部分有机污染物构效关系的研究
摘要
在环境化学中，有机污染物定量结构活性/性质相关(QSAR/QSPR)对于有机化合物的生态风险性评价、污染控制和预防等具有十分重要的意义。
量子化学计算是获得分子结构参数的重要手段。量子化学参数具有明确的物理化学意义，在有机污染物的QSAR/QSPR研究中，可用于探讨毒物与受体的作用方式，也可用于研究影响有机污染物理化性质的分子结构特征。量子化学计算方法中的密度泛函方法(DFT)在理论上很严格，已成为量子化学计算的主流。分子连接性指数是QSAR/QSPR研究中的另一类重要结构参数，其可以实现分子结构的定量描述，已广泛地用于有机化合物的QSAR/QSPR研究中。
温度限制串联相关网络以快速、强大及自组织结构而被设计，在串联相关网络的基础上引入了温度限制的概念，解决了过度训练的问题。Mark在径向基函数神经网络训练过程中引入前向选择而设计了改进的径向基函数神经网络，这样可以优化径向基函数的宽度，以控制模型的复杂性和性能。支持向量机是一种新的机器学习方法，其有良好的理论基础和泛化能力。本论文将这些方法引入到环境化学中，构建QSAR/QSPR模型，预测环境化学中的有机污染物的毒性及有机物的物理性质。
本论文的第一章简述了定量结构活性/性质相关的发展过程及研究现状。
在第二章中，介绍了定量结构活性/性质相关所使用的参数和研究方法，详细描述了改进的径向基函数神经网络、温度限制串联相关网络及支持向量机的原理及应用，这些理论和现状分析为我们开展本论文的研究工作提供了理论基础和依据。
在第三章中，我们采用B3LYP杂化密度泛函理论方法计算了35个硝基苯及其同系物的量子化学结构参数，通过逐步回归得到具有显著统计意义的QSAR方程，相关系数是0.925，交叉验证的相关系数是0.87。首次将TCCCN应用到QSAR研究中。用主成分分析选择参数，建立了BP网络和TCCCN网络非线性模型，其训练集的MSE分别为0.095和0.067，预测集的MSE分别为0.111和0.090。非线性的TCCCN模型较线性的MLR模型有更好的预测能力。
在第四章中，我们从分子结构计算分子的连接性指数出发，计算了25个酚类化合物的分子连接性指数，用逐步回归方法建立了4个参数的最佳方程，以此4个参数作为输入参数，将留一法（LOO）应用到BP网络、RBF网络及新颖的机器学习方法SVM中，建立了酚类化合物对黑呆头鱼的QSAR预测模型。应用非线性SVM方法建立模型的结果优于BP网络和RBF网络模型的结果，SVM、BP、RBF模型预测的相关系数分别为0.959，0.940和0.945，得到满意的结果。
在第五章中，我们采用密度泛函理论（DFT）方法计算了60个醇类化合物的量子化学结构参数，同时又计算了分子连接性指数，将量化参数和分子连接性指数联合应用到醇类的溶解度和辛醇/水分配系数的QSPR研究中，分别通过逐步回归得到具有显著统计意义的4个参数和5个参数的QSPR方程。以此4个参数和5个参数分别作为输入参
Quantitative structure-activity/property relationship（QSAR/QSPR）of organicpollutants is of great importance to ecological risk assessments of organic compounds,pollution control and pollution prevention, etc.
Quantum chemical calculation is an important way to get structural parameters ofspecific molecules in the QSAR/QSPR study. Quantum chemical parameters have explicitphysical chemistry interpretation, and they can be used in not only discussing effect modebetween toxicity and acceptor but also studying the molecular characters affecting physicalchemistry property of organic pollutants. Due to Density Functional Methods (DFT) inquantum chemistry calculation methods have very strict theory bases, they have become aneffective tool in quantum chemistry calculation worldwide. Molecular connectivity index isanother important structure parameters in QSAR/QSPR study. Because they can describemolecular structure in quantity, they have come into wide use in QSAR/QSPR study.
Temperature-constrained cascade correlation network (TCCCN) was devised based onfast, strong and self-organizational architecture. The use of temperature constraints in cascadecorrelation network can solve the effects of overfitting. Mark devised an improved radialbasis function neural network (RBFNN) based on forward selection, which can optimize theRBF widths to control model complexity. Support Vector Machine (SVM) is a novel type ofmachine learning method; it has rigorous theory background and remarkable generalizationperformance. This dissertation introduced these methods to environmental chemistry to buildQSAR/QSPR model, predict the toxicity of organic pollutants and physical properties oforganic compounds.
A brief description of QSAR/QSPR realization process and research status was given inChapter 1 of this dissertation.
In Chapter 2, firstly we introduced the parameters and research methods used inQSAR/QSPR. Then we described the principle of improved RBFNN, TCCCN and SVM indetail. At last we gave a review of the application of these methods, respectly.
In Chapter 3, TCCCN, back-propagation neural network (BP) and multiple linearregression (MLR) were applied to QSAR modeling based on a set of 35 nitrobenzenederivatives and their acute toxicities. These structure quantum-chemical descriptors wereobtained from density functional theory (DFT). Stepwise multiple regression analysis wasperformed and model was obtained. The conventional R was 0.925, and cross-validation Rwas 0.87.The principal component analysis is used for parameter selection. RMS for trainingset using TCCCN and BP were 0.067, 0.095 respectively, and RMS for testing set were 0.090,
量子化学计算是获得分子结构参数的重要手段。量子化学参数具有明确的物理化学意义，在有机污染物的QSAR/QSPR研究中，可用于探讨毒物与受体的作用方式，也可用于研究影响有机污染物理化性质的分子结构特征。量子化学计算方法中的密度泛函方法(DFT)在理论上很严格，已成为量子化学计算的主流。分子连接性指数是QSAR/QSPR研究中的另一类重要结构参数，其可以实现分子结构的定量描述，已广泛地用于有机化合物的QSAR/QSPR研究中。
温度限制串联相关网络以快速、强大及自组织结构而被设计，在串联相关网络的基础上引入了温度限制的概念，解决了过度训练的问题。Mark在径向基函数神经网络训练过程中引入前向选择而设计了改进的径向基函数神经网络，这样可以优化径向基函数的宽度，以控制模型的复杂性和性能。支持向量机是一种新的机器学习方法，其有良好的理论基础和泛化能力。本论文将这些方法引入到环境化学中，构建QSAR/QSPR模型，预测环境化学中的有机污染物的毒性及有机物的物理性质。
本论文的第一章简述了定量结构活性/性质相关的发展过程及研究现状。
在第二章中，介绍了定量结构活性/性质相关所使用的参数和研究方法，详细描述了改进的径向基函数神经网络、温度限制串联相关网络及支持向量机的原理及应用，这些理论和现状分析为我们开展本论文的研究工作提供了理论基础和依据。
在第三章中，我们采用B3LYP杂化密度泛函理论方法计算了35个硝基苯及其同系物的量子化学结构参数，通过逐步回归得到具有显著统计意义的QSAR方程，相关系数是0.925，交叉验证的相关系数是0.87。首次将TCCCN应用到QSAR研究中。用主成分分析选择参数，建立了BP网络和TCCCN网络非线性模型，其训练集的MSE分别为0.095和0.067，预测集的MSE分别为0.111和0.090。非线性的TCCCN模型较线性的MLR模型有更好的预测能力。
在第四章中，我们从分子结构计算分子的连接性指数出发，计算了25个酚类化合物的分子连接性指数，用逐步回归方法建立了4个参数的最佳方程，以此4个参数作为输入参数，将留一法（LOO）应用到BP网络、RBF网络及新颖的机器学习方法SVM中，建立了酚类化合物对黑呆头鱼的QSAR预测模型。应用非线性SVM方法建立模型的结果优于BP网络和RBF网络模型的结果，SVM、BP、RBF模型预测的相关系数分别为0.959，0.940和0.945，得到满意的结果。
在第五章中，我们采用密度泛函理论（DFT）方法计算了60个醇类化合物的量子化学结构参数，同时又计算了分子连接性指数，将量化参数和分子连接性指数联合应用到醇类的溶解度和辛醇/水分配系数的QSPR研究中，分别通过逐步回归得到具有显著统计意义的4个参数和5个参数的QSPR方程。以此4个参数和5个参数分别作为输入参
Quantitative structure-activity/property relationship（QSAR/QSPR）of organicpollutants is of great importance to ecological risk assessments of organic compounds,pollution control and pollution prevention, etc.
Quantum chemical calculation is an important way to get structural parameters ofspecific molecules in the QSAR/QSPR study. Quantum chemical parameters have explicitphysical chemistry interpretation, and they can be used in not only discussing effect modebetween toxicity and acceptor but also studying the molecular characters affecting physicalchemistry property of organic pollutants. Due to Density Functional Methods (DFT) inquantum chemistry calculation methods have very strict theory bases, they have become aneffective tool in quantum chemistry calculation worldwide. Molecular connectivity index isanother important structure parameters in QSAR/QSPR study. Because they can describemolecular structure in quantity, they have come into wide use in QSAR/QSPR study.
Temperature-constrained cascade correlation network (TCCCN) was devised based onfast, strong and self-organizational architecture. The use of temperature constraints in cascadecorrelation network can solve the effects of overfitting. Mark devised an improved radialbasis function neural network (RBFNN) based on forward selection, which can optimize theRBF widths to control model complexity. Support Vector Machine (SVM) is a novel type ofmachine learning method; it has rigorous theory background and remarkable generalizationperformance. This dissertation introduced these methods to environmental chemistry to buildQSAR/QSPR model, predict the toxicity of organic pollutants and physical properties oforganic compounds.
A brief description of QSAR/QSPR realization process and research status was given inChapter 1 of this dissertation.
In Chapter 2, firstly we introduced the parameters and research methods used inQSAR/QSPR. Then we described the principle of improved RBFNN, TCCCN and SVM indetail. At last we gave a review of the application of these methods, respectly.
In Chapter 3, TCCCN, back-propagation neural network (BP) and multiple linearregression (MLR) were applied to QSAR modeling based on a set of 35 nitrobenzenederivatives and their acute toxicities. These structure quantum-chemical descriptors wereobtained from density functional theory (DFT). Stepwise multiple regression analysis wasperformed and model was obtained. The conventional R was 0.925, and cross-validation Rwas 0.87.The principal component analysis is used for parameter selection. RMS for trainingset using TCCCN and BP were 0.067, 0.095 respectively, and RMS for testing set were 0.090,
引文
[1]沈德中. 污染环境的生物修复[M]. 北京：化学工业出版社，2002. 1-4.
[2]Blum D J W, Speece R E. Edtermining chemical toxicity of aquatic species[J]. EnvironSci Technol, 1990, 24: 284-293.
[3]王连生. 有机污染物化学[M]:上册. 北京：科学出版社，1990.1.
[4]梁逸曾, 俞汝勤. 分析化学手册[M]: 第十分册, 化学计量学. 北京: 化学工业出版社, 2000.392.
[5]Hansch C, Maloney P P, Fujita T, et al. Correlation of biological activity ofphenoxyacetic acids with Hammett substituent constants and partition coefficients [J].Nature, 1962,194: 178-180.
[6]Hancch C, Muir R M, Fujita T, et al. The correlation of biological activity of plantgrowth regulators and chloromycetin derivatives with Hammett constants and partitioncoefficient [J]. J Am Chem Soc, 1963, 85(18): 2817-2824.
[7]Fujita T, Iwasa J, Hansch C A new substituent constant, π,derived from partitioncoefficients[J]. J Am Chem Soc, 1964, 86(23): 5175-5180.
[8]Free S M, Wilson J M. A mathematical contribution to structure-activity studies [J].J Med Chem, 1964, 7(4): 395-399.
[9]王连生. 分子结构、性质与活性[M]. 北京：化学工业出版社，1997.
[10]唐桂刚，白乃彬. 遗传神经网络在 QSAR 中的应用研究[J]. 计算机与应用化学，1999, 16(6):435-440.
[11]周家驹，陈红明，谢桂荣，等. 三维定量结构-活性关系研究中的受体作用位点模型方法[J]. 化学进展, 1998,10(1): 55-62.
[12]李江波, 谢华庆, 张敬畅, 等. dT4 核苷酸类似物抗 HIV-1 活性的定量构效关系[J]. 北京化工大学学报, 2001, 28(1): 79-80.
[13]Suh Myung-Eun, Kang Min-Jung, Park So-Young. The 3-D QSAR Study of Anticancer1-N-substituted Imidazo- and Pyrrolo-quinoline-4,9-dione Derivatives by CoMFA andCoMSIA[J]. Bioorg Med Chem, 2001, 9: 2987-2991.
[14]李华, 许禄, 苏锵, 等. 甾体化合物的结构与抗炎活性相关性 3D-QSAR 研究[J]. 计算机与应用化学, 1997, 14(1): 27-30.
[15]Cong V Nguyen-, Dang G Van, BM Rode. Using Multivariate Adaptive Regression Splinesto QSAR Studies of Dihydroartemisinin Derivatives [J]. Eur J Med Chem, 1996, 31: 797-803.
[16]Kowalski B R, Bender C F. The application of pattern recognition to screeningprospective anticancer drugs [J]. J Am Chem Soc, 1974, 96:916-918.
[17]Lipnick R L, Overton C E. Narcosis studies and a contribution to general pharmacology[J]. Trends Pharmacol Sci, 1986, 7: 161–164.
[18]Cocchi M., Benedetti P G De. Use of the Supermolecule Approach to Derive MolecularSimilarity Descriptors for QSAR Analysis [J]. J Mol Model, 1998, 4: 113 – 131.
[19]蔡煜东, 杨兵, 孙红. 用自组织神经网络研究抗肿瘤药卡巴醌构效关系[J]. 计算机与应用化学, 1995, 12: 255-266.
[20]冯长君. 无机物晶格能及其它理化性质的定量构效关系研究[J]. 分析科学学报, 2002, 18(1):27-32.
[21]赵瑞环, 岳丙方, 单亦初, 等. 神经元网络用于 PCDD 定量构效关系的研究[J]. 色谱, 1999,17(2): 112-114.
[22]Kamlet M J, Doherty R M, Veith G D, et al. Solubility properties in polymers andbiological media[J]. Environ Sci Technol, 1986, 20(7): 690-695.
[23]Ovidiu Ivanciuc. The Neural Network MolNet Prediction of Alkane Enthalpies [J]. AnalChim Acta, 1999, 384: 271-284.
[24]Aleksandar S. QSAR Models for Estimating Properties of Present Organic PollutantsRequired in Evaluation of Their Environmental Fate and Risk [J]. Chemosphere, 2001, 43:363-375.
[25]Loukas Y L. Radial Basis Function Networks in Host-Guest Interactions: Instant andAccurate Formation Constant Calculations [J]. Anal Chim Acta, 2000, 417: 221-22.
[26]李颖娇, 叶非. 定量构效关系在农药设计合成中的应用进展[J]. 农药科学与管理, 2002,23(6): 20-23.
[27]Kishino T, Kobayshi K. Studies on the Mechanism of Toxicity of Chlorophenols Found inFish through Quantitative Structure-Activity Relationship [J]. Wat Res, 1996, 30(2):393-399.
[28]Fang Hong, Tong Weida, Welsh W J, et al. QSAR Models in Receptor-Mediated Effects: theNuclear Receptor Superfamily [J]. J Mol Struct (Theochem), 2003, 622: 113-125.
[29]Debnath K, Hansch C, Kim H K, et al. Mechanisic interpretation of the genotoxicity ofnitrofurans (antibacterial age nts) using quantitative structure-activity relationshipsand comparative molecular field analysis [J]. J Med Chem, 1993, 36:1007-1016.
[30]Norinder Ulf. Support Vector Machine Models in Drug Design: Applications to DrugTransport Processes and QSAR Using Simplex Optimizations and Variable Selection [J].Neurocomputing, 2003, 55: 337-346.
[31]Lewis D F V. Structural Characteristics of Human P450s Involved in Drug Metabolism:QSARs and Lipophilicity Profiles [J]. Toxicology, 2000, 144: 197-203.
[32]Dvorsky R, Balaz S, Sawchuk R J. Kinetics of Subcellular Distribution of Compounds inSimple Biosystems and its Use in QSAR [J]. J Theor Biol, 1997, 185: 213-222.
[33]高大文, 王鹏, 郑彤, 等. 多氯酚定量构效关系人工神经网络信息流分析[J]. 中国环境科学,2002, 22(6):561-564.
[34]Basak S C, Gute B D, Mills D, et al. Quantitative Molecular Similarity Methods in theProperty/Toxicity Estimation of Chemicals: a Comparison of Arbitrary versus TailoredSimilarity Spaces[J]. J Mol Struct (Theochem), 2003, 622: 127-145.
[35]Tuppurainen K. Frontier Orbital Energies, Hydrophobicity and Steric Facrtors asPhysical QSAR Descriprtors of Molecular Mutagenicity[J]. Chemosphere, 1999, 38(13):3015-3030.
[36]Walker J D. International workshops on QSARs in the environmental sciences-The first20 years [J]. QSAR & Combinatorial Science, 2003,22(4): 415-421.
