Prostate cancer is the most common non-cutaneous cancer in males and accounts for about 4% of all cancer-related deaths in males annually. In silico methods provide faster, economical, and environmentally friendly alternatives to the traditional trial and error method of lead identification and optimization. This study, therefore, was aimed at building a robust QSAR and QSTR model to predict the anti-proliferate activity and toxicity of some phenylpiperazine compounds against the DU145 prostate cancer cell lines and normal prostate epithelial cells as well as carry out molecular docking studies between the compounds and the androgen receptor. Genetic Function Algorithm–Multilinear Regression approach was employed in building the QSAR and QSTR model. The QSAR model built had statistical parameters R2 = 0.7792, R2adj. = 0.7240, Q2cv = 0.6607, and R2ext = 0.6049 and revealed the anti-proliferate activity to be strongly dependent on the molecular descriptors: VR3_Dzp, VE3_Dzi, Kier3, RHSA, and RDF55v. The QSTR model, on the other hand, had statistical parameters R2 = 0.8652, R2adj. = 0.8315, Q2cv = 0.7788, and R2ext = 0.6344. The toxicity of the compounds was observed to be dependent on the descriptors MATS8c, MATS3s, ETA_EtaP_F, and RDF95m. The molecular descriptors in both models were poorly correlated (R < 0.4) and had variance inflation factors < 3. Molecular docking studies between the androgen receptor and compounds 25 and 32 revealed the compounds primarily formed hydrogen, halogen, and hydrophobic interactions with the receptor. Findings from this study can be employed in in silico design of novel phenylpiperazine compounds. It can also be employed in predicting the toxicity and anti-proliferate activity of other phenylpiperazine compounds against DU145 prostate cancer cell lines.

中文翻译：

---

苯基哌嗪衍生物对 DU145 前列腺癌细胞系的抗增殖活性的 QSAR、QSTR 和分子对接研究

前列腺癌是男性最常见的非皮肤癌，每年约占男性所有癌症相关死亡的 4%。In silico 方法为铅识别和优化的传统试错法提供了更快、经济且环保的替代方法。因此，本研究旨在建立一个强大的 QSAR 和 QSTR 模型，以预测某些苯基哌嗪化合物对 DU145 前列腺癌细胞系和正常前列腺上皮细胞的抗增殖活性和毒性，并在两者之间进行分子对接研究。化合物和雄激素受体。在构建 QSAR 和 QSTR 模型时采用了遗传函数算法-多重线性回归方法。建立的 QSAR 模型具有统计参数 R2 = 0.7792，R2adj。= 0.7240，Q2cv = 0.6607，和 R2ext = 0.6049 并显示抗增殖活性强烈依赖于分子描述符：VR3_Dzp、VE3_Dzi、Kier3、RHSA 和 RDF55v。另一方面，QSTR 模型具有统计参数 R2 = 0.8652，R2adj。= 0.8315，Q2cv = 0.7788，R2ext = 0.6344。观察到化合物的毒性取决于描述符 MATS8c、MATS3s、ETA_EtaP_F 和 RDF95m。两种模型中的分子描述符相关性较差 (R < 0.4)，方差膨胀因子 < 3。雄激素受体与化合物 25 和 32 之间的分子对接研究表明，这些化合物主要与受体形成氢、卤素和疏水相互作用。这项研究的结果可用于新型苯基哌嗪化合物的计算机设计。