AIM: To develop an artificial neural network (ANN) model for predicting the resistance index (RI) of taxoids. METHODS: A dataset of 63 experimental data points were compiled from literatures and subdivided into traini...AIM: To develop an artificial neural network (ANN) model for predicting the resistance index (RI) of taxoids. METHODS: A dataset of 63 experimental data points were compiled from literatures and subdivided into training and external test sets randomly. Electrotopological state (E-state) indices were calculated to characterize molecular structure, together with a principle component analysis to reduce the variable space and analyze the relative importance of E-state indices. Back propagation neural network (BPNN) technique was used to build the models. Five-fold cross validation was performed and five models with different compounds composition in training and validation sets were built. The independent external test set was used to evaluate the predictive ability of models. RESULTS: The final model was proved to be good with the cross validation Qcv2 0.62, external testing R2 0.84 and the slope of the regression line through the origin for testing set is 0.9933. CONCLUSION: The QSAR model can predict the RI to a relative nicety, which will aid in the development of new anti-MDR taxoids.展开更多
基于拓扑化学理论,原子类型电拓扑态指数( E n )和电性距离矢量( M k )被用于表征19种苯基噻唑衍生物的化学微环境。采用最佳变量子集回归方法,分别建立上述化合物对HT29(人结肠癌细胞系)、Hela(人宫颈癌细胞系)和Karpas299(人淋巴瘤细...基于拓扑化学理论,原子类型电拓扑态指数( E n )和电性距离矢量( M k )被用于表征19种苯基噻唑衍生物的化学微环境。采用最佳变量子集回归方法,分别建立上述化合物对HT29(人结肠癌细胞系)、Hela(人宫颈癌细胞系)和Karpas299(人淋巴瘤细胞系)的体外细胞增殖抑制活性( L i :L t , L e , L a )与 E n 和 M k 的定量构效关系(QSAR)模型。它们的最佳三元QSAR模型的判定系数( R 2)依次为0.865、0.924、0.937,逐一剔除法交叉验证相关系数( R cv^2)依次为0.679、0.898、0.878。经 R cv 2、 V IF 、 F IT 、 A IC 等统计指标检验,该模型具有良好的稳健性及预测能力。结果显示—CH 3、—CH2—、—NH—、—N=(芳环中)和—OH等分子结构单元直接影响这些化合物的抑制活性。据此提出苯基噻唑衍生物分子与生物受体之间的主要作用是疏水作用、氢键。此与文献[1]的分子对接结果基本一致。展开更多
文摘AIM: To develop an artificial neural network (ANN) model for predicting the resistance index (RI) of taxoids. METHODS: A dataset of 63 experimental data points were compiled from literatures and subdivided into training and external test sets randomly. Electrotopological state (E-state) indices were calculated to characterize molecular structure, together with a principle component analysis to reduce the variable space and analyze the relative importance of E-state indices. Back propagation neural network (BPNN) technique was used to build the models. Five-fold cross validation was performed and five models with different compounds composition in training and validation sets were built. The independent external test set was used to evaluate the predictive ability of models. RESULTS: The final model was proved to be good with the cross validation Qcv2 0.62, external testing R2 0.84 and the slope of the regression line through the origin for testing set is 0.9933. CONCLUSION: The QSAR model can predict the RI to a relative nicety, which will aid in the development of new anti-MDR taxoids.
文摘基于拓扑化学理论,原子类型电拓扑态指数( E n )和电性距离矢量( M k )被用于表征19种苯基噻唑衍生物的化学微环境。采用最佳变量子集回归方法,分别建立上述化合物对HT29(人结肠癌细胞系)、Hela(人宫颈癌细胞系)和Karpas299(人淋巴瘤细胞系)的体外细胞增殖抑制活性( L i :L t , L e , L a )与 E n 和 M k 的定量构效关系(QSAR)模型。它们的最佳三元QSAR模型的判定系数( R 2)依次为0.865、0.924、0.937,逐一剔除法交叉验证相关系数( R cv^2)依次为0.679、0.898、0.878。经 R cv 2、 V IF 、 F IT 、 A IC 等统计指标检验,该模型具有良好的稳健性及预测能力。结果显示—CH 3、—CH2—、—NH—、—N=(芳环中)和—OH等分子结构单元直接影响这些化合物的抑制活性。据此提出苯基噻唑衍生物分子与生物受体之间的主要作用是疏水作用、氢键。此与文献[1]的分子对接结果基本一致。