[37]曲凡歧，戴志群，胡潇文，等. 植物抗病毒剂的探寻Ⅹ--1,5-二丙酸酯-1,3,5-均三嗪-2,4-二酮的合成[J]. 武汉大学学报（理学版），2002, 48(4): 435-428.
[38]周家驹，王亭. 药物设计中的非模型化方法[M]. 北京：科学出版社，2001. 104.
[39]Yuan H, Parrill A L. QSAR development to describe HIV-1 integrase inhibition [J]. JMol Strut (Themochem), 2000,529:273-282.
[40]Li H, Ung C Y , Yap C W , et al. Predictio of Genotoxicity of Chemical Compounds byStatistical Learnng Methods [J]. Chem Res Toxicol, 2005,18:1071-1080.
[41]杨光富，刘华银，杨秀凤，等. 以 ALS 为把标的新型除草剂的分子设计、合成及生物活性研究.VIII. Hansch 方法与 CoMFA 方法相结合研究稠杂磺酰胺类除草剂的构效关系 [J]. 化学学报，1999，57：706-711.
[42]王连生，支正良，高松亭. 分子结构于色谱保留[M]. 北京: 化学工业出版社，1994.2、113-119.
[43]Boyce C B C., Millborrow B V. A Simple Assessment of Partition Data for CorrelatingStructure and Biological Activity using Thin-layer Chromatography [J]. Nature, 1985, 537.
[44]罗北平，彭晓春，任碧野. 有机物气相色谱保留指数的 QSPR 研究[J]. 吉首大学学报（自然科学版）, 2000, 21(1): 43-49.
[45]刘够生，宋兴福，于建国，等. 人工神经网络模拟芳基脲土壤吸附系数研究[J]. 环境科学学报,2002, 22(3): 359-363.
[46]Tantishaiyakul Vimon. Prediction of the aqueous solubility of benzylamine salts usingQSPR model [J]. J Pharmaceut Biomed, 2005,37: 411-145.
[47]Raevsky Oleg A, Schaper Klaus J. Analysis of water solubility data on the basis of HYBOTdescriptors [J]. QSAR Comb Sci,(QSAR & Combinatorial Science,) 2003,22: 926-942.[QSARg31]
[48]Uddameri Venkatesh, Kuchanur Muthukumar. Fuzzy QSARs for predicting logKoc of persistentorganic pollutants [J]. Chemosphere, 2004,54: 771-776.
[49]Gramatica Paola, Giani Elisa, Papa Ester. Statistical external validation and consensusmodeling: A QSPR case study for Koc predictio [J]. Molecular Graphics and Modeling, 2006,in press.
[50]Schultz T Wayne, Mark T D Cronin, John D Walker, et al. Quantitative Structure-ActivityRelationships (QSARs) in Toxicology: a Historical Perspective [J]. J Mol Struct (Theochem),2003, 622: 1-22.
[51]刘征涛，张颖，徐静波，等. 烷基酚类的生殖干扰毒性与结构相关研究[J]. 环境科学研究,2002,15(6): 39-48.
[52]Toropov Andrey A, Toropova Alla P. QSAR modeling of toxicity on optimization ofcorrelation weights of Morgan extended connectivity [J]. J Mol Struct (Theochem), 2002,578:129-134.
[53]Argese Emanuele, Bettiol Cinzia, Fasolo Matteo, et al. Substituted aniline interactionwith submitochondrial particles and quantitative structure-activity relationshios [J].Biochimica et Biophysica Acta, 2002, 1558: 151-160.
[54]Kapur S, Shusterman A, Verma R P, et al. Toxicology of Benzyl Alcohols: a QSAR Analysis[J]. Chemosphere, 2000, 41: 1643-1649.
[55]Gokhale V M, Kulkarni V M. Understanding the Antifungal Activity of TerbinafineAnalogues Using Quantitative Structure–Activity Relationship (QSAR) Models [J]. BioorgMed Chem, 2000, 8: 2487-2499.
[56]Roy D R, Parthasarath R, Maiti B,et al. Electrophilicity as a possible descriptor fortoxicity prediction[J]. Bioorg Med Chem,2005,13: 3405-3412.
[57]Niederlehner B R, Cairns John, Jr, et al. Modeling Acute and Chronic Toxicity of NonpolarNarcotic Chemicals and Mixtures to Ceriodaphnia dubia [J]. Ecotoxicology and EnvironmentalSafety, 1998,39:136-146.
[58]Juranic Ivan O, Drakulic Branko J, Petrovic Slobodan D, et al. A QSAR study of axutetoxicity of N-substituted fluoroaceamides to rats [J]. Chemosphere, 2006, 62:641-649.
[59]Clare Brian W. QSAR of aromatic substances: Toxicity of polychlorodibenzofurans [J].Theochem, 2006,763: 205-213.
[60]Lin Zhifen, Yu Hongxia, Wei Dongbin, et al. Prediction of mixture toxicity with itstotal hydrophobicity [J]. Chemosphere, 2002,46:305-310.
[61]Pololth Claudia, Mangelsdorf Inge. Commentary on the application of QSAR to thetoxicological evaluation of existing chemicals [J]. Chemosphere, 1997,35(11): 2525-2542.
[62]Zhang Daren. QSPR studies of PCBs by the combination of genetic algorithms and PLSanalysis [J]. Comput Chem, 2001, 25: 197-204.
[63]Veith Gilman D, Mekenyan Ovanes G, Ankley Gerald T, et al. A QSAR analysis of substituenteffects on the photoinduced acute toxicity of PAHs [J]. Chemosphere, 1995,30(11):2129-2142.
[64]Sztandera Les, Garg Ashish, Hayik Seth, et al. Mutagenicity of aminoazo dyes and theirreductive-cleavage metabolites: a QSAR/QSPR investigation [J]. Dyes and Pigments, 2003,59: 117-133.
[65]Abraham M H, Kumarsingh R, Cometto-Muniz J E W S Cain. A Quantitative Structure-ActivityRelationship (QSAR) for a Draize Eye Irritation Database [J]. Toxicol in Vitro, 1998, 12:201-207.
[66]Lewis D F V, Lake B G. Quantitative Structure-Activity Relationship(QSAR) Analysis fora Series of Rodent Peroxisome Proliferators: Interaction with the Mouse Liver PerpxisomeProliferator-activated Receptor α(mPPARα)[J]. Toxicol in Vitro, 1997, 11: 99-105.
[67]Argese E, Bettiol C, Giurin G, et al. Quantitative Structure-Activity Relationshipsfor the Toxicity of Chlorophenols to Mammalian Submitochondrial Particles [J]. Chemosphere,1999, 38(13): 2281-2292.
[68]Ashby J, Tennant R W. Chemical Structure, Salmonella Mutagenicity and Extent ofCarcinogenicity as Indicators of Genotoxic Carcinogenesis among 222 Chemicals tested inRodents by the U.S. NCI/NTP [J]. Mut Res, 1988, 204: 17-115.
[69]Wang Xiaodong, Dong Yuying, Wang Liangsheng, et al. Acute toxicity of SubstitutedPhenols to Rana japonica tadpoles and Mechanism-based Quantitative Structure-activityRelationship (QSAR) Study [J]. Chemosphere, 2001, 44: 447-455.
[70]Ramos E Urrestarazu, Vaes W H J, Mayer P, et al. Algal growth inhibition of Chlorellapyrenoidosa by polar narcotic pollutants: toxic cell concentrations and QSAR modeling [J].Aquat Toxicol, 1999, 46: 1–10.
[71]Furay V J, Smith S. Toxicity and QSAR of Chlorobenzenes in Two Species of BenthicFlatfish, Flounder (Platichthys flesus L.) and Sole(Solea solea L.)[J]. Bull Enviorn Contam.Toxicol, 1995, 54(1): 36-42.
[72]袁星, 袁晓凡, 赵元慧. 取代苯胺、苯酚对鲤鱼毒性的定量构效关系[J]. 东北师大学报（自然科学版）, 2001, 33(1): 70-73.
[73]Kaiser K L E. The Use of Neural Networks in QSARs for Acute Aquatic ToxicologicalEndpoints [J]. J Mol Struct (Theochem), 2003, 622: 85-95.
[74]胥江河. 氯代苯对水生生物毒性的定量构效关系研究[J]. 渝州大学学报(自然科学版), 2000,17(3): 66-70.
[75]Tao S, Xi Xiaohuan, Xu Fuliu, et al. A QSAR model for predicting toxicity (LC 50) torainbow trout [J]. Wat Res, 2002, 36: 2926-2930.
[76]Diane J Blum W, Speece R E. Determining chemical toxicity to aquatic species [J]. EnvironSci Technol, 1990, 24(3): 284-293.
[77]Netzeva Tatiana I, Schultz T Wayne. QSARs for the aquatic toxicity of aromatic aldehydesfrom Tetrahymena data[J]. Chemosphere, 2005, 61: 1632-1643.
[78]Dimitrov S D, Mekenyan O G, Sinks G D, et al. Global Modeling of Narcotic Chemicals:Ciliate and Fish Toxicity [J]. J Mol Struct (Theochem), 2003, 622: 63–70.
[79]Bundy J G, Morriss A W J, Durham D G, et al. Development of QSARs to Investigate theBacterial Toxicity and Biotransformation Potential of Aromatic Heterocylic Compounds[J].Chemosphere, 2001, 42: 885-892.
[80]王连生, 黄庆国, 韩朔睽等. 有机污染化学进展[M], 北京: 化学工业出版社, 1998, 164-168.
[81]黄鸿新, 卢忠, 罗一帆. 氯代酚类化合物对发光细菌毒性的构效关系研究[J]. 西江大学学报,2000, 2: 69-72.
[82]刘够生，宋兴福，于建国，等. 氯代酚类衍生物对水生物发光细菌的定量结构—活性关系研究[J]. 江西师范大学学报（自然科学版），2001，25（4）：313-316.
[83]李卫华, 许旋, 徐志广, 罗一帆. 卤代苯对酵母菌毒性的定量构效关系研究. 数理医药学杂志[J], 2001, 14(2): 168-169.
[84]Patlewicz G Y, Rodford R A, Ellis G, et al. A QSAR Model for the Eye Irritation of CationicSurfactants[J]. Toxicol in Vitro, 2000, 14: 79-84.
[85]Barratt M D. Quantitative structure-activity relationships(QSARs) for skin Corrosivityof Organic acids, bases and phenols: principal components and neural network analysis ofextended datasets[J]. Toxicol in Vitro, 1996, 10: 85-94.
[86]Patel H, Berge W ten, Cronin M T D. Quantitative structure-activity relationships(QSARs)for the prediction of skin permeation of exogenous chemicals[J]. Chemosphere, 2002, 48:603-613.
[87]Whittle E, Battatt M D, Carter J A, et al. Skin Corrosivity Potential of Fatty Acids:In Vitro Rat and Human Skin Testing and QSAR Studies [J]. Toxicol in Vitro, 1996, 10: 95-100.
[88]Chamberlain M, Barratt M D. Practical Applications of QSAR to In Vitro ToxicologyIllustrated by Consideration of Eye Irritation [J]. Toxicol in Vitro, 1995, 9(4): 543-547.
[89]Cronin M T D, Schultz T W. Pitfalls in QSAR [J]. J Mol Struct (Theochem), 2003, 622:39-51.
[90]Walker John D. Applications of QSARs in toxicology: a US Government perspective [J].J Mol Struct (Theochem), 2003,622: 167-184.
[91]Kier L B, Hall L H. Molecular Connectivity in Chemistry and Drug Research [M]. New York,Academic Press: 1976.
[92]Karelson, M. Molecular Descriptors in QSAR/QSPR [M]. New York, John Wiley & Sons: 2000.
[93]Balaban A T. From chemical topology to 3D geometry [J]. J Chem Inf Comput Sci, 1997,37:645-650.
[94]Balaban A T. Topological and stereochemical molecular descriptors for dastabased usefulin QSAR, similarity/dissimilarity and drug design [J]. SAR QSAR Environ Res, 1998, 8: 1-21.
[95]Estrada E, Uriarte E. Recent advances on the role of topological indices in drugdiscovery research [J]. Curr Med Chem, 2001,8: 1699-1714.
[96]Estrada E, Molina E. 3D connectivity indices in QSPR/QSAR studies [J]. J Chem Inf ComputSci, 2001,41: 791-797.
[97]Gasteiger J, Sadowski J, Selzer P, et al. Chemical information in 3D space [J]. J ChemInf Comput Sci, 1996,36: 1030-1037.
[98]Gasteiger J, Burkard S U, Hemmer M C, et al. Decision support systems for chemicalstructure representation, reaction modeling and spectra simulation [J]. SAR & QSAR inEnviron Res, 2002, 13: 89-110.
[99]Engel T, Gasteiger J. Chemical structure representation for information exchange[J].Online Inform Rev, 2002, 26: 139-145.
[100]Taft R M. Polar and steric substituent constants for aliphatic and o-benzoate groupsfrom rates of esterification and hydrolysis of esters [J]. J Am Chem Soc, 1952, 74(12):3120-3128.
[101]Livingstone D J. The characterization of chemical structures using molecularproperties. A survey [J]. J Chem Info Compt Sci, 2000, 40(2): 195-209.
[102]Hansch C, Maloney P P, Fujita T, et al. Correlation of biological activity ofphenoxyacetic acids with Hammett substituent constants and partition coefficients [J].Nature, 1962, 194:178-180.
[103]Hansch C, Muir R M, Fujita T, et al. The correlation of biological activity of plantgrowth regulators and chloromycetin derivatives with Hammett constants and partitioncoefficients [J]. J Am Chem Soc, 1963，85(18)：2817-2824.
[104]Wang R, Fu Y, Lai L. A New Atom-additive Method for Calculating Partition Coefficients[J]. J Chem Inf Comput Sci, 1997, 37: 615-621.
[105]Klopman G, Namboodiri K, Schochet M. Simple Method of Computing the PartitionCoefficient [J]. J Chem Chem 1985, 6: 28-38.
[106]Bodor N, Gabanyi Z, Wong C K. A New Method for the Estimation of Partition Coefficient[J]. J Am Chem Soc, 1989, 111: 3783-3786.
[107]Ghose A K, Croppen G M. Atomic physicochemical parameters for three-dimensionalstructure-directed quantitative structure-activity relationships. 1. Partitioncoefficients as a measure of hydrophobicity [J]. J Comput Chem, 1986,7: 565-577
[108]Hansch C, Leo A J. Substituent Constants for Correlation Analysis in Chemistry andBiology [M]. New York: Wiley and Sons, 1979. 1-339.
[109]Livingstone D J. The characterization of chemical structures using molecularproperties. A survey [J]. J Chem Inf Comput Sci, 2000, 40(2): 195-209.
[110]Selwood D L, Livingstone D J, Comley J C W. Structure-activity relationships ofantifilarial antimycin analogs: a multivariate pattern recognition study [J]. J Med Chem.1990, 33(1): 136-142.
[111]Taft R W, Kamlet M J. The solvatochromic comparison method. 2. The alpha-scale ofsolvent hydrogen-bond donor (HBD) acidities [J]. J Am Chem Soc, 1976,98(10):2886-2894.
[112]Balaban A T. Using real numbers as vertex invariants for third-generation topologicalindexes [J]. J Chem Inf Comput Sci, 1992,32: 23-28.
[113]Wiener H J. Structural determination of paraffin boiling points [J]. J Chem Soc, 1947,69: 17-20.
[114]王连生. 环境化学进展[M]. 北京: 化学工业出版社, 1995. 105-106.
[115]王连生，支正良，高松亭. 分子结构于色谱保留[M]. 北京: 化学工业出版社，1994.216-217.
[116]Randic M. On characterization of molecular branching [J]. J Am Chem Soc, 1975, 97(23):6609-6615.
[117]Kier L B. A shape index from molecular graphs [J]. Quantitative Structure-ActivityRelationships, 1985, 4: 109-116.
[118]Kier L B. Shape indexes of orders one and three from molecular graphs [J]. QuantitativeStructure-Activity Relationships, 1986, 5(1): 1-7.
[119]Kier L B, Hall L H. The nature of structure-activity relationships and their relationto molecular connectivity [J]. Eur J Med Chem, 1977, 12: 307-312.
[120]Hu C Y, Xu L. J Chem Inf Comput Sci, 1996, 36: 82.
[121]Randic M. On molecular identification numbers [J]. J Chem Inf Comput Sci, 1984, 24(3):164-175.
[122]Hansen P J, Jurs P C. Chemical applications of graph theory. Part I. Fundamentals andtopological indices [J]. J Chem Educ, 1988,65(7): 574-580.
[123]Schultz H P. Topological organic chemistry. 1. Graph theory and topological indicesof alkanes [J]. J Chem Inf Comput Sci, 1989,29(3): 227-228.
[124]Muller W R, Szymanski K, Knop J V, et al. An algorithm for construction of the moleculardistance matrix [J]. J Comput Chem, 1987,8(2): 170-173.
[125]Schultz H P, Chultz E B, Schultz T P. Topological organic chemistry. 2. Graph theory,matrix determinants and eigenvalues, and topological indexes of alkanes [J]. J Chem InfComput Sci, 1990,30(1): 27-29.
[126]Schultz H P. Topological organic chemistry. 3. Graph theory, binary and decimaladjacency matrices, and topological indices of alkanes [J]. J Chem Inf Comput Sci,1991,31(1): 144-147.
