茶叶上农药生物降解性的电性距离矢量预测  

Biodegradability Prediction of Pesticides in Tea Plant Using Molecular Electronegativity Distance Vector

作　　者：冯长君[1,2] 

机构地区：[1]徐州工程学院化学化工学院,江苏徐州221008 [2]徐州市生物质燃料工程技术研究中心,江苏徐州221008

出　　处：《南京理工大学学报》2011年第5期722-725,共4页Journal of Nanjing University of Science and Technology

基　　金：国家自然科学基金(21075138);江苏省高校自然科学基金(08KJD610003);贾汪科技局基金(XM10A05)

摘　　要：为了建立茶叶上农药的生物降解性与分子电性距离矢量MK的定量构效关系模型(QSAR),揭示影响生物降解性的结构因素,运用分子MK关联11种农药(含拟除虫菊酯和有机磷农药)在茶叶上的生物降解半衰期t0.5,经逐步回归建立了最佳二元数学模型。传统的判定系数R2为0.954,逐一剔除法的交互验证系数Q2为0.923,证明该模型具有良好的稳健性及预测能力。根据进入该模型的2个电性距离矢量M10、M20可知,影响农药生物降解性的主要因素是分子的二维结构特征-CH3,-O-,-C-,P等结构碎片。以M10,M20为人工神经网络的输入层,设定2∶2∶1的网络结构,所建BP模型的传统相关系数R2为0.977。研究结果表明,电性距离矢量对部分农药生物降解性的表征是合理有效的。In order to study the quantitative structure-activity relationship（QSAR）between biodegradability and molecular electronegativity distance vector MK of pesticides in tea plant,and analyze the decisive factors affecting the biodegradation half lives of pesticides,a two-variabe quantitative structure-biodegradability relationship（QSBR）model is established by applying the molecular electronegativity distance vector MK to simulate the biodegradation half lives t0.5 of 11 pesticides（such as pyrethroid and organophosphorus compounds）in tea plant.The traditional correlation coefficient R2 and the cross-validation correlation coefficient Q2 of leave-one-out are 0.954 and 0.923.The result demonstrates that the model is reliable and has good predictive ability.From the two parameters M10,M20 of the model,it can be seen that the 2D-molecular structure characteristics,such as-CH3,-O-,-C-,P,are the decisive factors affecting the half lives of the pesticides.The two structure parameters are used as the input neurons of artificial neural network,and the 2∶2∶1 network architecture is employed.A satisfying model can be constructed with the back-propagation algorithm,and the correlation coefficient R2 is 0.977.The results show that MK has good rationality and efficiency for predicting the half lives of some pesticides.

关 键 词：农药 生物降解 半衰期 电性距离矢量 多元回归分析 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论] X172[自动化与计算机技术—计算机科学与技术]