Discovery of ANO1 Inhibitors based on Machine learning and molecule docking simulation approaches

• DOI: 10.1016/j.ejps.2023.106408
• Publication Date: 2023-02-25T00:24:34Z
• Authors: Junjie Zhong; Wendi Xuan; Sheng Lu; Shihao Cui; Yuhang Zhou; Mengting Tang; Xiaosheng Qu; Wencong Lu; Haizhong Huo; Chi Zhang; Ning Zhang; Bing Niu
• Keywords: Docking (animal)

Calcium-activated chloride channels (CaCCs) are that regulated according to intracellular calcium ion concentrations. The channel protein ANO1 is widely present in cells and involved physiological activities including cellular secretion, signaling, cell proliferation vasoconstriction diastole. In this study, the inhibitors were investigated with machine learning molecular simulation. Two-dimensional structure-activity relationship (2D-SAR) three-dimensional quantitative (3D-QSAR) models developed for qualitative prediction of inhibitors. results showed accuracies model 85.9% 87.8% training test sets, respectively, rotating forest (RF) 2D-SAR model. CoMFA CoMSIA methods then used 3D QSAR modeling inhibitors, respectively. q2 coefficients cross-validation all greater than 0.5, implying we able obtain a stable drug activity prediction. Molecular docking was further simulate interactions between five most promising compounds predicted by protein. total score target 6, indicating they interacted strongly form hydrogen bonds. Finally, simulations amino acid mutations around cavity proteins each molecule had two or more sites reduced affinity following single mutation, outstanding specificity screened molecules their ligands.