[127]Heitler W, London F. Interaction Between Neutral Atoms and Homopolar H2[J]. Zeitschriftfür Physik, 1927, 44: 455.
[128]封继康. 基础量子化学原理[M]. 北京: 高等教育出版社, 1984. 346-351
[129]唐敖庆，杨忠志，李前树. 量子化学[M]. 北京：科学出版社. 1982.
[130]Hartree D. Calculations of atomic structure. Wiley. 1957.
[131]Born M, Oppenheimer J R, Quantum theory of molecules. Ann Physik, 1927, 84: 457
[132]Born M, Huang K, Dynamical theory of crystal lattices. Oxford University Press, NewYork, 1954.
[133]Lowdin P O.，Correlation problem in many-electron quantum mechanics [J]. I.Review ofdifferent approaches and discussion of some current ideas [J]. Adv Chem Phys, 1959, 2:207-332.
[134]Bauschlicher C W Jr, Langhoff S R, Accurate quantum chemical calculation [J]. Adv ChemPhys, 1990, 77: 103-161.
[135]Hohenberg P, Kohn W, Inhomogeneous electron gas [J]. Phys Rev B, 1964, 136: 864-871.
[136]Kohn W, Sham L J. Self-consistent equations including exchange and correlation effects[J]. Phys Rev, 1965,140:1133.
[137]Becke A D, Density-functional exchange-energy approximation with correct asymptoticbehavior [J]. Phys Rev A, 1988, 38: 3098-3100.
[138]Perdew J P, Wang Y., Phys. Rev. B, 1986, 33: 8800.
[139]Perdew J P, Chevary J A, Vosko S H, et al, Phys. Rev. B, 1992, 46:6671.
[140]Lee C, Yang W, Parr R G, Development of the Colle-Salvetti correlation-energy formulainto a functional of the electron density [J]. Phys Rev B, 1988, 37: 785-789.
[141]Stevens P J, Devlin J F, Chabalowski C F, et al. J. Phys. Chem., 1994, 98: 11623.
[142]Becke A D, J. Chem. Phys., 1995, 104: 1040.
[143]Bader R F W, Beddall P M. J. Chem. Phys., 1972,56:3320.
[144]Karelson M, Loganov VS. Quantum-chemical descriptors in QSAR/QSPR studies [J]. ChemRev, 1996,96:1027-1043.
[145]Franke R. Theoretial drug design methods [M]. Elsevier: Amsterdam, 1984.11.
[146]Fukui K. Theory of orientation and stereoselection [M]. New York: Springer Verlag,1975.34.
[147]刘次全. 量子生物学[M].北京：高等教育出版社，1990.12-79.
[148]朱永，韩世纲，朱平仇.量子有机化学[M].上海:上海科学出版社,1986.1177~183.
[149]Green M Stuart, Maeshall R Garland. 3D-QSAR: a current perspertive [J]. Tips, 1995,16:28-291.
[150]Vijay M Gokhale, Vithal M Kulkami. Comparative molecular field analysis of fungalsqualene epoxidase inhibitors [J]. J Med Chem, 1999,42:5348-5358.
[151]Elizabeth R Collantes, Li Xing. Comparative molecular field analysis as a tool toevaluate mode of action of chemical hybridization agents [J]. J Agric Food Chem, 1999,47:5245-5251.
[152]Kamlet M J, Abraham M H, Doherty R M, et al. Solubility properties in polymers andbiological media. 4. Correlations of octanol/water patition coefficients withsolvatochromic parameters [J]. J Am Chem Soc, 1984, 106(2): 464-466.
[153]Kamlet M J, Doherty R M, Abboud J-L M, et al. Solubility: a new look [J]. Chemtech,1986,16(9): 566-576.
[154]Kamlet M J, Doherty R M, Beith G D et al. Solubility properties in polymers andbiological media. 7. An analysis of toxicant properties that influence inhibition ofbioluminescence in Photobacterium phosphoreum (the Microtox test)[J]. Environ Sci Tech,1986,20(7):690-695.
[155]Kamlet M J, Doherty R M, Taft R W, et al. Solubility properties in polymers andbiological media. 8. An analysis of the factors that influence toxicities of organicnonelectrolytes to the Golden Orfe Fish (Leuciscus idus melanotus)[J]. Environ Sci Tech,1987,21(2): 149-155.
[156]Kamlet M J, Doherty R M, Carr P W, Linear solvation energy relationships, 44. Parameterestimation rules that allow accurate prediction of octanol/water partition coeffients andother solubility and toxicity properties of polychlorinated biphenyls and polycyclicaromatic hudrocarbons [J]. Environ Sci Tech, 1988,22(5): 503-509.
[157]Kier L B, Hall L H. Molecular Connectivity in Chemistry and Drug Research [M]. NewYork: Academic Press, 1996.
[158]Kier L B, Hall L H. Molecular Connectivity in Structure-Activity Analysis [M]. NewYork: John Wiley & Sons, 1986.
[159]Free S M, Wilson J M. A mathematical contribution to structure-activbity studies [J].J Med Chem, 1964,7(4):395-399.
[160]Draper N R, Smith H. Applied Regression Analysis[J]. New Youk: John Wiley & Sons, Inc.1981.
[161]张尧庭，方开泰.多元统计分析引论[M].北京：科学出版社，1983.
[162]Eriksson L, Johansson E, Kettaneh-Wold N, et al. Multi- and Megavariate Data Analysis:Principles and Applications [M]. Ume?: Umetrics, 2001.
[163]Livingstone D J, Salt D W. Judging the significance of multiple linear regressionmodels [J]. J Med Chem, 2005,48(3): 661-663.
[164]Emanuele Argese, Cinzia Bettiol, Matteo Fasolo, et al. Substituted aniline interactionwith submitochondrial particles and quantitative structure-activity relationshios [J].Biochimica et Biophysica Acta, 2002, 1558: 151-160.
[165]Cronin M T D, Netzeva T I, Dearden J C, et al. Assessment and modeling of the toxicityof organic chemicals to Chlorella vulgaris: development of a novel database [J]. Chem ResToxicol, 2004,17(4)545-554.
[166]Thomas Leonard J., Roy Kunal. QSAR by LFER model of HIV protease inhibitor mannitolderivatives using FA-MLR, PCRA, and PLS techniques [J]. Bioorgan Med Chem, 2006,14:1039-1046.
[167]Shen Qi, Jiang Jian Hui, Jiao Chen Xu, et al. Modified particle swarm optimizationalgorithm for variable selection in MLR and PLS modeling: QSAR studies of antagonism ofangiotensin Ⅱ antagonists [J]. Eur J Pharm Sci, 2004,22: 145-152.
[168]Lü Jian Xia, Shen Qi, Jiang Jian Hui, et al. QSAR analysis of cyclooxygenase inhibitorusing particle searm optimization and multiple linear regression[J]. J Pharmaceut Biomed,2004,35:679-687.
[169]Hemmateenejad Bahram，Miri Ramin, Akhond Morteza, et al. QSAR study of the calciumchannel antagonist activity of some recently synthesized dihydropyridine derivatives. Anapplication of genetic algorithm for variable selection in MLR and PLS methods [J].Chemometr Intell Lab, 2002,64: 91-99.
[170]Franke R., Gruska A. In Chemometric Methods in Molecular Design [M]. VCH: Weinheim,1995. 113.
[171]Franke R. Theoretical Drug Design Methods [M]. Elsevier: Amsterdam, 1984. 184.
[172]Kirschner G, Kowalski B R. The application of pattern recognition on drug design, inDrug Design [M]. New York: Academic Press, 1979. Vol.Ⅲ.
[173]Brown R D, Martin Y C. Use of structure-activity data to compare structure-basedclustering methods and descriptors for use in compound sdelection [J]. J Chem Inf ComputSci, 1996,36(3):572-584.
[174]Senese C L, Hopfinger A J. A simple clustering technique to improve QSAR model selectionand predictivity: application to a receptor independent 4D-QSAR analysis of Cyclic Ureaderived inhibitors of HIV-1 protease [J]. J Chem Inf Comput Sci, 2003,43(6):2180-2193.
[175]Klein C T, Kaiser D, Ecker G. Topological distance based 3D descriptors for use inQSAR and diversity analysis [J]. J Chem Inf Comput Sci, 2004,44(1):200-209.
[176]Barnard J M, Downs G M. Clustering of chemical structures on the basis oftwo-dimensional similarity measures [J]. J Chem Inf Comput Sci, 1992,32(6):644-649.
[177]任若恩，王惠文.多元统计数据分析-理论、方法、实例.北京：国防工业科学出版社，1999.
[178]刘增良, 刘有才. 模糊逻辑与神经网络技术[M]. 北京航天航空大学出版社, 1996.121.
[179]焦李成.神经网络系统理论[M].西安：西安电子科技大学出版社，1992.
[180]焦李成.神经网络计算[M].西安：西安电子科技大学出版社，1993.
[181]焦李成.神经网络的应用与实现[M].西安：西安电子科技大学出版社，1993.
[182]蒋宗礼.人工神经网络导论[M].北京：高等教育出版社，2001.
[183]Broonhead, Lowe D. Multivariable function interpretation and adaptive network [J].Complex Systems, 1988,2.
[184]Moody J E, Darken C.J. Fast algorithms in networks of locally tuned processing units[J]. Neural Computation, 1989,1:282-289.
[185]Jackon I R H. Convergence properties of radial basis function [J]. Approx,1988,4:243-264.
[186]Tyler Holcomb, Manfred Morari. Local training for radial basis function metworkstoward solving the hidden unit problem[C]. American Control Conference,1993,3(1).
[187]Mark J L Orr. Introduction to radial basis function networks [G].1996.
[188]Mark Orr, Hallam J. Assessing RBF networks using DELVE [J]. Intermational Journal ofNeural System, 2000, 10(5): 397-415.
[189]吴海龙，梁逸曾，俞汝勤.分析化学计量学[M].分析实验室，1999,18(6):94-102.
[190]Harrington Peter de B..Temperature-Constrained Cascade Correlation Networks [J]. Anal.Chem., 1998,70:1297-1306.
[191]Cai Chunsheng, Harrington Peter de B. Wavelet Transform Preprocessing for TemperatureConstrained Cascade Correlation Neural Networks[J].J Chem InfComput Sci,1999,39:874-880.
[192]Harrington Peter de B, Urbas Aaron, Wan Chuanhao. Evaluation of Neural Network Modelswith Generalized Sensitivity Analysis [J]. Anal Chem, 2000,72:50004-5013.
[193]Wan Chuanhao, Harrington Peter de B. Screening GC-MS data for carbamate pesticideswith temperature-constrained-cascade correlation neural networks [J].Analytical ChinicaActa ,2000, 408:1-12.
[194]刘思东，崔秀君，张卓勇等.温度限制串联相关网络用于有机环境污染物紫外光谱的识别[J].光谱学与光谱分析， 2003（1）：119-122.
[195]崔秀君，张卓勇，Harrington Peter de B，等. 温度限制串联相关网络-近红外光谱法用于药物甲硝唑的质量控制[J].高等学校化学学报，2004，25（7）：1251-1253.
[196]Cui Xiujun, Zhang Zhuoyong, Ren Yulin, et al. Quality control of the powderpharmaceutical samples of sulfaguanidine by using NIR reflectance spectrometry andtemperature-constrained cascade correlation networks [J]. Talanta, 2004,64:943-948.
[197]张卓勇.温度限制串联相关神经网络及其在细菌辨识中的应用，高等学校化学学报[J].2002（4）：570-572.
[198]Wang Fengxia, Zhang Zhuoyong, Cui Xiujun, et al. Identification of rhubarbs by usingNIR spectrometry and temperature-constrained cascade correlation networks [J]. Talanta,2006, (in press).
[199]Vapnik V. The Nature of Statistical Learning Theory [M]. Springer Verlag, 1995.
[200]Vapnik V. Statistical Learning Theory[M].New York: Wiley,1998.
[201]Burges C J C . A tutorial on support vector machines for pattern recognition [J]. DataMin Knowl Discov,1998,2:121-167.
[202]Cristianini N, Shawe-Taylor J. Introduction to support vector machine and other kernelbased learning methods [M]. Cambridge：Cambridge University Press, 2000.
[203]张学工.关于统计学习理论与支持向量机[J].自动化学报，2000, 26 (1): 32-42.
[204]边肇棋.模式识别[M]. 北京:清华大学出版社，2000.
[205]Vapnik V.Statistical learning theory [M].New York:John Wiley and Sons,1998.
[206]Wolfe P. A duality theory for nonlinear programming. Quarterly of Mathematics [J].1961, 19:239-244.
[207]Chinnasamy Arunkumar, Mittal Ankush, Sung WingKin Probabilistic prediction ofprotein-proteininteractionsformtheproteinsequences[J].ComputBiolMed,2006,36:1143-1154.
[208]Lumini Alessandra, Nanni Loris. Machine learning for HIV-1 protease cleavage siteprediction [J].Pattern Recognition Letters,2006,27:1537-1544.
[209]Chen Chao, Zhou Xibin, Tian Yuanxin, et al. Predicting Protein structural class withpseudo-amino acid composition and suppout vector machine fusion network[J]. AnalBiochem,2006,357:116-121.
[210]Dubey Anshul, Realff Matthew J, Lee Jay H, et al. Identifying the interacting positionsof a protein using Boolean learning and support vector machines [J]. Comput. Biol. Chem.,2006,30:268-279.
[211]Brown M, Grundy W, Lin D, et al. Knowledge-based analysis of microarry gene expressiondata using support vector machines [J]. Proc Natl Acad Sci USA.,2000,97:262-267.
[212]Guyon I, Weston J, Barnhill S, et al. Gene selection for cancer classification usingsupport vector machines [J]. mach Learn, 2002,46:389-422.
[213]Cai Y D, Liu X J, Xu X B, et al. Prediction of protein structural classes by supportvector machines [J]. Comput Chem 2002,26:293-296.
[214]Cai Y D, Liu X J, Xu X B, et al. Support vector machines for the classification andprediction of beta-turn types [J]. J Pept Sci, 2002,8:297-301.
[215]Hua S, Sun Z. Novel method of protein secondary structure prediction with high segmentoverlap measure: support vector machine approach [J]. J Mol Bio, 2001,308:397-407.
[216]Ding C H Q, Dubchak I. Multi-class protein fokd recognition using support vectormachines and neural networks [J]. bioinfornatics, 2001,17:349-358.
[217]Cai YD, Liu X J, Xu X B, et al. Support vector machines for prediction of proteinsubcellular location by incorporation gquasi-sequence-order effect [J]. J Cell Biochem,2002,84:343-348.
[218]Hua S, Sun Z. Support vector machine approach for protein subcellular localizationprediction [J]. Bioinformatics, 2001,17:721-728.
[219]Bock J R, Gough S A. Predictiong protein-protein interactions from primary structure[J]. Bioinformatics, 2001,17:455-460.
[220]Darchin, R, Darplus K, Huassler D. Classifying G-protein coupled receptors withsupport vector machines [J]. Nioinformatics, 2002,18:147-159.
[221]Lee D, choi . W, Kim M, et al. Discovery of differentially expressed genes relatedtohistologicalsubtypeofhepatocellularcarcinoma[J].biotechnolProg,2003,19:1011-1015.
[222]Xue Y M, Yap C W, Sun L Z, et al. Prediction of P-glycoprotein substrates by a supportvector machine approach [J]. J Chem Inf Comput Sci, 2004,44:1497-1505.
[223]Zhao C Y., Zhang R S, Liu H X, et al. Diagnosing anorexia based on partial least squares,back propagation neural network and support vector machines [J]. J Chem Inf Comput Sci,2004,44:2040-2046.
[224]Xue C X, Zhang R S, Liu H X, et al. An accurate QSPRstudy of O-H bond dissociationenergy in substituted phenol base on support vector machines [J]. J Chem Inf Comput Sci,2004,44:669-677.
[225]Xue C X, Zhang R S, Liu M C, et al. Study of the quantitative structure-mobilityrelationship of carboxylic acids in capillary electrophoresis baded on support vectormachines [J]. J Chem Inf Comput Sci,2004,44:950-957.
[226]Xue C X, Zhang R S, Liu H X, et al. Support vector machines-based quantitativestructure-property relationship for the prediction of heat capacity [J]. J Chem Inf ComputSci, 2004,44:1267-1274.
[227]Xue C X, Zhang R S, Yao X J, et al. prediction of the isoelectric point of an aminoacid based on GA-PLS and SVMs [J]. J Chem Inf Comput. Sci, 2004,44:161-167.
[228]Luan F, Zhang R, Yao X, et al. support vector machine-based QSPR for the predictionof Van de Waal’constants [J]. QSAR & comb Sci, 2005,24:227-239.
[229]Sadik O, Land Jr, Wanekaya A K, et al. Detection and classification of organophosphatenerve agent simulants using support vector machines with multiarray sensors [J]. J chem.Inf Comput Sci, 2004,44:499-507.
[230]Liu Y. A comparative study on feature selection methods for drug discovery [J]. J ChemInf Comput Sci, 2004,44:1823-1828
[231]Xue Y, Li Z R, Yap C W, et al. Effect of molecular descriptor feature selection insupport vector machine classification of pharmacokinetic and toxicological properties ofchemical agents [J]. J Chem Inf Comput Sci, 2004,44:1630-1638.
