Abstract

Increasing diabetic population is one of the major health concerns all over the world. Inhibition of α-glucosidase is a clinically proved and attractive strategy to manage diabetes. In this study, robust and reliable QSAR models to predict α-glucosidase inhibitory potential of xanthone derivatives are developed by the Monte Carlo technique. The chemical structures are represented by SMILES notation without any 3D-optimization. The significance of the index of ideality correlation (IIC) with applicability domain (AD) is also studied in depth. The models developed using CORAL software by considering IIC criteria are found to be statistically more significant and robust than simple balance of correlation. The QSAR models are validated by both internal and external validation methods. The promoters of increase and decrease of activity are also extracted and interpreted in detail. The interpretation of developed models explains the role of different structural attributes in predicting the pIC₅₀ of xanthone derivatives as α-glucosidase inhibitors. Based on the results of model interpretation, modifications are done on some xanthone derivatives and 15 new molecules are designed. The α-glucosidase inhibitory activity of novel molecules is further supported by docking studies.

摘要

糖尿病人口的增加是全世界关注的主要健康问题之一。抑制α-葡萄糖苷酶是一种临床证明且具有吸引力的糖尿病管理策略。在这项研究中，通过 Monte Carlo 技术开发了可靠且可靠的 QSAR 模型来预测呫吨酮衍生物的 α-葡萄糖苷酶抑制潜力。化学结构由 SMILES 符号表示，没有任何 3D 优化。还深入研究了理想相关指数（IIC）与适用域（AD）的重要性。发现使用 CORAL 软件通过考虑 IIC 标准开发的模型比简单的相关性平衡在统计上更显着和稳健。QSAR 模型通过内部和外部验证方法进行验证。还详细提取和解释了活性增加和减少的促进剂。已开发模型的解释解释了不同结构属性在预测 pIC 中的作用₅₀的黄酮衍生物作为 α-葡萄糖苷酶抑制剂。根据模型解释的结果，对部分呫吨酮衍生物进行了修饰，设计了15个新分子。对接研究进一步支持了新分子的α-葡萄糖苷酶抑制活性。