Predictive 3D-QSAR and HQSAR model generation of isocitrate lyase (ICL) inhibitors by various alignment methods combined with docking study

[Ensemble hologram quantitative structure activity relationship model of the chromatographic retention index of aldehydes and ketones]

色谱保留指数(retention index, RI)是色谱分析中的重要参数,不同化合物在不同极性固定相上具有不同的保留行为。醛酮化合物种类众多,实验测定其RI值的时间和经济成本高。该论文采用集成建模(ensemble modeling)结合全息定量构效关系(HQSAR)方法研究了醛酮化合物在2种固定相(DB-210和HP-Innowax)上色谱保留指数的定量构效关系(QSAR)模型。用外部测试集验证法和留一交叉验证法评估了所建立模型的预测能力。首先建立了34种被研究化合物的个体HQSAR模型。在固定相DB-210上,片段特性(FD)为“供体/受体原子(DA)”且片段尺寸(FS)为1~9时可得到最优个体模型,在固定相HP-Innowax上,FD为“DA”且FS为4~7时可得到最优个体模型,这两个模型的交叉验证相关系数( qcv2)分别为0.935和0.909,外部验证相关系数( qext2)分别为0.925和0.927,一致性相关系数(CCC)分别为0.953和0.960,预测平方相关系数F2( QF22)分别为0.922和0.918,预测平方相关系数F3( QF32)分别为0.931和0.927。研究结果表明醛酮化合物的分子结构与RI值之间存在定量关系,用HQSAR方法可以建立二者之间的QSAR模型。其次,以4个预测准确度最高的个体HQSAR模型作为子模型通过算术平均建立了集成HQSAR模型。建立的集成HQSAR模型预测被研究化合物在DB-210和HP-Innowax固定相上RI值的 qcv2分别为0.927和0.919, qext2分别为0.929和0.963, CCC分别为0.956和0.979, QF22分别为0.927和0.958, QF32分别为0.935和0.963。与个体HQSAR模型相比,建立的集成HQSAR模型预测准确度更高。这说明集成建模是提高HQSAR模型预测能力的有效方法,HQSAR与集成建模方法相结合可以用于研究和预测醛酮化合物的RI值。

Design of 2-Nitroimidazooxazine Derivatives as Deazaflavin-Dependent Nitroreductase (Ddn) Activators as Anti-Mycobacterial Agents Based on 3D QSAR, HQSAR, and Docking Study with In Silico Prediction of Activity and Toxicity

Deazaflavin-dependent nitroreductase (Ddn) is an emerging target in the field of anti-tuberculosis agents. In the present study, 2-nitroimidazooxazine derivatives as Ddn activators were aligned for CoMFA, CoMSIA and HQSAR analysis. The best CoMFA and CoMSIA model were generated with leave-one-out correlation coefficients (q²) of 0.585 and 0.571, respectively. Both the CoMFA and CoMSIA models were also validated by a test set of 11 compounds with satisfactory [Formula: see text] value of 0.701 and 0.667, respectively. Results of 3D QSAR and HQSAR study were used for the designing of novel and potent nitroimidazooxazine derivatives as Ddn activators. 21 novel compounds were designed, and docked into the Ddn enzyme. In docking study compound ng11 showed interaction with key amino acid residues such as Tyr65 and Tyr133, and also showed better ADMET compatibility. The ADMET prediction, docking study and the predicted activity of novel designed compounds revealed that compound ng11 showed good potential as Ddn activators for the treatment of tuberculosis.