[232]Sorich M J, Miners J O, Mckinnon R A, et al. Comparison of linear and nonlinearclassification algorithms for the prediction of drug and chemical metabolism by humanUDP-Glucuronosylltransferase isoforms [J]. J Chem Inf Comput Sci, 2003,43:2019-2024.
[233]Yao X J, Panaye A, Doucet J P, et al. Comparative study of QSAR/QSPR correlations usingsupport vector machines, radial basis function neural networks, and multiple linearregression[J]. J Chem Inf Comput Sci, 2004,44:1257-1266.
[234]Liu H X, Zhang R S, Yao X J, et al. QSAR study of ethyl2-[(3-methyl-2,5-dioxo(3-pyrrolinyl))amino]-4-(trifluoromethyl)pyrimidine-5-carboxylate:an inhibitor of AP-1 and NF-1 and NF-kB mediated gene expression based on support vectormachines[J]. J Chem Inf Comput Sci, 2003,43:1288-1296.
[235]刘亮，包新华，冯建星，等,α-唑基-α-芳氧烷基频哪酮（芳乙酮）及其醇式衍生物抗真菌活性的分子筛选[J].计算机与应用化学，2002,19(4):476-479.
[236]Shen Qi, Shi WeiMin, Kong Wei, et al. A conbination of modified particle swarmoptimization algorithm and support vector machine for gene selection and tumorclassification [J]. Talanta, 2006,(in press)
[237]Furey T, Cristianini N, Cuffy N, et al. Support vector machine classification andvalidation of cancer tissue samples using microarray expression data [J]. Bioinformatics,2000,16:906-914.
[238]Ramaswamy S, Tamays P, Rifkin R, et al. Multiclass cancer diagnosis using tumor geneexpression signatures [J]. Proc Natl Aatl Acad Sci USA, 2001,98:15149-15154.
[239]Luan F, Zhang R, Zhao C, et al. Classification of the carcinogenicity of N-nitrosocompounds based on support vector machines and linear discriminant analysis [J]. Chem ResToxicol, 2005,18:198-203
[240]Takaoka Y, Endo Y, Yamanobe S, et al. Development of a Method for EvaluatingDrug-Likeness and Ease of Synthesis Using a Data Set in Which Compounds Are Assigned ScoresBased on Chemists’ Intuition [J]. J Chem Inf Comput Sci, 2003,43:1269-1275.
[241]Ivancius O. Support vector machine identification of the aquatic toxicity mechanismof organic compounds[J]. Internet Electronic Journal of Molecular Design, 2002,1:151-172.
[242]Li H, Ung C Y, Yap C W, et al. Prediction of Genotoxicity of Chemical Compounds byStatistical Learning Methods [J]. Chem Res Toxicol,2005,18:1071-1080.
[243]Liu H X, Zhang R S, Yao X J, et al. QSAR and classification models of a novel seriesof COX-2 selective inhibitors: 1,5-Diarylimidazoles based on support vector machines[J].J Comput Aid Mol Des,2004,18:389-399.
[244]Yang S S, Lu W C, Chen N Y, et al. Support vector regression based QSPR for the predictionof some physicochemical properties of alkyl benzenes [J]. J Mol Struct(Theochem),2005,719:119-127.
[245]Tugcu N, Song M, Breneman C M, et al. Prediction of the effect of mobile-phase salttype on protein retention and selectivity in anion exchange systems [J]. Anal Chem,2003,75:3563-3572.
[246]Yao X J, Panaye A, Doucet J P, et al. Comparative classification study of toxicitymechanisms using support vector machines and radial basis function neural networks [J].Anal Chim Acta, 2005,535,259-273.
[247]Hartter D R. The use and importance of nitroaromatic chemicals in the chemical industry[A]. In: Rickert D (Ed). Toxicity of nitroaromatic compounds[C]. Washington, DC: Hemisphere,1985,1-3.
[248]Rosenblatt D H, Burrows E P, Mitchell W R, et al. Organic explosives and relatedcompounds [A]. In: Hutzinger O(Ed). The Handbook of Environmental Chemistry—AnthropogenicCompounds [C].Berlin:spinger Verlag,1991,195-237.
[249]Adkins R L. Nitrobenze and nitrotoluene [A]. In Grant MH (Hd). Kirk Othmer encyclopediaof chemical technology 4ed[C]. New York: Wiley Interscience,1994,133-152.
[250]Yamagisshi T, Miyazaki T, Horii S, et al. Identificatin of musk Xylene and musk ketonein fresh water fish collected from the Tama River Tokyo [J]. Bull Environ Contem Toxic,1981,26(10):656-662.
[251]Yurawecz M P, Puma B J. Nitro musks fragrances as potential contaminants in pesticideresidue analysis [J]. J. Ass. Off. Analyst. Chem., 1983,66(4): 241-247.
[252]Gatermann R, Huhnerfuss H, Rimkus G, et al. The distribution of nitrobenzene and othernitroaromatic compounds in the North Sea [J]. Marine Pollution Bulletin,1994,30(3):221-227.
[253]郎佩珍，龙凤山，袁星，等. 松花江中游（哨口-松花江村段）水中有毒有机物污染研究[M]. 环境科学进展，1993，1（6）：47-55.
[254]于常荣，曹喆，王炜，等. 松花江鱼类有机污染物的研究[J]. 中国环境科学，1994，14（4）：283-287.
[255]康跃惠，宫正宇，王子建，等. 宫厅水库及永定河水库中挥发性有机物分布规律研究[J]. 环境科学学报，2001，21（3）：338-343.
[256]王子建，吕怡兵，王毅，等. 淮河水取代苯污染物及其生态风险[J]. 环境科学学报，2002，22（3）：300-304.
[257]王宏，杨霓云，沈英娃，等. 海河流域几种典型的有机污染物环境安全性评价[J]. 环境科学研究，2003，16（6）：35-36.
[258]胡国华，赵沛伦. 黄河猛津-花园口区间有机污染分析及防治对策[J]. 人民黄河，1995，11：6-9。
[259]田怀军，舒为群，张学奎，等. 长江、嘉陵江（重庆段）源水有机污染物的研究[J]. 长江流域资源与环境，2003，12（2）：118-123.
[260]Huang Hong, Wang Xiaodong, Ou Wenhua,et al., Acute toxicity of benzene derivativesto the tadpoles (Rana japonica) and QSAR analyses[J]. Chemosphere, 2003,53: 963-970.
[261]Ren Shijin. Ecotoxicity prediction using mechanism – and non-mechanism-based QSARs:a preliminary study [J]. Chemosphere, 2003,53: 1053-1069.
[262]Trohalaki S, Gifford E, Pachter R. Improved QSARs for predictive toxicology ofhalogenated hydrocarbons [J]. Comput Chem, 2000, 24:421-427.
[263]Renata Dias de Mello Castanho Amboni, Berenice da Silva Junkes, Vilma Edite FonsecaHeinzen, et al., Semi-empirical topological method for prediction of the chromatographicretention of esters[J]. J Mol Struc-Theochem. 2002,579:53-62.
[264]齐玉华，许禄，王淑云.量化参数在黄酮类化合物构效关系研究中的应用[J]. 计算机与应用化学，2000，17（1）：29-31.
[265]Carbó-Dorca R, Amat L, Besalú E, et al. Quantum mechanical origin of QSAR: theory andapplications [J]. J. Mol. Strut.（Theochem）, 2000,504:181-228.
[266]籍国东，赵元慧，袁星. 量化参数及其在定量结构-活性-性质相关研究中的应用[J]. 东北师大学报自然科学版，1998，4:47-53.
[267]Worrall Fred, Thomsen Marianne. Quantum vs. topological descriptors in the developmentof molecular models of groundwater pollution by pesticides [J]. Chemophere,2004,54:585-596.
[268]周文富，戈芳.Ph(x)n位电荷密度能与生物毒性定构活性相关研究[J]. 广西大学学报（自然科学版），1997,9:218-221.
[269]齐玉华，许禄.应用量化参数和CoMFA法研究苯甲酸类化合物的结构和其pKa值的相关性[J]. 应用化学，2002,19(11):1054-1058.
[270]裴洪平，许高金.量子化学参数用于苯胺类化合物的 QSAR 毒性研究[J]. 浙江大学学报（理学版）2003,30(3):310-313。
[271]Brian W Clare, claudiu T Supuran, Semi-empirical atomic charges and dipole momentsin hypervalent sulfonamide molecules: descriptors in QSAR studies [J]. J Mol Struc-Theochem,1998, 428:109-121.
[272]Li Laicai, Maoshuang, Zhao Keqing, et al. Semi-empirical quantum chemical study onstudy on structure-activity relationship in monocyclic-β-lactam antibiotics [J]. J MolStruc- Theochem, 2001,545: 1-5.
[273]Emanuele Argese, Cinzia Bettiol, Matteo Fasolo, et al. Substituted aniline interactionwith submitochondrial particles and quantitative structure-activity relationshios [J].Biochimica et Biophysica Acta, 2002, 1558: 151-160.
[274]李卫华, 许旋. 氯苯胺类化合物对发光细菌毒性的量子化学研究[J]. 华南示范大学学报(自然科学版), 2001, 1: 76-78.
[275]Stefan Sixt, Joachim Altschuh, Rainer Brüggemann. Quantitative structure-toxicityrelationships for 80 chlorinated compounds using quantum chemical descriptors [J].Chemosphere, 1995, 30(12): 2397-2414.
[276]Li Zhang, Jian Wan, Guanfu Yang. A DFT-based QSARs study of protoporphyrinogen oxidaseinhibitors [J]. phenyl triazolinones. 2004, 12:6183-6191.
[277]Stefan P. Niculescu. Artificial neural networks and genetic algorithms in QSAR [J].J Mol Struc- Theochem, 2003, 622: 71-83.
[278]Tomas-Vert F Perez-Gimenez, Ma T Salabert-Salvador, Garcia-March F J, et al.Artificial neural network applied to the discrimination of antibacterial activity bytopological methods [J]. J Mol Struc- Theochem, 2000,504:249-259.
[279]Yan Aixia, Jiao Guimei, Hu Zhide, et al. Use of artificial neural networks to predictthe gas chromatographic retention index data of alkylbenzenes on carbowax-20M [J]. ComputChem, 2000, 24:171-179.
[280]Gonzalez-Arjona D, Lopez-Perez G, Gustavo Gonzalez A. Non-linear QSAR modeling byusing multiplayer perceptron feedforward neural networks trained by back-propagation [J].Talanta, 2002,56:79-90.
[281]Agatonovic-Kustrin S, Beresford R, Pauzi A Yusof M. ANN modeling of the penetrationacross a polydimethylsiloxane membrane from theoretically derived molecular descriptors[J]. J Pharmaceut Biomed, 2001,26: 241-254.
[282]章文军，许禄.定量结构-活性/性质相关性中变量的选择----正交变换法与最优子集回归法的比较[J]. 高等学校化学学报，2001，22:1134-1136.
[283]Becke, A D. Density-functional thermochemistry. III. The role of exact exchange [J].J Chem Phys, 1993, 98(7): 5648–5652.
[284]Frisch M J， Trucks G W， Schlegel H B， et al． Gaussian 98, Revision A.9， Gaussian,Inc, Pittsburgh PA, 1998.
[285]许禄，邵学广. 化学计量学方法（第二版）[M]. 北京:科学出版社， 2004,291.
[286]金相灿.有机化合物污染化学[M].北京：清华大学出版社，1990.
[287]王红娟，奚红霞，夏启斌等.含酚废水处理技术的现状与开发前景[J].工业水处理，2002,22(6):6-9.
[288]巨华东.水环境污染概论[M].北京:北京师范大学出版社，1989:112-157.
[289]尹伊伟. 用氯苯及酚类化合物研究大鳞副泥鳅急性毒性的定量结构一活性关系[J].暨南大学学报，1998，19：112-115.
[290]Nimrod A C, Bcnson W H. Environmental estrogenic effects of Ikylphenol ethoxylates[J]. Crit Rew Toxicol, 1996,2 (3): 335-364.
[291]Jobling S Sheahao D, Osbome J A, et al. Inhabition of testicular growth in minbow troutexposed to estrogenic alkylphenolic chemicals [J].Environ Toxicol chem,1996, 15(2):194-202.
[292]Harries J E, Sheahan D A, Jobling S. Estrogenic activity in five United Kingdom riversdetected by measurement of vitellogenexis in caged trout [J].Environ Health Toxicol.Chem.1997, 16(3):534-542.
[293]Kaiser K L, Dixon D G, Hodson P V. SAR in Environmental Toxicology[M].Boston: D. ReidelPublishing Company, 1984,189-206.
[294]许旋，罗一帆，徐志广，等.氯苯酚电子结构对水生物毒性的研究[J]. 卫生毒理学杂志，2001，15（3）：167-170.
[295]刘够生，宋兴福，于建国，等.氯代苯酚类衍生物对水生物发光细菌的定量结构-活性关系研究[J].江西师范大学学报（自然科学版），2001，25（4）：313-316.
[296]许禄，吴亚平.苯酚类化合物的三维定量构效关系研究[J].计算机与应用化学，2000,17(1)：15-17.
[297]张大仁.酚取代衍生物的 QSAR 研究[J]. 环境科学，1995，16（2）：4-6.
[298]秦正龙，冯长君.取代苯酚的定量结构-活性/性质相关性研究[J].有机化学，2003，23（7）：654-658.
[299]王桂莲，白乃彬.多氯酚 QSAR 数值模型比较研究[J]. 环境科学学报，1996，16（2）：190-194.
[300]Kishino Takuo, Kobayshi Kunio. Studies on the mechanism of toxicity of chlorophenolsfound in fish through Quantitative Structure-Activity Relationships [J]. Wat Res,1996,30(2):393-399
[301]Cronin Mark T D, Aptula Aynur O, Duffy Judith C, et al. Comparative assessment ofmethods to develop QSARs for the prediction of toxicity of phenols to Tetrahymena pyriformis[J]. Chemosphere, 2002,49:1201-1221.
[302]Okey Rober W, Stensel H.David. A QSAR-based biodegradablity model-A QSBR [J]. Wat Res,1996, 30(9): 2206-2214.
[303]Adams Craig D, Cozzens Randall A, Kim Byung J. Effects of ozonation on thebiodegradability of substituted phenols [J]. Wat. Res., 1997, 31(10): 2655-2663.
[304]Damborsky Jiri, Schultz T.Wayne. Comparison of the QSAR models for toxicity andbiodegradability of anilines and phenols [J]. Chemosphere, 1997,34(2): 429-446.
[305]王连生，支正良. 分子连接性与分子结构-活性[M]. 北京：中国环境科学出社出版，1992.
[306]Bodor N Harget A，Huang M. Neural network studies. 1.Estimation of the aqueoussolubility of organic compounds [J]. J Am Chem Soc, 1991,113:9480-9483.
[307]Mitchell B E, Jurs PC. Prediction of aqueous solubility of organic compounds frommolecular structure [J]. J Chem Inf Comput Sci, 1988, 38: 489-496.
[308]Huuskonen J. Estimation of aqueous solubility for a diverse set of organic compoundsbased on molecular topology [J]. J Chem Inf Comput Sci, 2000, 400:773-777.
[309]Liu R, So S-S. Development of quantitative structure-property relationship models forearly ADME evaluation in drug discovery. 1.Aqueous solubility [J]. J Chem Inf Comput Sci,2001, 41: 1633 一 1639.
[310]Yalkowsky S H, Pinal R. Estimation of the aqueous solubility of complex organicmolecules [J]. Chemosphere, 1993, 260: 1239-1261.
[311]Huibers PT, Katritzky AR. Correlation of the aqueous solubility of hydrocarbons andhalogenated hydrocarbons with molecular structure [J]. J Chem Inf Comput Sci, 1998, 38:283-292.
[312]Tolls J, van Dijk J, Verbruggen EJM, et al. Aqueous solubility-molecular sizerelationships: A mechanistic case study using C10- to C19-alkanes [J]. J Phys Chem: A, 2002,1060: 2760-2765.
[313]Klopman G, Zhu H. Estimation of the aqueous solubility of organic molecules by thegroup contribution approach [J]. J Chem Inf Comput Sci, 2001, 410: 439-445.
[314]Jain N, Yalkowsky SH. Estimation of the aqueous solubility I: Application to organicnonelectrolytes [J]. J Pharm Sci, 2001, 90: 234-252.
[315]Bodor N, Huang M-J. A new method for the estimation of the aqueous solubility of organiccompounds [J]. J Pharm Sci, 1992, 81:954-960.
[316]Nelson TM, Jurs PC. Prediction of aqueous solubility of organic compounds [J]. J ChemInf Comput Sci, 1994, 34: 601-609.
[317]Nirmalakhandan NN, Speece R.E. Prediction of aqueous solubility of organic chemicalsbased on molecular structure [J]. Environ Sci Technol, 1988, 22 :328-338.
[318]Yalkowsky S.H, Pinal R. Estimation of the aqueous solubility of complex organicmolecules [J]. Chemosphere, 1993, 26: 1239-1261.
[319]王福安，蒋登高，傅举孚. 化工数据导引[M]. 北京：化学工业出版社，1995，156-184.
