苯砜基羧酸酯类化合物急性毒性的预测模型
|
摘要
用MATLAB软件编程计算了56个苯砜基羧酸酯类化合物分子电性距离矢量(mk),同时用Hyper chem8.0程序包计算了它们的理化参数.这两类结构参数被用于建立苯砜基羧酸酯类化合物急性毒性的预测模型.通过最佳变量子集回归的方法构建多元线性回归模型:-lg EC50=4.724+30.275m7+0.061m24+6.468m85+0.880m90-0.003V-0.096(lg P)2.该模型具有良好的稳健性和较强的预测能力.以模型中的6个参数为人工神经网络(ANN)输入层,设定6∶4∶1的网络结构,用BP算法构建人工神经网络模型,其相关系数R2为0.986.结果表明,神经网络BP算法模型的预测结果优于多元线性回归模型的预测结果.
The molecular electrongativity distance vector( mk) and physicochemical parameters of 56 phenylsulfonyl carboxylates were calculated by the software of MATLAB and Hyper chem8.0 for establishing the prediction model of the acute toxicity(-lg EC50) of these compounds. The multiple liner regression( MLR) model was constructed by leaps-and-bounds regression:-lg EC50= 4.724+30.275m7+0.061m24+6.468m85+ 0. 880m90- 0. 003V- 0. 096( lg P)2. The model is highly reliable and has good predictive ability. The six structural parameters were used as the input neurons of artificial neural network,and a 6 ∶ 4 ∶ 1 network architecture was employed. A satisfied model was constructed with the back-propagation algorithm,the correlation coefficient( R2) was 0.986. It can be concluded that the prediction results of BP-ANN model are better than MLR-QSAR model.
The molecular electrongativity distance vector( mk) and physicochemical parameters of 56 phenylsulfonyl carboxylates were calculated by the software of MATLAB and Hyper chem8.0 for establishing the prediction model of the acute toxicity(-lg EC50) of these compounds. The multiple liner regression( MLR) model was constructed by leaps-and-bounds regression:-lg EC50= 4.724+30.275m7+0.061m24+6.468m85+ 0. 880m90- 0. 003V- 0. 096( lg P)2. The model is highly reliable and has good predictive ability. The six structural parameters were used as the input neurons of artificial neural network,and a 6 ∶ 4 ∶ 1 network architecture was employed. A satisfied model was constructed with the back-propagation algorithm,the correlation coefficient( R2) was 0.986. It can be concluded that the prediction results of BP-ANN model are better than MLR-QSAR model.
引文
[1]杭烨超,李方实.苯砜基羧酸酯类化合物急性毒性的QSAR研究[J].南京工业大学学报,2006,28(4):104-107.
[2]李吉来,杭烨超,耿彩云,等.苯砜基羧酸酯类急性毒性的QSAR研究[J].高等学校化学学报,2007,28(1):117-120.
[3]陈艳,岳玮,王彬.含氮杂环化合物理化性质和生物活性的QSPR/QSAR分析[J].武汉大学学报(理学版),2014,50(1):52-56.
[4]刘树深,刘堰,李志良,等.一个新的分子电性距离矢量(MEDV)[J].化学学报,2000,58(11):1 353-1 357.
[5]LIU S S,YIN C S,LI Z L,et al.QSAR study of steroid benchmark and dipeptides based on MEDV-13[J].Journal of Chemical Information and Computer Sciences,2001,41(2):321-329.
[6]LIU S S,LIU H L,YIN C S,et al.VSMP:a novel variable selection and modeling method based on the prediction[J].Journal of Chemical Information and Computer Sciences,2003,43(3):964-969.
[7]胡黔楠,梁逸曾,王亚丽,等.直观队列命名法的基本原理及其在矩阵与拓扑指数计算中的应用[J].计算机与应用化学,2003,20(4):386-390.
[8]张婷,梁逸曾,赵晨曦,等.基于分子结构预测气相色谱程序升温保留指数[J].分析化学,2006,34(11):1 607-1 610.
[9]DOUGLAS M H,SUBHASH C B,DENISE M.Assessing model fit by cross-validation[J].Journal of Chemical Information and Computer Sciences,2003,43(2):579-586.
[10]冯长君.手性有机酸保留指数的手性指数及原子类型电拓扑指数模型[J].物理化学学报,2010,26(1):193-198.
[11]冯子雅,冯长君.卤代苯对呆头鱼、发光菌急性毒性的构效关系研究[J].广州化工,2013,41(6):16-18.
