a precision of 0.77, and an MCC of 0.42. The model predicted the potential sedative effects of citric acid and Pogostone with probabilities of 0.51 and 0.29, respectively. Molecular docking results indicated that citric acid and Pogostone had strong interactions with tyrosinase and gamma‐aminobutyric acid type A (GABAA) receptors, with MOE software scores of −3.83 and −4.18, respectively. This study concludes that a rapid assessment method was developed to evaluate the pharmacological potential of natural products, suggesting that Jasmine Leaves may exert sedative effects through interactions between citric acid and Pogostone with tyrosinase and GABAA receptors. Additionally, Jasmine Leaves show promise as a cost‐effective sleep aid beverage.