[320]Wienke G, Gmehling J, Prediction of Octanol-Water Partition Coefficients, HenryCoefficients and Water Solubilitis Using UNIFAC[J]. Toxicol Environ Chem., 1998, 65:57-65.
[321]郭明，刘文杰.有机化合物水溶解性的 QSPR 研究[J]. 计算机与应用化学，2000，17（1）：57-58.
[322]陈艳，冯长君.连接性指数对脂肪醇的 QSAR 研究[J]. 环境化学，2000，19（6）：538-543.
[323]任碧野，许有.一个新的拓扑指数用于有机化合物 QSAR/QSPR 研究[J].化学学报，1999，57（6）：563-571.
[324]张旭，余训民. 一个新的连接性指数对脂肪醇的 QSPR/QSAR 研究[J]. 化学研究，2002，13（1）：49-57.
[325]HAN J，KAMBER M．Data Mining：Concepts and Techniques[M]．Morgan Kaufmaml Publishers，2001．
[326]徐泽柱，王林. 基于粗糙集理论和 BP 神经网络的数据挖掘算法[J]．计算机工程与应用，2004，40(31)：169-172，175．
[327]刘钊，蒋良孝．基于神经网络的数据挖掘研究[J]．计算机工程与应用，2004，40(3)：172-173，190．
[328]Orlinskii D. Influence of environmental contamination with PCBs on human health [J].Environ Geochemistry and Health, 2001，23:317-332.
[329]Sayles G D. DDT, DDD, and DDF Dechlorination by Zero-Valent Iron [J]. EnvironSciTechnol, 1997, 31(12):3448-3454.
[330]Squillace P J. Volatile Organic Copounds in Untreated Ambient Groudwater of the UnitedStates [J]. 1985-1995 Environ Sci Technol 1999,33(3):4176-4187.
[331]李伟民，尹大强，李时银，等. 氯代苯胺对斑马鱼的急性毒性及 3D-QSAR 分析[J]. 环境科学研究，2002，15（2）：6-7.
[332]张玲，张爱茜，韩朔睽，等. 卤代苯及其衍生物的比较分子场研究[J]. 环境化学，2002，21（3）：250-253.
[333]周文富. 卤代苯和苯酚衍生物的位电荷密度能和logKow 与-logLC50的QSAR研究[J]. 有机化学，2002，22（9）：658-662.
[334]Blum D J W, Speece R E. Quantitative Structure-activity Relationships for ChemicalsToxicity to Environmental Bacteria [J]. Ecotoxi Environ Saf, 1991，22:198-224.
[335]Blum D J W, Speece R E. A Database of Chemical Toxicity to Environmental Bacterialand Its Use in Interspecies Comparisons and Correlations [J]. Research Journal WPCF, 1991，63 (3):198-207.
[336]舒元梯.卤代芳烃生物活性的分子拓扑研究[J]. 西南民族大学学报（自然科学版），2006，32（3）：502-207.
[337]袁星, 郎佩珍, 洪晖，等. 应用发光菌测定有机化合物的毒性, 见：王连生等编. 有机污染化学进展[M]. 北京: 化学工业出版社, 1998, 90-92.
[338]Sixt Stefan, Altschuh Joachim, Brüggemann Rainer. Quantitative structure-toxicityrelationships for 80 chlorinated compounds using quantum chemical descriptors[J].Chemosphere, 1995,30(12): 2397-2414.
[339]Becke A D. Density-functional thermochemistry. III. The role of exact exchange [J].J Chem Phys, 1993, 98: 5648-5652.
[340]Stephens P J, Devlin F J, Chabalowski C K, et al. Ab initio Calculation of VibrationalAbsorption and circular Dichroism Spectra Using Density Functional Force Fields [J]. J PhysChem, 1994, 98: 11623-16539.
[341]Zhhao Q, Bao Z. Target recognition based on radial basis functio network. In:Proceeding of international joint conference on neural network V3 [M]. Nagoya:1993,2735-2738.
[342]Hartman E J, Keeler J D, Kowalski J M. Layered neural networks with Guassian hiddenunits as universal approximations [J]. Neural Comput.,1990,2:210-215.
[343]Park J, Sandberg J W. Universal approximation using radial basis functions network[J].Neural Comput, 1991,3:246-257.
[344]Poggio T, Girosi F. Networks for approximation and learning [J]. Proc of the IEEE,1990,78(9):1481-1497.
[345]Schwarzenbach R P, Gschwend P M, Imboden D M. 王连生等译. 环境有机化学[M]. 北京:化学工业出版社, 环境科学与工程出版中心, 2004.
[346]林红卫，杨胜喜，李志良. 卤代苯和苯酚的衍生物的结构表征和毒性预测[J]. 湘潭大学自然科学学报，20059，27（3）：71-76。
[347]舒元梯. 分子拓扑指数与对卤代苯酚生物活性的预测[J]. 西南民族大学学报（自然科学版），2006，32：78-82.
[348]Trohalaki S, Gifford E, Pachter R. Improved QSARs for predictive toxicology ofhalogenated hydrocarbons [J]. Comput Chem, 2000,24:421-427.
[349]Gute Brian D, Balasubramanian K, Geiss K T, et al. Prediction of halocarbon toxicityfrom structure: a hierarchical QSAR approach [J]. Environ Toxicol Phar, 2004,16:121-129.
[2]Blum D J W, Speece R E. Edtermining chemical toxicity of aquatic species[J]. EnvironSci Technol, 1990, 24: 284-293.
[3]王连生. 有机污染物化学[M]:上册. 北京：科学出版社，1990.1.
[4]梁逸曾, 俞汝勤. 分析化学手册[M]: 第十分册, 化学计量学. 北京: 化学工业出版社, 2000.392.
[5]Hansch C, Maloney P P, Fujita T, et al. Correlation of biological activity ofphenoxyacetic acids with Hammett substituent constants and partition coefficients [J].Nature, 1962,194: 178-180.
[6]Hancch C, Muir R M, Fujita T, et al. The correlation of biological activity of plantgrowth regulators and chloromycetin derivatives with Hammett constants and partitioncoefficient [J]. J Am Chem Soc, 1963, 85(18): 2817-2824.
[7]Fujita T, Iwasa J, Hansch C A new substituent constant, π,derived from partitioncoefficients[J]. J Am Chem Soc, 1964, 86(23): 5175-5180.
[8]Free S M, Wilson J M. A mathematical contribution to structure-activity studies [J].J Med Chem, 1964, 7(4): 395-399.
[9]王连生. 分子结构、性质与活性[M]. 北京：化学工业出版社，1997.
[10]唐桂刚，白乃彬. 遗传神经网络在 QSAR 中的应用研究[J]. 计算机与应用化学，1999, 16(6):435-440.
[11]周家驹，陈红明，谢桂荣，等. 三维定量结构-活性关系研究中的受体作用位点模型方法[J]. 化学进展, 1998,10(1): 55-62.
[12]李江波, 谢华庆, 张敬畅, 等. dT4 核苷酸类似物抗 HIV-1 活性的定量构效关系[J]. 北京化工大学学报, 2001, 28(1): 79-80.
[13]Suh Myung-Eun, Kang Min-Jung, Park So-Young. The 3-D QSAR Study of Anticancer1-N-substituted Imidazo- and Pyrrolo-quinoline-4,9-dione Derivatives by CoMFA andCoMSIA[J]. Bioorg Med Chem, 2001, 9: 2987-2991.
[14]李华, 许禄, 苏锵, 等. 甾体化合物的结构与抗炎活性相关性 3D-QSAR 研究[J]. 计算机与应用化学, 1997, 14(1): 27-30.
[15]Cong V Nguyen-, Dang G Van, BM Rode. Using Multivariate Adaptive Regression Splinesto QSAR Studies of Dihydroartemisinin Derivatives [J]. Eur J Med Chem, 1996, 31: 797-803.
[16]Kowalski B R, Bender C F. The application of pattern recognition to screeningprospective anticancer drugs [J]. J Am Chem Soc, 1974, 96:916-918.
[17]Lipnick R L, Overton C E. Narcosis studies and a contribution to general pharmacology[J]. Trends Pharmacol Sci, 1986, 7: 161–164.
[18]Cocchi M., Benedetti P G De. Use of the Supermolecule Approach to Derive MolecularSimilarity Descriptors for QSAR Analysis [J]. J Mol Model, 1998, 4: 113 – 131.
[19]蔡煜东, 杨兵, 孙红. 用自组织神经网络研究抗肿瘤药卡巴醌构效关系[J]. 计算机与应用化学, 1995, 12: 255-266.
[20]冯长君. 无机物晶格能及其它理化性质的定量构效关系研究[J]. 分析科学学报, 2002, 18(1):27-32.
[21]赵瑞环, 岳丙方, 单亦初, 等. 神经元网络用于 PCDD 定量构效关系的研究[J]. 色谱, 1999,17(2): 112-114.
[22]Kamlet M J, Doherty R M, Veith G D, et al. Solubility properties in polymers andbiological media[J]. Environ Sci Technol, 1986, 20(7): 690-695.
[23]Ovidiu Ivanciuc. The Neural Network MolNet Prediction of Alkane Enthalpies [J]. AnalChim Acta, 1999, 384: 271-284.
[24]Aleksandar S. QSAR Models for Estimating Properties of Present Organic PollutantsRequired in Evaluation of Their Environmental Fate and Risk [J]. Chemosphere, 2001, 43:363-375.
[25]Loukas Y L. Radial Basis Function Networks in Host-Guest Interactions: Instant andAccurate Formation Constant Calculations [J]. Anal Chim Acta, 2000, 417: 221-22.
[26]李颖娇, 叶非. 定量构效关系在农药设计合成中的应用进展[J]. 农药科学与管理, 2002,23(6): 20-23.
[27]Kishino T, Kobayshi K. Studies on the Mechanism of Toxicity of Chlorophenols Found inFish through Quantitative Structure-Activity Relationship [J]. Wat Res, 1996, 30(2):393-399.
[28]Fang Hong, Tong Weida, Welsh W J, et al. QSAR Models in Receptor-Mediated Effects: theNuclear Receptor Superfamily [J]. J Mol Struct (Theochem), 2003, 622: 113-125.
[29]Debnath K, Hansch C, Kim H K, et al. Mechanisic interpretation of the genotoxicity ofnitrofurans (antibacterial age nts) using quantitative structure-activity relationshipsand comparative molecular field analysis [J]. J Med Chem, 1993, 36:1007-1016.
[30]Norinder Ulf. Support Vector Machine Models in Drug Design: Applications to DrugTransport Processes and QSAR Using Simplex Optimizations and Variable Selection [J].Neurocomputing, 2003, 55: 337-346.
[31]Lewis D F V. Structural Characteristics of Human P450s Involved in Drug Metabolism:QSARs and Lipophilicity Profiles [J]. Toxicology, 2000, 144: 197-203.
[32]Dvorsky R, Balaz S, Sawchuk R J. Kinetics of Subcellular Distribution of Compounds inSimple Biosystems and its Use in QSAR [J]. J Theor Biol, 1997, 185: 213-222.
[33]高大文, 王鹏, 郑彤, 等. 多氯酚定量构效关系人工神经网络信息流分析[J]. 中国环境科学,2002, 22(6):561-564.
[34]Basak S C, Gute B D, Mills D, et al. Quantitative Molecular Similarity Methods in theProperty/Toxicity Estimation of Chemicals: a Comparison of Arbitrary versus TailoredSimilarity Spaces[J]. J Mol Struct (Theochem), 2003, 622: 127-145.
[35]Tuppurainen K. Frontier Orbital Energies, Hydrophobicity and Steric Facrtors asPhysical QSAR Descriprtors of Molecular Mutagenicity[J]. Chemosphere, 1999, 38(13):3015-3030.
[36]Walker J D. International workshops on QSARs in the environmental sciences-The first20 years [J]. QSAR & Combinatorial Science, 2003,22(4): 415-421.
[37]曲凡歧，戴志群，胡潇文，等. 植物抗病毒剂的探寻Ⅹ--1,5-二丙酸酯-1,3,5-均三嗪-2,4-二酮的合成[J]. 武汉大学学报（理学版），2002, 48(4): 435-428.
[38]周家驹，王亭. 药物设计中的非模型化方法[M]. 北京：科学出版社，2001. 104.
[39]Yuan H, Parrill A L. QSAR development to describe HIV-1 integrase inhibition [J]. JMol Strut (Themochem), 2000,529:273-282.
[40]Li H, Ung C Y , Yap C W , et al. Predictio of Genotoxicity of Chemical Compounds byStatistical Learnng Methods [J]. Chem Res Toxicol, 2005,18:1071-1080.
[41]杨光富，刘华银，杨秀凤，等. 以 ALS 为把标的新型除草剂的分子设计、合成及生物活性研究.VIII. Hansch 方法与 CoMFA 方法相结合研究稠杂磺酰胺类除草剂的构效关系 [J]. 化学学报，1999，57：706-711.
[42]王连生，支正良，高松亭. 分子结构于色谱保留[M]. 北京: 化学工业出版社，1994.2、113-119.
[43]Boyce C B C., Millborrow B V. A Simple Assessment of Partition Data for CorrelatingStructure and Biological Activity using Thin-layer Chromatography [J]. Nature, 1985, 537.
[44]罗北平，彭晓春，任碧野. 有机物气相色谱保留指数的 QSPR 研究[J]. 吉首大学学报（自然科学版）, 2000, 21(1): 43-49.
[45]刘够生，宋兴福，于建国，等. 人工神经网络模拟芳基脲土壤吸附系数研究[J]. 环境科学学报,2002, 22(3): 359-363.
[46]Tantishaiyakul Vimon. Prediction of the aqueous solubility of benzylamine salts usingQSPR model [J]. J Pharmaceut Biomed, 2005,37: 411-145.
[47]Raevsky Oleg A, Schaper Klaus J. Analysis of water solubility data on the basis of HYBOTdescriptors [J]. QSAR Comb Sci,(QSAR & Combinatorial Science,) 2003,22: 926-942.[QSARg31]
[48]Uddameri Venkatesh, Kuchanur Muthukumar. Fuzzy QSARs for predicting logKoc of persistentorganic pollutants [J]. Chemosphere, 2004,54: 771-776.
[49]Gramatica Paola, Giani Elisa, Papa Ester. Statistical external validation and consensusmodeling: A QSPR case study for Koc predictio [J]. Molecular Graphics and Modeling, 2006,in press.
[50]Schultz T Wayne, Mark T D Cronin, John D Walker, et al. Quantitative Structure-ActivityRelationships (QSARs) in Toxicology: a Historical Perspective [J]. J Mol Struct (Theochem),2003, 622: 1-22.
[51]刘征涛，张颖，徐静波，等. 烷基酚类的生殖干扰毒性与结构相关研究[J]. 环境科学研究,2002,15(6): 39-48.
[52]Toropov Andrey A, Toropova Alla P. QSAR modeling of toxicity on optimization ofcorrelation weights of Morgan extended connectivity [J]. J Mol Struct (Theochem), 2002,578:129-134.
[53]Argese Emanuele, Bettiol Cinzia, Fasolo Matteo, et al. Substituted aniline interactionwith submitochondrial particles and quantitative structure-activity relationshios [J].Biochimica et Biophysica Acta, 2002, 1558: 151-160.
[54]Kapur S, Shusterman A, Verma R P, et al. Toxicology of Benzyl Alcohols: a QSAR Analysis[J]. Chemosphere, 2000, 41: 1643-1649.
[55]Gokhale V M, Kulkarni V M. Understanding the Antifungal Activity of TerbinafineAnalogues Using Quantitative Structure–Activity Relationship (QSAR) Models [J]. BioorgMed Chem, 2000, 8: 2487-2499.
[56]Roy D R, Parthasarath R, Maiti B,et al. Electrophilicity as a possible descriptor fortoxicity prediction[J]. Bioorg Med Chem,2005,13: 3405-3412.
[57]Niederlehner B R, Cairns John, Jr, et al. Modeling Acute and Chronic Toxicity of NonpolarNarcotic Chemicals and Mixtures to Ceriodaphnia dubia [J]. Ecotoxicology and EnvironmentalSafety, 1998,39:136-146.
[58]Juranic Ivan O, Drakulic Branko J, Petrovic Slobodan D, et al. A QSAR study of axutetoxicity of N-substituted fluoroaceamides to rats [J]. Chemosphere, 2006, 62:641-649.
[59]Clare Brian W. QSAR of aromatic substances: Toxicity of polychlorodibenzofurans [J].Theochem, 2006,763: 205-213.
[60]Lin Zhifen, Yu Hongxia, Wei Dongbin, et al. Prediction of mixture toxicity with itstotal hydrophobicity [J]. Chemosphere, 2002,46:305-310.
[61]Pololth Claudia, Mangelsdorf Inge. Commentary on the application of QSAR to thetoxicological evaluation of existing chemicals [J]. Chemosphere, 1997,35(11): 2525-2542.
[62]Zhang Daren. QSPR studies of PCBs by the combination of genetic algorithms and PLSanalysis [J]. Comput Chem, 2001, 25: 197-204.