[12]WANG B C,QIAN J Z,FAN Y,et al.The QSAR study of flavonoid-metal complexes scavenging OH free radical[J].Journal of Molecular Structure,2014,1 075(5):204-212.
[13]ZHENG F,ZHAN M,HUANG X Q,et al.Modeling in vitro inhibition of butyrylcholinesterase using molecular docking,multilinear regression and artificial neural network approaches[J].Bioorganic&Medicinal Chemistry,2014,22(1):538-549.
[14]堵锡华,史小琴,冯长君,等.基于野韭菜挥发性成分的色谱保留指数神经网络预测[J].山东大学学报(理学版),2014,49(1):468-473.
[15]冯长君,沐来龙,杨伟华,等.用拓扑指数和神经网络研究有机污染物的生物富集因子[J].化学学报,2008,66(19):2 093-2 098.
[16]陈艳,堵锡华,吴雅萍.用拓扑指数和神经网络研究拟除虫菊酯类农药的急性毒性[J].化学通报,2014,77(4):341-345.
[17]许禄,邵学广.化学计量学方法[M].2版.北京:科学出版社,2004:441.
[18]ANDREA T A,KALAYEH H.Applications of neural networks in quantitative structure-activity relationships of dihydrofolate reductase inhibitors[J].J Med Chem,1991,34(9),2 824-2 836.
[2]李吉来,杭烨超,耿彩云,等.苯砜基羧酸酯类急性毒性的QSAR研究[J].高等学校化学学报,2007,28(1):117-120.
[3]陈艳,岳玮,王彬.含氮杂环化合物理化性质和生物活性的QSPR/QSAR分析[J].武汉大学学报(理学版),2014,50(1):52-56.
[4]刘树深,刘堰,李志良,等.一个新的分子电性距离矢量(MEDV)[J].化学学报,2000,58(11):1 353-1 357.
[5]LIU S S,YIN C S,LI Z L,et al.QSAR study of steroid benchmark and dipeptides based on MEDV-13[J].Journal of Chemical Information and Computer Sciences,2001,41(2):321-329.
[6]LIU S S,LIU H L,YIN C S,et al.VSMP:a novel variable selection and modeling method based on the prediction[J].Journal of Chemical Information and Computer Sciences,2003,43(3):964-969.
[7]胡黔楠,梁逸曾,王亚丽,等.直观队列命名法的基本原理及其在矩阵与拓扑指数计算中的应用[J].计算机与应用化学,2003,20(4):386-390.
[8]张婷,梁逸曾,赵晨曦,等.基于分子结构预测气相色谱程序升温保留指数[J].分析化学,2006,34(11):1 607-1 610.
[9]DOUGLAS M H,SUBHASH C B,DENISE M.Assessing model fit by cross-validation[J].Journal of Chemical Information and Computer Sciences,2003,43(2):579-586.
[10]冯长君.手性有机酸保留指数的手性指数及原子类型电拓扑指数模型[J].物理化学学报,2010,26(1):193-198.
[11]冯子雅,冯长君.卤代苯对呆头鱼、发光菌急性毒性的构效关系研究[J].广州化工,2013,41(6):16-18.
[12]WANG B C,QIAN J Z,FAN Y,et al.The QSAR study of flavonoid-metal complexes scavenging OH free radical[J].Journal of Molecular Structure,2014,1 075(5):204-212.
[13]ZHENG F,ZHAN M,HUANG X Q,et al.Modeling in vitro inhibition of butyrylcholinesterase using molecular docking,multilinear regression and artificial neural network approaches[J].Bioorganic&Medicinal Chemistry,2014,22(1):538-549.
[14]堵锡华,史小琴,冯长君,等.基于野韭菜挥发性成分的色谱保留指数神经网络预测[J].山东大学学报(理学版),2014,49(1):468-473.
[15]冯长君,沐来龙,杨伟华,等.用拓扑指数和神经网络研究有机污染物的生物富集因子[J].化学学报,2008,66(19):2 093-2 098.
[16]陈艳,堵锡华,吴雅萍.用拓扑指数和神经网络研究拟除虫菊酯类农药的急性毒性[J].化学通报,2014,77(4):341-345.
[17]许禄,邵学广.化学计量学方法[M].2版.北京:科学出版社,2004:441.
[18]ANDREA T A,KALAYEH H.Applications of neural networks in quantitative structure-activity relationships of dihydrofolate reductase inhibitors[J].J Med Chem,1991,34(9),2 824-2 836.