毕业设计说明书中文摘要LogKpew为研究疏水性有机污染物(HOCs)进行了描述。前人通过建立QSAR模型来预测logKpew，进行了一些有意义的探索。在环境科学中，QSAR指关联有机污染物分子结构与其理化性质、环境行为和毒理学参数的定量预测模型。QSAR可以弥补基础数据的缺失、降低昂贵的测试费用、减少实验次数。78507

     本研究依照OECD关于QSAR模型构建和使用的导则，首先采用线性溶解能关系（LSER）采用逐步回归法建立QSAR模型，得到了较好的结果（相关系数R2=0。94，交叉验证Q2LOO=0。91）为了提高模型的预测精准度，根据LSER理论选择理论计算的分子结构描述符，采用最小二乘法建立QSAR模型，其具有良好的拟合能力(R2=0。919 RMSEtra=0。379),模型有良好的稳健性（交叉验证Q2tra=0。918）和预测能力（Q2ext=0。755  RMSEext=0。385）。最后采用William图和欧几里得距离图法对模型的应用域进行了表征。所建立的QSAR模型，可以应用于应用域内有机化合物的logKpew的预测，具有潜在应用价值。

毕业论文关键词 logKpew 有机污染物  应用域  QSAR模型

毕业设计说明书外文摘要

Title  Establishment and evaluation of a quantitative model for the equilibrium partition coefficient of organic pollutants in the passive sampling materials                 

Abstract LogKpew were reviewed for trace hydrophobic organic contaminants (HOCs)。 Predecessors by building QSAR models to predict logKpew, made some meaningful exploration。 In environmental science, QSAR quantitative prediction model refers to the molecular structure of organic pollutants associated with their physicochemical properties,environmental behavior and toxicological parameters。 QSAR can compensate for the lack of basic data and reduce expensive testing costs, reduce the number of experiments。

    In this study, in accordance with the OECD on QSAR model building and use of guidelines, linear solvation energy relationship (LSER) QSAR models were established by stepwise regression method to obtain good results (correlation coefficient R2 = 0。94, cross-validation Q2LOO = 0。91) in order to improve the prediction accuracy of the model, select the theoretical calculations based on LSER theoretical molecular descriptors, QSAR models were established using the least squares method, which has a good ability to fit (R2 = 0。919 RMSEtra = 0。379), the model has good robustness (cross-validation Q2tra = 0。918) and predictive capacity (Q2ext = 0。755 RMSEext = 0。385)。 Finally, William Euclidean distance map and the application domain models were characterized。 QSAR model established, the application can be used to predict the domain of organic compounds logKpew having potential applications。

Keywords： logKpew   Organic Pollutants   AD   QSAR

目 次

1 引言 1

1。1 被动采样技术的重要性以及材料的选择 1

1。2平衡分配常数 3

1。3 QSAR模型的优点及其类型 4

2。材料与方法 5

2。1数据来源 5

2。2数据处理方式 14

2。3描述符的计算 21

2。4 QSAR模型的建立 21

2。5 QSAR模型离域点的诊断 24

3。低疏水性数据集的QSAR模型 26

3。1 QSAR模型的建立及其相关系数