[63]Veith Gilman D, Mekenyan Ovanes G, Ankley Gerald T, et al. A QSAR analysis of substituenteffects on the photoinduced acute toxicity of PAHs [J]. Chemosphere, 1995,30(11):2129-2142.
[64]Sztandera Les, Garg Ashish, Hayik Seth, et al. Mutagenicity of aminoazo dyes and theirreductive-cleavage metabolites: a QSAR/QSPR investigation [J]. Dyes and Pigments, 2003,59: 117-133.
[65]Abraham M H, Kumarsingh R, Cometto-Muniz J E W S Cain. A Quantitative Structure-ActivityRelationship (QSAR) for a Draize Eye Irritation Database [J]. Toxicol in Vitro, 1998, 12:201-207.
[66]Lewis D F V, Lake B G. Quantitative Structure-Activity Relationship(QSAR) Analysis fora Series of Rodent Peroxisome Proliferators: Interaction with the Mouse Liver PerpxisomeProliferator-activated Receptor α(mPPARα)[J]. Toxicol in Vitro, 1997, 11: 99-105.
[67]Argese E, Bettiol C, Giurin G, et al. Quantitative Structure-Activity Relationshipsfor the Toxicity of Chlorophenols to Mammalian Submitochondrial Particles [J]. Chemosphere,1999, 38(13): 2281-2292.
[68]Ashby J, Tennant R W. Chemical Structure, Salmonella Mutagenicity and Extent ofCarcinogenicity as Indicators of Genotoxic Carcinogenesis among 222 Chemicals tested inRodents by the U.S. NCI/NTP [J]. Mut Res, 1988, 204: 17-115.
[69]Wang Xiaodong, Dong Yuying, Wang Liangsheng, et al. Acute toxicity of SubstitutedPhenols to Rana japonica tadpoles and Mechanism-based Quantitative Structure-activityRelationship (QSAR) Study [J]. Chemosphere, 2001, 44: 447-455.
[70]Ramos E Urrestarazu, Vaes W H J, Mayer P, et al. Algal growth inhibition of Chlorellapyrenoidosa by polar narcotic pollutants: toxic cell concentrations and QSAR modeling [J].Aquat Toxicol, 1999, 46: 1–10.
[71]Furay V J, Smith S. Toxicity and QSAR of Chlorobenzenes in Two Species of BenthicFlatfish, Flounder (Platichthys flesus L.) and Sole(Solea solea L.)[J]. Bull Enviorn Contam.Toxicol, 1995, 54(1): 36-42.
[72]袁星, 袁晓凡, 赵元慧. 取代苯胺、苯酚对鲤鱼毒性的定量构效关系[J]. 东北师大学报（自然科学版）, 2001, 33(1): 70-73.
[73]Kaiser K L E. The Use of Neural Networks in QSARs for Acute Aquatic ToxicologicalEndpoints [J]. J Mol Struct (Theochem), 2003, 622: 85-95.
[74]胥江河. 氯代苯对水生生物毒性的定量构效关系研究[J]. 渝州大学学报(自然科学版), 2000,17(3): 66-70.
[75]Tao S, Xi Xiaohuan, Xu Fuliu, et al. A QSAR model for predicting toxicity (LC 50) torainbow trout [J]. Wat Res, 2002, 36: 2926-2930.
[76]Diane J Blum W, Speece R E. Determining chemical toxicity to aquatic species [J]. EnvironSci Technol, 1990, 24(3): 284-293.
[77]Netzeva Tatiana I, Schultz T Wayne. QSARs for the aquatic toxicity of aromatic aldehydesfrom Tetrahymena data[J]. Chemosphere, 2005, 61: 1632-1643.
[78]Dimitrov S D, Mekenyan O G, Sinks G D, et al. Global Modeling of Narcotic Chemicals:Ciliate and Fish Toxicity [J]. J Mol Struct (Theochem), 2003, 622: 63–70.
[79]Bundy J G, Morriss A W J, Durham D G, et al. Development of QSARs to Investigate theBacterial Toxicity and Biotransformation Potential of Aromatic Heterocylic Compounds[J].Chemosphere, 2001, 42: 885-892.
[80]王连生, 黄庆国, 韩朔睽等. 有机污染化学进展[M], 北京: 化学工业出版社, 1998, 164-168.
[81]黄鸿新, 卢忠, 罗一帆. 氯代酚类化合物对发光细菌毒性的构效关系研究[J]. 西江大学学报,2000, 2: 69-72.
[82]刘够生，宋兴福，于建国，等. 氯代酚类衍生物对水生物发光细菌的定量结构—活性关系研究[J]. 江西师范大学学报（自然科学版），2001，25（4）：313-316.
[83]李卫华, 许旋, 徐志广, 罗一帆. 卤代苯对酵母菌毒性的定量构效关系研究. 数理医药学杂志[J], 2001, 14(2): 168-169.
[84]Patlewicz G Y, Rodford R A, Ellis G, et al. A QSAR Model for the Eye Irritation of CationicSurfactants[J]. Toxicol in Vitro, 2000, 14: 79-84.
[85]Barratt M D. Quantitative structure-activity relationships(QSARs) for skin Corrosivityof Organic acids, bases and phenols: principal components and neural network analysis ofextended datasets[J]. Toxicol in Vitro, 1996, 10: 85-94.
[86]Patel H, Berge W ten, Cronin M T D. Quantitative structure-activity relationships(QSARs)for the prediction of skin permeation of exogenous chemicals[J]. Chemosphere, 2002, 48:603-613.
[87]Whittle E, Battatt M D, Carter J A, et al. Skin Corrosivity Potential of Fatty Acids:In Vitro Rat and Human Skin Testing and QSAR Studies [J]. Toxicol in Vitro, 1996, 10: 95-100.
[88]Chamberlain M, Barratt M D. Practical Applications of QSAR to In Vitro ToxicologyIllustrated by Consideration of Eye Irritation [J]. Toxicol in Vitro, 1995, 9(4): 543-547.
[89]Cronin M T D, Schultz T W. Pitfalls in QSAR [J]. J Mol Struct (Theochem), 2003, 622:39-51.
[90]Walker John D. Applications of QSARs in toxicology: a US Government perspective [J].J Mol Struct (Theochem), 2003,622: 167-184.
[91]Kier L B, Hall L H. Molecular Connectivity in Chemistry and Drug Research [M]. New York,Academic Press: 1976.
[92]Karelson, M. Molecular Descriptors in QSAR/QSPR [M]. New York, John Wiley & Sons: 2000.
[93]Balaban A T. From chemical topology to 3D geometry [J]. J Chem Inf Comput Sci, 1997,37:645-650.
[94]Balaban A T. Topological and stereochemical molecular descriptors for dastabased usefulin QSAR, similarity/dissimilarity and drug design [J]. SAR QSAR Environ Res, 1998, 8: 1-21.
[95]Estrada E, Uriarte E. Recent advances on the role of topological indices in drugdiscovery research [J]. Curr Med Chem, 2001,8: 1699-1714.
[96]Estrada E, Molina E. 3D connectivity indices in QSPR/QSAR studies [J]. J Chem Inf ComputSci, 2001,41: 791-797.
[97]Gasteiger J, Sadowski J, Selzer P, et al. Chemical information in 3D space [J]. J ChemInf Comput Sci, 1996,36: 1030-1037.
[98]Gasteiger J, Burkard S U, Hemmer M C, et al. Decision support systems for chemicalstructure representation, reaction modeling and spectra simulation [J]. SAR & QSAR inEnviron Res, 2002, 13: 89-110.
[99]Engel T, Gasteiger J. Chemical structure representation for information exchange[J].Online Inform Rev, 2002, 26: 139-145.
[100]Taft R M. Polar and steric substituent constants for aliphatic and o-benzoate groupsfrom rates of esterification and hydrolysis of esters [J]. J Am Chem Soc, 1952, 74(12):3120-3128.
[101]Livingstone D J. The characterization of chemical structures using molecularproperties. A survey [J]. J Chem Info Compt Sci, 2000, 40(2): 195-209.
[102]Hansch C, Maloney P P, Fujita T, et al. Correlation of biological activity ofphenoxyacetic acids with Hammett substituent constants and partition coefficients [J].Nature, 1962, 194:178-180.
[103]Hansch C, Muir R M, Fujita T, et al. The correlation of biological activity of plantgrowth regulators and chloromycetin derivatives with Hammett constants and partitioncoefficients [J]. J Am Chem Soc, 1963，85(18)：2817-2824.
[104]Wang R, Fu Y, Lai L. A New Atom-additive Method for Calculating Partition Coefficients[J]. J Chem Inf Comput Sci, 1997, 37: 615-621.
[105]Klopman G, Namboodiri K, Schochet M. Simple Method of Computing the PartitionCoefficient [J]. J Chem Chem 1985, 6: 28-38.
[106]Bodor N, Gabanyi Z, Wong C K. A New Method for the Estimation of Partition Coefficient[J]. J Am Chem Soc, 1989, 111: 3783-3786.
[107]Ghose A K, Croppen G M. Atomic physicochemical parameters for three-dimensionalstructure-directed quantitative structure-activity relationships. 1. Partitioncoefficients as a measure of hydrophobicity [J]. J Comput Chem, 1986,7: 565-577
[108]Hansch C, Leo A J. Substituent Constants for Correlation Analysis in Chemistry andBiology [M]. New York: Wiley and Sons, 1979. 1-339.
[109]Livingstone D J. The characterization of chemical structures using molecularproperties. A survey [J]. J Chem Inf Comput Sci, 2000, 40(2): 195-209.
[110]Selwood D L, Livingstone D J, Comley J C W. Structure-activity relationships ofantifilarial antimycin analogs: a multivariate pattern recognition study [J]. J Med Chem.1990, 33(1): 136-142.
[111]Taft R W, Kamlet M J. The solvatochromic comparison method. 2. The alpha-scale ofsolvent hydrogen-bond donor (HBD) acidities [J]. J Am Chem Soc, 1976,98(10):2886-2894.
[112]Balaban A T. Using real numbers as vertex invariants for third-generation topologicalindexes [J]. J Chem Inf Comput Sci, 1992,32: 23-28.
[113]Wiener H J. Structural determination of paraffin boiling points [J]. J Chem Soc, 1947,69: 17-20.
[114]王连生. 环境化学进展[M]. 北京: 化学工业出版社, 1995. 105-106.
[115]王连生，支正良，高松亭. 分子结构于色谱保留[M]. 北京: 化学工业出版社，1994.216-217.
[116]Randic M. On characterization of molecular branching [J]. J Am Chem Soc, 1975, 97(23):6609-6615.
[117]Kier L B. A shape index from molecular graphs [J]. Quantitative Structure-ActivityRelationships, 1985, 4: 109-116.
[118]Kier L B. Shape indexes of orders one and three from molecular graphs [J]. QuantitativeStructure-Activity Relationships, 1986, 5(1): 1-7.
[119]Kier L B, Hall L H. The nature of structure-activity relationships and their relationto molecular connectivity [J]. Eur J Med Chem, 1977, 12: 307-312.
[120]Hu C Y, Xu L. J Chem Inf Comput Sci, 1996, 36: 82.
[121]Randic M. On molecular identification numbers [J]. J Chem Inf Comput Sci, 1984, 24(3):164-175.
[122]Hansen P J, Jurs P C. Chemical applications of graph theory. Part I. Fundamentals andtopological indices [J]. J Chem Educ, 1988,65(7): 574-580.
[123]Schultz H P. Topological organic chemistry. 1. Graph theory and topological indicesof alkanes [J]. J Chem Inf Comput Sci, 1989,29(3): 227-228.
[124]Muller W R, Szymanski K, Knop J V, et al. An algorithm for construction of the moleculardistance matrix [J]. J Comput Chem, 1987,8(2): 170-173.
[125]Schultz H P, Chultz E B, Schultz T P. Topological organic chemistry. 2. Graph theory,matrix determinants and eigenvalues, and topological indexes of alkanes [J]. J Chem InfComput Sci, 1990,30(1): 27-29.
[126]Schultz H P. Topological organic chemistry. 3. Graph theory, binary and decimaladjacency matrices, and topological indices of alkanes [J]. J Chem Inf Comput Sci,1991,31(1): 144-147.
[127]Heitler W, London F. Interaction Between Neutral Atoms and Homopolar H2[J]. Zeitschriftfür Physik, 1927, 44: 455.
[128]封继康. 基础量子化学原理[M]. 北京: 高等教育出版社, 1984. 346-351
[129]唐敖庆，杨忠志，李前树. 量子化学[M]. 北京：科学出版社. 1982.
[130]Hartree D. Calculations of atomic structure. Wiley. 1957.
[131]Born M, Oppenheimer J R, Quantum theory of molecules. Ann Physik, 1927, 84: 457
[132]Born M, Huang K, Dynamical theory of crystal lattices. Oxford University Press, NewYork, 1954.
[133]Lowdin P O.，Correlation problem in many-electron quantum mechanics [J]. I.Review ofdifferent approaches and discussion of some current ideas [J]. Adv Chem Phys, 1959, 2:207-332.
[134]Bauschlicher C W Jr, Langhoff S R, Accurate quantum chemical calculation [J]. Adv ChemPhys, 1990, 77: 103-161.
[135]Hohenberg P, Kohn W, Inhomogeneous electron gas [J]. Phys Rev B, 1964, 136: 864-871.
[136]Kohn W, Sham L J. Self-consistent equations including exchange and correlation effects[J]. Phys Rev, 1965,140:1133.
[137]Becke A D, Density-functional exchange-energy approximation with correct asymptoticbehavior [J]. Phys Rev A, 1988, 38: 3098-3100.
[138]Perdew J P, Wang Y., Phys. Rev. B, 1986, 33: 8800.
[139]Perdew J P, Chevary J A, Vosko S H, et al, Phys. Rev. B, 1992, 46:6671.
[140]Lee C, Yang W, Parr R G, Development of the Colle-Salvetti correlation-energy formulainto a functional of the electron density [J]. Phys Rev B, 1988, 37: 785-789.
[141]Stevens P J, Devlin J F, Chabalowski C F, et al. J. Phys. Chem., 1994, 98: 11623.
[142]Becke A D, J. Chem. Phys., 1995, 104: 1040.
[143]Bader R F W, Beddall P M. J. Chem. Phys., 1972,56:3320.
[144]Karelson M, Loganov VS. Quantum-chemical descriptors in QSAR/QSPR studies [J]. ChemRev, 1996,96:1027-1043.
[145]Franke R. Theoretial drug design methods [M]. Elsevier: Amsterdam, 1984.11.
[146]Fukui K. Theory of orientation and stereoselection [M]. New York: Springer Verlag,1975.34.
[147]刘次全. 量子生物学[M].北京：高等教育出版社，1990.12-79.
[148]朱永，韩世纲，朱平仇.量子有机化学[M].上海:上海科学出版社,1986.1177~183.
[149]Green M Stuart, Maeshall R Garland. 3D-QSAR: a current perspertive [J]. Tips, 1995,16:28-291.
[150]Vijay M Gokhale, Vithal M Kulkami. Comparative molecular field analysis of fungalsqualene epoxidase inhibitors [J]. J Med Chem, 1999,42:5348-5358.
[151]Elizabeth R Collantes, Li Xing. Comparative molecular field analysis as a tool toevaluate mode of action of chemical hybridization agents [J]. J Agric Food Chem, 1999,47:5245-5251.
[152]Kamlet M J, Abraham M H, Doherty R M, et al. Solubility properties in polymers andbiological media. 4. Correlations of octanol/water patition coefficients withsolvatochromic parameters [J]. J Am Chem Soc, 1984, 106(2): 464-466.
[153]Kamlet M J, Doherty R M, Abboud J-L M, et al. Solubility: a new look [J]. Chemtech,1986,16(9): 566-576.
[154]Kamlet M J, Doherty R M, Beith G D et al. Solubility properties in polymers andbiological media. 7. An analysis of toxicant properties that influence inhibition ofbioluminescence in Photobacterium phosphoreum (the Microtox test)[J]. Environ Sci Tech,1986,20(7):690-695.
[155]Kamlet M J, Doherty R M, Taft R W, et al. Solubility properties in polymers andbiological media. 8. An analysis of the factors that influence toxicities of organicnonelectrolytes to the Golden Orfe Fish (Leuciscus idus melanotus)[J]. Environ Sci Tech,1987,21(2): 149-155.
[156]Kamlet M J, Doherty R M, Carr P W, Linear solvation energy relationships, 44. Parameterestimation rules that allow accurate prediction of octanol/water partition coeffients andother solubility and toxicity properties of polychlorinated biphenyls and polycyclicaromatic hudrocarbons [J]. Environ Sci Tech, 1988,22(5): 503-509.
[157]Kier L B, Hall L H. Molecular Connectivity in Chemistry and Drug Research [M]. NewYork: Academic Press, 1996.
[158]Kier L B, Hall L H. Molecular Connectivity in Structure-Activity Analysis [M]. NewYork: John Wiley & Sons, 1986.
[159]Free S M, Wilson J M. A mathematical contribution to structure-activbity studies [J].J Med Chem, 1964,7(4):395-399.
[160]Draper N R, Smith H. Applied Regression Analysis[J]. New Youk: John Wiley & Sons, Inc.1981.
[161]张尧庭，方开泰.多元统计分析引论[M].北京：科学出版社，1983.
