Abstract：
Within the realm of pharmacological strategies for cardiovascular diseases (CVD) like hypertension, stroke, and heart failure, targeting the angiotensin-converting enzyme I (ACE-I) stands out as a significant treatment approach. This study employs QSAR modeling using Monte Carlo optimization techniques to investigate a range of compounds known for their ACE-I inhibiting properties. The modeling process involved leveraging local molecular graph invariants and SMILES notation as descriptors to develop conformation-independent QSAR models. The dataset was segmented into distinct sets for training, calibration, and testing to ensure model accuracy. Through the application of various statistical analyses, the efficacy, reliability, and predictive capability of the models were evaluated, showcasing promising outcomes. Additionally, molecular fragments derived from SMILES notation descriptors were identified to elucidate the activity changes observed in the compounds. The validation of the QSAR model and designed inhibitors was carried out via molecular docking, aligning well with the QSAR results. To ascertain the drug-worthiness of the designed molecules, their physicochemical properties were computed, aiding in the prediction of ADME parameters, pharmacokinetic attributes, drug-likeness, and medicinal chemistry compatibility.
摘要：
在心血管疾病（CVD）如高血压的药理学策略范围内，中风,心力衰竭,靶向血管紧张素转换酶I（ACE-I）是一种重要的治疗方法。本研究使用蒙特卡罗优化技术进行QSAR建模，以研究一系列以ACE-I抑制特性而闻名的化合物。建模过程涉及利用局部分子图不变量和SMILES符号作为描述符来开发与构象无关的QSAR模型。数据集被分割成不同的集合进行训练，校准,和测试，以确保模型的准确性。通过各种统计分析的应用，功效,可靠性,并对模型的预测能力进行了评估，展示有希望的结果。此外,鉴定了源自SMILES符号描述符的分子片段，以阐明在化合物中观察到的活性变化。通过分子对接对QSAR模型和设计的抑制剂进行了验证，与QSAR结果吻合良好。为了确定设计分子的药物价值，计算了它们的物理化学性质，帮助预测ADME参数，药代动力学属性，药物相似，和药物化学相容性。
