基于迭代自组织数据分析算法与蚁群算法建立有机物黏度的QSPR模型

doi:10.3866/PKU.WHXB201403181

摘要/Abstract

摘要：

为了构建310个有机物分子结构与其黏度之间的定量结构-性质关系（QSPR）模型，探讨影响有机物液体黏度的结构因素，首先运用迭代自组织数据分析技术（ISODATA）将样本集初步分类，划分为训练集和测试集，进而应用DRAGON₂.1 软件计算310个有机物分子的分子结构描述符，以蚁群算法（ACO）筛选分子描述符，得到5 个参数，随后分别采用多元线性回归法（MLR）和支持向量机法（SVM）建立ACO-MLR模型和ACOSVM模型. 结果表明，非线性ACO-SVM 模型（相关系数R_train²=0.9013，R_test²=0.9026）的性能优于线性ACOMLR模型（R_train²=0.7680，R_test²=0.8725）. ACO-MLR模型和ACO-SVM模型对测试集所得预测值与实验值的相关系数分别为0.934和0.950，预测效果令人满意. 本文应用Williams 图对模型的应用域进行了一定的研究，所建立的模型为工程上提供了一种根据分子结构预测有机物黏度的有效方法.

关键词: 黏度, ISODATA, 蚁群算法, 多元线性回归, 支持向量机

Abstract:

The aim of this study was to construct a quantitative structure-property relationship model to identify relationships between the molecular structures and viscosities of 310 compounds, as well as specific structural factors that could affect the viscosities of the compounds. Using an iterative self-organizing data analysis technique, the sample set was preliminarily classified into two sets, including a training set and a test set. The molecular structure descriptors of 310 compounds were calculated using version 2.1 of the Dragon software and subsequently sifted using an ant colony algorithm (ACO), which resulted in the selection of five parameters. Multiple linear regression (MLR) and the support vector machine (SVM) techniques were then used to establish ACO-MLR and ACO-SVMmodels, respectively. The results showed that the performance of the non-linear ACOSVMmodel (correlation coefficient R_train²=0.9013, R_test²=0.9026) was superior to the linearACO-MLRmodel (R_train²=0.7680, R_test²=0.8725). The correlation coefficients between the experimental and predicted values of the ACOMLR and ACO-SVM models for the test set were 0.934 and 0.950, respectively. The predictive properties of the two models were therefore determined to be satisfying. The application domain of the model was also studied using a Williams graph, which demonstrated that the models established in this study provide effective methods for predicting the viscosities of specific compounds based on their molecular structure.

Key words: Viscosity, ISODATA, Ant colony algorithm, Multiple linear regression, Support vector machine

点击下载补充材料ZIP格式文件

时静洁, 陈利平, 陈网桦. 基于迭代自组织数据分析算法与蚁群算法建立有机物黏度的QSPR模型[J]. 物理化学学报, 2014, 30(5), 803-810. doi: 10.3866/PKU.WHXB201403181

SHI Jing-Jie, CHEN Li-Ping, CHEN Wang-Hua. QSPR Models of Compound Viscosity Based on Iterative Self-Organizing Data Analysis Technique and Ant Colony Algorithm[J]. Acta Phys. -Chim. Sin. 2014, 30(5), 803-810. doi: 10.3866/PKU.WHXB201403181