[162]Eriksson L, Johansson E, Kettaneh-Wold N, et al. Multi- and Megavariate Data Analysis:Principles and Applications [M]. Ume?: Umetrics, 2001.
[163]Livingstone D J, Salt D W. Judging the significance of multiple linear regressionmodels [J]. J Med Chem, 2005,48(3): 661-663.
[164]Emanuele Argese, Cinzia Bettiol, Matteo Fasolo, et al. Substituted aniline interactionwith submitochondrial particles and quantitative structure-activity relationshios [J].Biochimica et Biophysica Acta, 2002, 1558: 151-160.
[165]Cronin M T D, Netzeva T I, Dearden J C, et al. Assessment and modeling of the toxicityof organic chemicals to Chlorella vulgaris: development of a novel database [J]. Chem ResToxicol, 2004,17(4)545-554.
[166]Thomas Leonard J., Roy Kunal. QSAR by LFER model of HIV protease inhibitor mannitolderivatives using FA-MLR, PCRA, and PLS techniques [J]. Bioorgan Med Chem, 2006,14:1039-1046.
[167]Shen Qi, Jiang Jian Hui, Jiao Chen Xu, et al. Modified particle swarm optimizationalgorithm for variable selection in MLR and PLS modeling: QSAR studies of antagonism ofangiotensin Ⅱ antagonists [J]. Eur J Pharm Sci, 2004,22: 145-152.
[168]Lü Jian Xia, Shen Qi, Jiang Jian Hui, et al. QSAR analysis of cyclooxygenase inhibitorusing particle searm optimization and multiple linear regression[J]. J Pharmaceut Biomed,2004,35:679-687.
[169]Hemmateenejad Bahram，Miri Ramin, Akhond Morteza, et al. QSAR study of the calciumchannel antagonist activity of some recently synthesized dihydropyridine derivatives. Anapplication of genetic algorithm for variable selection in MLR and PLS methods [J].Chemometr Intell Lab, 2002,64: 91-99.
[170]Franke R., Gruska A. In Chemometric Methods in Molecular Design [M]. VCH: Weinheim,1995. 113.
[171]Franke R. Theoretical Drug Design Methods [M]. Elsevier: Amsterdam, 1984. 184.
[172]Kirschner G, Kowalski B R. The application of pattern recognition on drug design, inDrug Design [M]. New York: Academic Press, 1979. Vol.Ⅲ.
[173]Brown R D, Martin Y C. Use of structure-activity data to compare structure-basedclustering methods and descriptors for use in compound sdelection [J]. J Chem Inf ComputSci, 1996,36(3):572-584.
[174]Senese C L, Hopfinger A J. A simple clustering technique to improve QSAR model selectionand predictivity: application to a receptor independent 4D-QSAR analysis of Cyclic Ureaderived inhibitors of HIV-1 protease [J]. J Chem Inf Comput Sci, 2003,43(6):2180-2193.
[175]Klein C T, Kaiser D, Ecker G. Topological distance based 3D descriptors for use inQSAR and diversity analysis [J]. J Chem Inf Comput Sci, 2004,44(1):200-209.
[176]Barnard J M, Downs G M. Clustering of chemical structures on the basis oftwo-dimensional similarity measures [J]. J Chem Inf Comput Sci, 1992,32(6):644-649.
[177]任若恩，王惠文.多元统计数据分析-理论、方法、实例.北京：国防工业科学出版社，1999.
[178]刘增良, 刘有才. 模糊逻辑与神经网络技术[M]. 北京航天航空大学出版社, 1996.121.
[179]焦李成.神经网络系统理论[M].西安：西安电子科技大学出版社，1992.
[180]焦李成.神经网络计算[M].西安：西安电子科技大学出版社，1993.
[181]焦李成.神经网络的应用与实现[M].西安：西安电子科技大学出版社，1993.
[182]蒋宗礼.人工神经网络导论[M].北京：高等教育出版社，2001.
[183]Broonhead, Lowe D. Multivariable function interpretation and adaptive network [J].Complex Systems, 1988,2.
[184]Moody J E, Darken C.J. Fast algorithms in networks of locally tuned processing units[J]. Neural Computation, 1989,1:282-289.
[185]Jackon I R H. Convergence properties of radial basis function [J]. Approx,1988,4:243-264.
[186]Tyler Holcomb, Manfred Morari. Local training for radial basis function metworkstoward solving the hidden unit problem[C]. American Control Conference,1993,3(1).
[187]Mark J L Orr. Introduction to radial basis function networks [G].1996.
[188]Mark Orr, Hallam J. Assessing RBF networks using DELVE [J]. Intermational Journal ofNeural System, 2000, 10(5): 397-415.
[189]吴海龙，梁逸曾，俞汝勤.分析化学计量学[M].分析实验室，1999,18(6):94-102.
[190]Harrington Peter de B..Temperature-Constrained Cascade Correlation Networks [J]. Anal.Chem., 1998,70:1297-1306.
[191]Cai Chunsheng, Harrington Peter de B. Wavelet Transform Preprocessing for TemperatureConstrained Cascade Correlation Neural Networks[J].J Chem InfComput Sci,1999,39:874-880.
[192]Harrington Peter de B, Urbas Aaron, Wan Chuanhao. Evaluation of Neural Network Modelswith Generalized Sensitivity Analysis [J]. Anal Chem, 2000,72:50004-5013.
[193]Wan Chuanhao, Harrington Peter de B. Screening GC-MS data for carbamate pesticideswith temperature-constrained-cascade correlation neural networks [J].Analytical ChinicaActa ,2000, 408:1-12.
[194]刘思东，崔秀君，张卓勇等.温度限制串联相关网络用于有机环境污染物紫外光谱的识别[J].光谱学与光谱分析， 2003（1）：119-122.
[195]崔秀君，张卓勇，Harrington Peter de B，等. 温度限制串联相关网络-近红外光谱法用于药物甲硝唑的质量控制[J].高等学校化学学报，2004，25（7）：1251-1253.
[196]Cui Xiujun, Zhang Zhuoyong, Ren Yulin, et al. Quality control of the powderpharmaceutical samples of sulfaguanidine by using NIR reflectance spectrometry andtemperature-constrained cascade correlation networks [J]. Talanta, 2004,64:943-948.
[197]张卓勇.温度限制串联相关神经网络及其在细菌辨识中的应用，高等学校化学学报[J].2002（4）：570-572.
[198]Wang Fengxia, Zhang Zhuoyong, Cui Xiujun, et al. Identification of rhubarbs by usingNIR spectrometry and temperature-constrained cascade correlation networks [J]. Talanta,2006, (in press).
[199]Vapnik V. The Nature of Statistical Learning Theory [M]. Springer Verlag, 1995.
[200]Vapnik V. Statistical Learning Theory[M].New York: Wiley,1998.
[201]Burges C J C . A tutorial on support vector machines for pattern recognition [J]. DataMin Knowl Discov,1998,2:121-167.
[202]Cristianini N, Shawe-Taylor J. Introduction to support vector machine and other kernelbased learning methods [M]. Cambridge：Cambridge University Press, 2000.
[203]张学工.关于统计学习理论与支持向量机[J].自动化学报，2000, 26 (1): 32-42.
[204]边肇棋.模式识别[M]. 北京:清华大学出版社，2000.
[205]Vapnik V.Statistical learning theory [M].New York:John Wiley and Sons,1998.
[206]Wolfe P. A duality theory for nonlinear programming. Quarterly of Mathematics [J].1961, 19:239-244.
[207]Chinnasamy Arunkumar, Mittal Ankush, Sung WingKin Probabilistic prediction ofprotein-proteininteractionsformtheproteinsequences[J].ComputBiolMed,2006,36:1143-1154.
[208]Lumini Alessandra, Nanni Loris. Machine learning for HIV-1 protease cleavage siteprediction [J].Pattern Recognition Letters,2006,27:1537-1544.
[209]Chen Chao, Zhou Xibin, Tian Yuanxin, et al. Predicting Protein structural class withpseudo-amino acid composition and suppout vector machine fusion network[J]. AnalBiochem,2006,357:116-121.
[210]Dubey Anshul, Realff Matthew J, Lee Jay H, et al. Identifying the interacting positionsof a protein using Boolean learning and support vector machines [J]. Comput. Biol. Chem.,2006,30:268-279.
[211]Brown M, Grundy W, Lin D, et al. Knowledge-based analysis of microarry gene expressiondata using support vector machines [J]. Proc Natl Acad Sci USA.,2000,97:262-267.
[212]Guyon I, Weston J, Barnhill S, et al. Gene selection for cancer classification usingsupport vector machines [J]. mach Learn, 2002,46:389-422.
[213]Cai Y D, Liu X J, Xu X B, et al. Prediction of protein structural classes by supportvector machines [J]. Comput Chem 2002,26:293-296.
[214]Cai Y D, Liu X J, Xu X B, et al. Support vector machines for the classification andprediction of beta-turn types [J]. J Pept Sci, 2002,8:297-301.
[215]Hua S, Sun Z. Novel method of protein secondary structure prediction with high segmentoverlap measure: support vector machine approach [J]. J Mol Bio, 2001,308:397-407.
[216]Ding C H Q, Dubchak I. Multi-class protein fokd recognition using support vectormachines and neural networks [J]. bioinfornatics, 2001,17:349-358.
[217]Cai YD, Liu X J, Xu X B, et al. Support vector machines for prediction of proteinsubcellular location by incorporation gquasi-sequence-order effect [J]. J Cell Biochem,2002,84:343-348.
[218]Hua S, Sun Z. Support vector machine approach for protein subcellular localizationprediction [J]. Bioinformatics, 2001,17:721-728.
[219]Bock J R, Gough S A. Predictiong protein-protein interactions from primary structure[J]. Bioinformatics, 2001,17:455-460.
[220]Darchin, R, Darplus K, Huassler D. Classifying G-protein coupled receptors withsupport vector machines [J]. Nioinformatics, 2002,18:147-159.
[221]Lee D, choi . W, Kim M, et al. Discovery of differentially expressed genes relatedtohistologicalsubtypeofhepatocellularcarcinoma[J].biotechnolProg,2003,19:1011-1015.
[222]Xue Y M, Yap C W, Sun L Z, et al. Prediction of P-glycoprotein substrates by a supportvector machine approach [J]. J Chem Inf Comput Sci, 2004,44:1497-1505.
[223]Zhao C Y., Zhang R S, Liu H X, et al. Diagnosing anorexia based on partial least squares,back propagation neural network and support vector machines [J]. J Chem Inf Comput Sci,2004,44:2040-2046.
[224]Xue C X, Zhang R S, Liu H X, et al. An accurate QSPRstudy of O-H bond dissociationenergy in substituted phenol base on support vector machines [J]. J Chem Inf Comput Sci,2004,44:669-677.
[225]Xue C X, Zhang R S, Liu M C, et al. Study of the quantitative structure-mobilityrelationship of carboxylic acids in capillary electrophoresis baded on support vectormachines [J]. J Chem Inf Comput Sci,2004,44:950-957.
[226]Xue C X, Zhang R S, Liu H X, et al. Support vector machines-based quantitativestructure-property relationship for the prediction of heat capacity [J]. J Chem Inf ComputSci, 2004,44:1267-1274.
[227]Xue C X, Zhang R S, Yao X J, et al. prediction of the isoelectric point of an aminoacid based on GA-PLS and SVMs [J]. J Chem Inf Comput. Sci, 2004,44:161-167.
[228]Luan F, Zhang R, Yao X, et al. support vector machine-based QSPR for the predictionof Van de Waal’constants [J]. QSAR & comb Sci, 2005,24:227-239.
[229]Sadik O, Land Jr, Wanekaya A K, et al. Detection and classification of organophosphatenerve agent simulants using support vector machines with multiarray sensors [J]. J chem.Inf Comput Sci, 2004,44:499-507.
[230]Liu Y. A comparative study on feature selection methods for drug discovery [J]. J ChemInf Comput Sci, 2004,44:1823-1828
[231]Xue Y, Li Z R, Yap C W, et al. Effect of molecular descriptor feature selection insupport vector machine classification of pharmacokinetic and toxicological properties ofchemical agents [J]. J Chem Inf Comput Sci, 2004,44:1630-1638.
[232]Sorich M J, Miners J O, Mckinnon R A, et al. Comparison of linear and nonlinearclassification algorithms for the prediction of drug and chemical metabolism by humanUDP-Glucuronosylltransferase isoforms [J]. J Chem Inf Comput Sci, 2003,43:2019-2024.
[233]Yao X J, Panaye A, Doucet J P, et al. Comparative study of QSAR/QSPR correlations usingsupport vector machines, radial basis function neural networks, and multiple linearregression[J]. J Chem Inf Comput Sci, 2004,44:1257-1266.
[234]Liu H X, Zhang R S, Yao X J, et al. QSAR study of ethyl2-[(3-methyl-2,5-dioxo(3-pyrrolinyl))amino]-4-(trifluoromethyl)pyrimidine-5-carboxylate:an inhibitor of AP-1 and NF-1 and NF-kB mediated gene expression based on support vectormachines[J]. J Chem Inf Comput Sci, 2003,43:1288-1296.
[235]刘亮，包新华，冯建星，等,α-唑基-α-芳氧烷基频哪酮（芳乙酮）及其醇式衍生物抗真菌活性的分子筛选[J].计算机与应用化学，2002,19(4):476-479.
[236]Shen Qi, Shi WeiMin, Kong Wei, et al. A conbination of modified particle swarmoptimization algorithm and support vector machine for gene selection and tumorclassification [J]. Talanta, 2006,(in press)
[237]Furey T, Cristianini N, Cuffy N, et al. Support vector machine classification andvalidation of cancer tissue samples using microarray expression data [J]. Bioinformatics,2000,16:906-914.
[238]Ramaswamy S, Tamays P, Rifkin R, et al. Multiclass cancer diagnosis using tumor geneexpression signatures [J]. Proc Natl Aatl Acad Sci USA, 2001,98:15149-15154.
[239]Luan F, Zhang R, Zhao C, et al. Classification of the carcinogenicity of N-nitrosocompounds based on support vector machines and linear discriminant analysis [J]. Chem ResToxicol, 2005,18:198-203
[240]Takaoka Y, Endo Y, Yamanobe S, et al. Development of a Method for EvaluatingDrug-Likeness and Ease of Synthesis Using a Data Set in Which Compounds Are Assigned ScoresBased on Chemists’ Intuition [J]. J Chem Inf Comput Sci, 2003,43:1269-1275.
[241]Ivancius O. Support vector machine identification of the aquatic toxicity mechanismof organic compounds[J]. Internet Electronic Journal of Molecular Design, 2002,1:151-172.
[242]Li H, Ung C Y, Yap C W, et al. Prediction of Genotoxicity of Chemical Compounds byStatistical Learning Methods [J]. Chem Res Toxicol,2005,18:1071-1080.
[243]Liu H X, Zhang R S, Yao X J, et al. QSAR and classification models of a novel seriesof COX-2 selective inhibitors: 1,5-Diarylimidazoles based on support vector machines[J].J Comput Aid Mol Des,2004,18:389-399.
[244]Yang S S, Lu W C, Chen N Y, et al. Support vector regression based QSPR for the predictionof some physicochemical properties of alkyl benzenes [J]. J Mol Struct(Theochem),2005,719:119-127.
[245]Tugcu N, Song M, Breneman C M, et al. Prediction of the effect of mobile-phase salttype on protein retention and selectivity in anion exchange systems [J]. Anal Chem,2003,75:3563-3572.
[246]Yao X J, Panaye A, Doucet J P, et al. Comparative classification study of toxicitymechanisms using support vector machines and radial basis function neural networks [J].Anal Chim Acta, 2005,535,259-273.
[247]Hartter D R. The use and importance of nitroaromatic chemicals in the chemical industry[A]. In: Rickert D (Ed). Toxicity of nitroaromatic compounds[C]. Washington, DC: Hemisphere,1985,1-3.
[248]Rosenblatt D H, Burrows E P, Mitchell W R, et al. Organic explosives and relatedcompounds [A]. In: Hutzinger O(Ed). The Handbook of Environmental Chemistry—AnthropogenicCompounds [C].Berlin:spinger Verlag,1991,195-237.
[249]Adkins R L. Nitrobenze and nitrotoluene [A]. In Grant MH (Hd). Kirk Othmer encyclopediaof chemical technology 4ed[C]. New York: Wiley Interscience,1994,133-152.
[250]Yamagisshi T, Miyazaki T, Horii S, et al. Identificatin of musk Xylene and musk ketonein fresh water fish collected from the Tama River Tokyo [J]. Bull Environ Contem Toxic,1981,26(10):656-662.
[251]Yurawecz M P, Puma B J. Nitro musks fragrances as potential contaminants in pesticideresidue analysis [J]. J. Ass. Off. Analyst. Chem., 1983,66(4): 241-247.
[252]Gatermann R, Huhnerfuss H, Rimkus G, et al. The distribution of nitrobenzene and othernitroaromatic compounds in the North Sea [J]. Marine Pollution Bulletin,1994,30(3):221-227.
[253]郎佩珍，龙凤山，袁星，等. 松花江中游（哨口-松花江村段）水中有毒有机物污染研究[M]. 环境科学进展，1993，1（6）：47-55.
