Integrated In Silico Prioritization of Antidiabetic Phytochemicals from Uvaria chamae P. Beauv. Based on Docking, Induced-Fit Docking, QSAR, and ADMET Analyses

Background: Diabetes mellitus remains a major public health concern, particularly in sub-Saharan Africa where type 2 diabetes predominates. In West Africa, Uvaria chamae P. Beauv. is traditionally used for diabetes management. This study investigates previously reported metabolites from Uvaria chamae using an integrated in silico approach to explore their potential antidiabetic activity and underlying mechanisms. Methods: A comprehensive literature survey identified 106 phytochemicals from stems, roots, leaves, and seeds. Diabetes-related protein targets were retrieved from the RCSB Protein Data Bank, while ligand structures were obtained from PubChem and the COCONUT database. Molecular docking, MM-GBSA rescoring, induced-fit docking, QSAR, and ADMET analyses were performed to evaluate interaction profiles, predicted activity, and developability. Results: The integrated analysis supports a polypharmacological mixture-based profile with organ-associated trends. Stem- and root-derived flavonoids, particularly isouvaretin and diuvaretin, showed the most consistent profiles for PPARγ-related pathways, while uvarinol was associated with PTP1B. Leaf alkaloids were mainly linked to DPP-4 and digestive enzyme inhibition. These compounds displayed more favorable predicted pharmacokinetic and toxicity profiles compared to acetogenins, which, despite favorable binding energies, were not prioritized as drug-like candidates due to their high lipophilicity, low QED values, and predicted toxicity liabilities, but may contribute to extract-level activity. Conclusion: These findings provide a hypothesis-generating and hierarchical framework for the prioritization of Uvaria chamae metabolites and extracts, supporting further experimental validation through enzymatic, cellular, and gene expression studies.