An AI ‐Guided Structure‐Based Strategy for the Discovery and Biological Validation of Natural Products Targeting the Vitamin D Receptor From YSQYP in Diabetic Kidney Disease

Diabetic kidney disease (DKD) is among the most common and severe complications of diabetes mellitus, with current treatments providing limited efficacy. Yi‐Shen‐Qu‐Yu prescription (YSQYP), a traditional Chinese herbal formula, has demonstrated clinical efficacy in the treatment of DKD; however, its bioactive constituents and underlying mechanisms remain poorly defined. This study proposes a structure‐based and deep learning–guided screening strategy to identify active compounds within YSQYP that target the vitamin D receptor (VDR). An integrated screening strategy was established, combining virtual screening using conventional docking methods and deep learning tools (CarsiDock and RTMScore), as well as biophysical validation through surface plasmon resonance (SPR). Subsequently, the protective effects of candidate compounds against high‐glucose‐induced podocyte injury were further validated using biological assays. Among the screened constituents, isorhamnetin, hederagenin, and (+)‐catechin exhibited the strongest binding affinities for VDR. Functional assays confirmed their capacity to mitigate high‐glucose‐induced cytotoxicity and apoptosis in podocytes, with isorhamnetin showing the most potent protective effect. Collectively, this study establishes an efficient strategy for natural product screening and identifies promising VDR‐targeting compounds from YSQYP, providing novel therapeutic candidates for the treatment of DKD.