[254]于常荣，曹喆，王炜，等. 松花江鱼类有机污染物的研究[J]. 中国环境科学，1994，14（4）：283-287.
[255]康跃惠，宫正宇，王子建，等. 宫厅水库及永定河水库中挥发性有机物分布规律研究[J]. 环境科学学报，2001，21（3）：338-343.
[256]王子建，吕怡兵，王毅，等. 淮河水取代苯污染物及其生态风险[J]. 环境科学学报，2002，22（3）：300-304.
[257]王宏，杨霓云，沈英娃，等. 海河流域几种典型的有机污染物环境安全性评价[J]. 环境科学研究，2003，16（6）：35-36.
[258]胡国华，赵沛伦. 黄河猛津-花园口区间有机污染分析及防治对策[J]. 人民黄河，1995，11：6-9。
[259]田怀军，舒为群，张学奎，等. 长江、嘉陵江（重庆段）源水有机污染物的研究[J]. 长江流域资源与环境，2003，12（2）：118-123.
[260]Huang Hong, Wang Xiaodong, Ou Wenhua,et al., Acute toxicity of benzene derivativesto the tadpoles (Rana japonica) and QSAR analyses[J]. Chemosphere, 2003,53: 963-970.
[261]Ren Shijin. Ecotoxicity prediction using mechanism – and non-mechanism-based QSARs:a preliminary study [J]. Chemosphere, 2003,53: 1053-1069.
[262]Trohalaki S, Gifford E, Pachter R. Improved QSARs for predictive toxicology ofhalogenated hydrocarbons [J]. Comput Chem, 2000, 24:421-427.
[263]Renata Dias de Mello Castanho Amboni, Berenice da Silva Junkes, Vilma Edite FonsecaHeinzen, et al., Semi-empirical topological method for prediction of the chromatographicretention of esters[J]. J Mol Struc-Theochem. 2002,579:53-62.
[264]齐玉华，许禄，王淑云.量化参数在黄酮类化合物构效关系研究中的应用[J]. 计算机与应用化学，2000，17（1）：29-31.
[265]Carbó-Dorca R, Amat L, Besalú E, et al. Quantum mechanical origin of QSAR: theory andapplications [J]. J. Mol. Strut.（Theochem）, 2000,504:181-228.
[266]籍国东，赵元慧，袁星. 量化参数及其在定量结构-活性-性质相关研究中的应用[J]. 东北师大学报自然科学版，1998，4:47-53.
[267]Worrall Fred, Thomsen Marianne. Quantum vs. topological descriptors in the developmentof molecular models of groundwater pollution by pesticides [J]. Chemophere,2004,54:585-596.
[268]周文富，戈芳.Ph(x)n位电荷密度能与生物毒性定构活性相关研究[J]. 广西大学学报（自然科学版），1997,9:218-221.
[269]齐玉华，许禄.应用量化参数和CoMFA法研究苯甲酸类化合物的结构和其pKa值的相关性[J]. 应用化学，2002,19(11):1054-1058.
[270]裴洪平，许高金.量子化学参数用于苯胺类化合物的 QSAR 毒性研究[J]. 浙江大学学报（理学版）2003,30(3):310-313。
[271]Brian W Clare, claudiu T Supuran, Semi-empirical atomic charges and dipole momentsin hypervalent sulfonamide molecules: descriptors in QSAR studies [J]. J Mol Struc-Theochem,1998, 428:109-121.
[272]Li Laicai, Maoshuang, Zhao Keqing, et al. Semi-empirical quantum chemical study onstudy on structure-activity relationship in monocyclic-β-lactam antibiotics [J]. J MolStruc- Theochem, 2001,545: 1-5.
[273]Emanuele Argese, Cinzia Bettiol, Matteo Fasolo, et al. Substituted aniline interactionwith submitochondrial particles and quantitative structure-activity relationshios [J].Biochimica et Biophysica Acta, 2002, 1558: 151-160.
[274]李卫华, 许旋. 氯苯胺类化合物对发光细菌毒性的量子化学研究[J]. 华南示范大学学报(自然科学版), 2001, 1: 76-78.
[275]Stefan Sixt, Joachim Altschuh, Rainer Brüggemann. Quantitative structure-toxicityrelationships for 80 chlorinated compounds using quantum chemical descriptors [J].Chemosphere, 1995, 30(12): 2397-2414.
[276]Li Zhang, Jian Wan, Guanfu Yang. A DFT-based QSARs study of protoporphyrinogen oxidaseinhibitors [J]. phenyl triazolinones. 2004, 12:6183-6191.
[277]Stefan P. Niculescu. Artificial neural networks and genetic algorithms in QSAR [J].J Mol Struc- Theochem, 2003, 622: 71-83.
[278]Tomas-Vert F Perez-Gimenez, Ma T Salabert-Salvador, Garcia-March F J, et al.Artificial neural network applied to the discrimination of antibacterial activity bytopological methods [J]. J Mol Struc- Theochem, 2000,504:249-259.
[279]Yan Aixia, Jiao Guimei, Hu Zhide, et al. Use of artificial neural networks to predictthe gas chromatographic retention index data of alkylbenzenes on carbowax-20M [J]. ComputChem, 2000, 24:171-179.
[280]Gonzalez-Arjona D, Lopez-Perez G, Gustavo Gonzalez A. Non-linear QSAR modeling byusing multiplayer perceptron feedforward neural networks trained by back-propagation [J].Talanta, 2002,56:79-90.
[281]Agatonovic-Kustrin S, Beresford R, Pauzi A Yusof M. ANN modeling of the penetrationacross a polydimethylsiloxane membrane from theoretically derived molecular descriptors[J]. J Pharmaceut Biomed, 2001,26: 241-254.
[282]章文军，许禄.定量结构-活性/性质相关性中变量的选择----正交变换法与最优子集回归法的比较[J]. 高等学校化学学报，2001，22:1134-1136.
[283]Becke, A D. Density-functional thermochemistry. III. The role of exact exchange [J].J Chem Phys, 1993, 98(7): 5648–5652.
[284]Frisch M J， Trucks G W， Schlegel H B， et al． Gaussian 98, Revision A.9， Gaussian,Inc, Pittsburgh PA, 1998.
[285]许禄，邵学广. 化学计量学方法（第二版）[M]. 北京:科学出版社， 2004,291.
[286]金相灿.有机化合物污染化学[M].北京：清华大学出版社，1990.
[287]王红娟，奚红霞，夏启斌等.含酚废水处理技术的现状与开发前景[J].工业水处理，2002,22(6):6-9.
[288]巨华东.水环境污染概论[M].北京:北京师范大学出版社，1989:112-157.
[289]尹伊伟. 用氯苯及酚类化合物研究大鳞副泥鳅急性毒性的定量结构一活性关系[J].暨南大学学报，1998，19：112-115.
[290]Nimrod A C, Bcnson W H. Environmental estrogenic effects of Ikylphenol ethoxylates[J]. Crit Rew Toxicol, 1996,2 (3): 335-364.
[291]Jobling S Sheahao D, Osbome J A, et al. Inhabition of testicular growth in minbow troutexposed to estrogenic alkylphenolic chemicals [J].Environ Toxicol chem,1996, 15(2):194-202.
[292]Harries J E, Sheahan D A, Jobling S. Estrogenic activity in five United Kingdom riversdetected by measurement of vitellogenexis in caged trout [J].Environ Health Toxicol.Chem.1997, 16(3):534-542.
[293]Kaiser K L, Dixon D G, Hodson P V. SAR in Environmental Toxicology[M].Boston: D. ReidelPublishing Company, 1984,189-206.
[294]许旋，罗一帆，徐志广，等.氯苯酚电子结构对水生物毒性的研究[J]. 卫生毒理学杂志，2001，15（3）：167-170.
[295]刘够生，宋兴福，于建国，等.氯代苯酚类衍生物对水生物发光细菌的定量结构-活性关系研究[J].江西师范大学学报（自然科学版），2001，25（4）：313-316.
[296]许禄，吴亚平.苯酚类化合物的三维定量构效关系研究[J].计算机与应用化学，2000,17(1)：15-17.
[297]张大仁.酚取代衍生物的 QSAR 研究[J]. 环境科学，1995，16（2）：4-6.
[298]秦正龙，冯长君.取代苯酚的定量结构-活性/性质相关性研究[J].有机化学，2003，23（7）：654-658.
[299]王桂莲，白乃彬.多氯酚 QSAR 数值模型比较研究[J]. 环境科学学报，1996，16（2）：190-194.
[300]Kishino Takuo, Kobayshi Kunio. Studies on the mechanism of toxicity of chlorophenolsfound in fish through Quantitative Structure-Activity Relationships [J]. Wat Res,1996,30(2):393-399
[301]Cronin Mark T D, Aptula Aynur O, Duffy Judith C, et al. Comparative assessment ofmethods to develop QSARs for the prediction of toxicity of phenols to Tetrahymena pyriformis[J]. Chemosphere, 2002,49:1201-1221.
[302]Okey Rober W, Stensel H.David. A QSAR-based biodegradablity model-A QSBR [J]. Wat Res,1996, 30(9): 2206-2214.
[303]Adams Craig D, Cozzens Randall A, Kim Byung J. Effects of ozonation on thebiodegradability of substituted phenols [J]. Wat. Res., 1997, 31(10): 2655-2663.
[304]Damborsky Jiri, Schultz T.Wayne. Comparison of the QSAR models for toxicity andbiodegradability of anilines and phenols [J]. Chemosphere, 1997,34(2): 429-446.
[305]王连生，支正良. 分子连接性与分子结构-活性[M]. 北京：中国环境科学出社出版，1992.
[306]Bodor N Harget A，Huang M. Neural network studies. 1.Estimation of the aqueoussolubility of organic compounds [J]. J Am Chem Soc, 1991,113:9480-9483.
[307]Mitchell B E, Jurs PC. Prediction of aqueous solubility of organic compounds frommolecular structure [J]. J Chem Inf Comput Sci, 1988, 38: 489-496.
[308]Huuskonen J. Estimation of aqueous solubility for a diverse set of organic compoundsbased on molecular topology [J]. J Chem Inf Comput Sci, 2000, 400:773-777.
[309]Liu R, So S-S. Development of quantitative structure-property relationship models forearly ADME evaluation in drug discovery. 1.Aqueous solubility [J]. J Chem Inf Comput Sci,2001, 41: 1633 一 1639.
[310]Yalkowsky S H, Pinal R. Estimation of the aqueous solubility of complex organicmolecules [J]. Chemosphere, 1993, 260: 1239-1261.
[311]Huibers PT, Katritzky AR. Correlation of the aqueous solubility of hydrocarbons andhalogenated hydrocarbons with molecular structure [J]. J Chem Inf Comput Sci, 1998, 38:283-292.
[312]Tolls J, van Dijk J, Verbruggen EJM, et al. Aqueous solubility-molecular sizerelationships: A mechanistic case study using C10- to C19-alkanes [J]. J Phys Chem: A, 2002,1060: 2760-2765.
[313]Klopman G, Zhu H. Estimation of the aqueous solubility of organic molecules by thegroup contribution approach [J]. J Chem Inf Comput Sci, 2001, 410: 439-445.
[314]Jain N, Yalkowsky SH. Estimation of the aqueous solubility I: Application to organicnonelectrolytes [J]. J Pharm Sci, 2001, 90: 234-252.
[315]Bodor N, Huang M-J. A new method for the estimation of the aqueous solubility of organiccompounds [J]. J Pharm Sci, 1992, 81:954-960.
[316]Nelson TM, Jurs PC. Prediction of aqueous solubility of organic compounds [J]. J ChemInf Comput Sci, 1994, 34: 601-609.
[317]Nirmalakhandan NN, Speece R.E. Prediction of aqueous solubility of organic chemicalsbased on molecular structure [J]. Environ Sci Technol, 1988, 22 :328-338.
[318]Yalkowsky S.H, Pinal R. Estimation of the aqueous solubility of complex organicmolecules [J]. Chemosphere, 1993, 26: 1239-1261.
[319]王福安，蒋登高，傅举孚. 化工数据导引[M]. 北京：化学工业出版社，1995，156-184.
[320]Wienke G, Gmehling J, Prediction of Octanol-Water Partition Coefficients, HenryCoefficients and Water Solubilitis Using UNIFAC[J]. Toxicol Environ Chem., 1998, 65:57-65.
[321]郭明，刘文杰.有机化合物水溶解性的 QSPR 研究[J]. 计算机与应用化学，2000，17（1）：57-58.
[322]陈艳，冯长君.连接性指数对脂肪醇的 QSAR 研究[J]. 环境化学，2000，19（6）：538-543.
[323]任碧野，许有.一个新的拓扑指数用于有机化合物 QSAR/QSPR 研究[J].化学学报，1999，57（6）：563-571.
[324]张旭，余训民. 一个新的连接性指数对脂肪醇的 QSPR/QSAR 研究[J]. 化学研究，2002，13（1）：49-57.
[325]HAN J，KAMBER M．Data Mining：Concepts and Techniques[M]．Morgan Kaufmaml Publishers，2001．
[326]徐泽柱，王林. 基于粗糙集理论和 BP 神经网络的数据挖掘算法[J]．计算机工程与应用，2004，40(31)：169-172，175．
[327]刘钊，蒋良孝．基于神经网络的数据挖掘研究[J]．计算机工程与应用，2004，40(3)：172-173，190．
[328]Orlinskii D. Influence of environmental contamination with PCBs on human health [J].Environ Geochemistry and Health, 2001，23:317-332.
[329]Sayles G D. DDT, DDD, and DDF Dechlorination by Zero-Valent Iron [J]. EnvironSciTechnol, 1997, 31(12):3448-3454.
[330]Squillace P J. Volatile Organic Copounds in Untreated Ambient Groudwater of the UnitedStates [J]. 1985-1995 Environ Sci Technol 1999,33(3):4176-4187.
[331]李伟民，尹大强，李时银，等. 氯代苯胺对斑马鱼的急性毒性及 3D-QSAR 分析[J]. 环境科学研究，2002，15（2）：6-7.
[332]张玲，张爱茜，韩朔睽，等. 卤代苯及其衍生物的比较分子场研究[J]. 环境化学，2002，21（3）：250-253.
[333]周文富. 卤代苯和苯酚衍生物的位电荷密度能和logKow 与-logLC50的QSAR研究[J]. 有机化学，2002，22（9）：658-662.
[334]Blum D J W, Speece R E. Quantitative Structure-activity Relationships for ChemicalsToxicity to Environmental Bacteria [J]. Ecotoxi Environ Saf, 1991，22:198-224.
[335]Blum D J W, Speece R E. A Database of Chemical Toxicity to Environmental Bacterialand Its Use in Interspecies Comparisons and Correlations [J]. Research Journal WPCF, 1991，63 (3):198-207.
[336]舒元梯.卤代芳烃生物活性的分子拓扑研究[J]. 西南民族大学学报（自然科学版），2006，32（3）：502-207.
[337]袁星, 郎佩珍, 洪晖，等. 应用发光菌测定有机化合物的毒性, 见：王连生等编. 有机污染化学进展[M]. 北京: 化学工业出版社, 1998, 90-92.
[338]Sixt Stefan, Altschuh Joachim, Brüggemann Rainer. Quantitative structure-toxicityrelationships for 80 chlorinated compounds using quantum chemical descriptors[J].Chemosphere, 1995,30(12): 2397-2414.
[339]Becke A D. Density-functional thermochemistry. III. The role of exact exchange [J].J Chem Phys, 1993, 98: 5648-5652.
[340]Stephens P J, Devlin F J, Chabalowski C K, et al. Ab initio Calculation of VibrationalAbsorption and circular Dichroism Spectra Using Density Functional Force Fields [J]. J PhysChem, 1994, 98: 11623-16539.
[341]Zhhao Q, Bao Z. Target recognition based on radial basis functio network. In:Proceeding of international joint conference on neural network V3 [M]. Nagoya:1993,2735-2738.
[342]Hartman E J, Keeler J D, Kowalski J M. Layered neural networks with Guassian hiddenunits as universal approximations [J]. Neural Comput.,1990,2:210-215.
[343]Park J, Sandberg J W. Universal approximation using radial basis functions network[J].Neural Comput, 1991,3:246-257.
[344]Poggio T, Girosi F. Networks for approximation and learning [J]. Proc of the IEEE,1990,78(9):1481-1497.
[345]Schwarzenbach R P, Gschwend P M, Imboden D M. 王连生等译. 环境有机化学[M]. 北京:化学工业出版社, 环境科学与工程出版中心, 2004.
[346]林红卫，杨胜喜，李志良. 卤代苯和苯酚的衍生物的结构表征和毒性预测[J]. 湘潭大学自然科学学报，20059，27（3）：71-76。
[347]舒元梯. 分子拓扑指数与对卤代苯酚生物活性的预测[J]. 西南民族大学学报（自然科学版），2006，32：78-82.
[348]Trohalaki S, Gifford E, Pachter R. Improved QSARs for predictive toxicology ofhalogenated hydrocarbons [J]. Comput Chem, 2000,24:421-427.
[349]Gute Brian D, Balasubramanian K, Geiss K T, et al. Prediction of halocarbon toxicityfrom structure: a hierarchical QSAR approach [J]. Environ Toxicol Phar, 2004,16:121-129.