链接本文: https://www.whxb.pku.edu.cn/CN/10.3866/PKU.WHXB201403181

https://www.whxb.pku.edu.cn/CN/Y2014/V30/I5/803

参考文献

(1) Baronavski, A. P.; McDonald, J. R. J. Chem. Phys. 1977, 66, 3300. doi: 10.1063/1.434306
(2) McKellar, A. The Journal of the Royal Astronomical Society of Canada 1960, 54, 97.
(3) Brault, J.W.; Delbouille, L.; Grevesse, N.; Roland, G.; Sauval, A. J.; Testerman, L. Astron. Astrophys.1982, 108, 201.
(4) A'Hearn, M. F.; Millis, R. C.; Schleicher, D. O.; Osip, D. J.; Birch, P. V. Icarus 1995, 118, 223. doi: 10.1006/icar.1995.1190
(5) Bakker, E. J.; van-Dishoeck, E. F.; Waters, L. B. F. M.; Schoenmaker, T. Astron. Astrophys. 1997, 323, 469.
(6) Cecchi-Pestellini, C.; Dalgarno, A. Mon. Not. Roy. Astron. Soc. 2002, 331, L31.
(7) Oka, T.; Thorburn, J. A.; McCall, B. J.; Friedman, S. D.; Hobbs, L. M.; Sonnentrucker, P.; Welty, D. E.; York, D. G. Astrophys. J. 2003, 582, 823. doi: 10.1086/apj.2003.582.issue-2
(8) Rennick, C. J.; Smith, J. A.; Ashfold, M. N. R.; Orr-Ewing, A. J. Chem. Phys. Lett. 2004, 383, 518. doi: 10.1016/j.cplett.2003.11.089
(9) Gordillo-Vazquez, F. J.; Albella, J. M. J. Appl. Phys. 2003, 94, 6085. doi: 10.1063/1.1617362
(10) Chernicharo, J.; Guelin, M.; Hein, H.; Kahane, C. Astron. Astrophys. 1987, 181, L9.
(11) Saito, S.; Kawaguchi, K.; Yamamoto, S.; Ohishi, M.; Suzuki, H.; Kaifu, N. Astrophys. J. 1987, 317, L115.
(12) Millar, T. J.; Herbst, E. Astron. Astrophys. 1990, 231, 466.
(13) Petrie, S. Mon. Not. Roy. Astron. Soc. 1996, 281, 666. doi: 10.1093/mnras/281.2.666
(14) Velusamy, T.; Kuiper, T. B. H.; Langer, W. D. Astrophys. J. 1995, 451, L75.
(15) Markwick, A. J.; Millar, T. J.; Charnley, S. B. Astrophys. J. 2000, 535, 256. doi: 10.1086/apj.2000.535.issue-1
(16) Hausmann, M.; Homann, B. K. H. Bunsen-Ges. Phys. Chem. 1990, 94, 1308.
(17) Harris, S. J.; Shin, H. S.; Goodwin, D. G. Appl. Phys. Lett. 1995, 66, 891. doi: 10.1063/1.113422
(18) Jones, J. M.; Harding, A.W.; Brown, S. D.; Thomas, K. M. Carbon 1995, 33, 833. doi: 10.1016/0008-6223(94)00183-Z
(19) Schofield, K. Combust. Flame 2001, 124, 137. doi: 10.1016/S0010-2180(00)00189-9
(20) Donnelly, V. M.; Pasternack, L. Chem. Phys. 1979, 39, 427. doi: 10.1016/0301-0104(79)80160-3
(21) Pasternack, L.; McDonald, J. R. Chem. Phys. 1979, 43, 173. doi: 10.1016/0301-0104(79)85185-X
(22) Reisler, H.; Mangir, M.; Wittig, C. J. Chem. Phys. 1979, 71, 2109. doi: 10.1063/1.438583
(23) Reisler, H.; Mangir, M. S.; Wittig, C. J. Chem. Phys. 1980, 73, 2280. doi: 10.1063/1.440377
(24) Pasternack, L.; Baronavski, A. P.; McDonald, J. R. J. Chem. Phys. 1980, 73, 3508. doi: 10.1063/1.440508
(25) Pasternack, L.; Pitts, W. M.; McDonald, J. R. Chem. Phys. 1981, 57, 19. doi: 10.1016/0301-0104(81)80017-1
(26) Pitts, W. M.; Pasternack, L.; McDonald, J. R. Chem. Phys. 1982, 68, 417. doi: 10.1016/0301-0104(82)87050-X
(27) Kruse, T.; Roth, P. Int. J. Chem. Kin. 1999, 31, 11.
(28) Becker, K. H.; Donner, B.; Dinis, C. M. F.; Geiger, H.; Schmidt, F.; Wiesen, P. Z. Phys. Chem. 2000, 214, 503.
(29) Fontijn, A.; Fernandez, A.; Ristanovic, A.; Randall, M. Y.; Jankowiak, J. T. J. Phys. Chem. A 2001, 105, 3182.
(30) Ristanovic, A.; Fernandez, A.; Fontijn, A. J. Phys. Chem. A 2002, 106, 8291. doi: 10.1021/jp014553n
(31) Huang, C. S.; Zhu, Z. Q.; Xin, Y.; Pei, L. S.; Chen, C. X.; Chen, Y. J. Chem. Phys. 2004, 120, 2225. doi: 10.1063/1.1636692
(32) Huang, C. S.; Zhao, D. F.; Pei, L. S.; Chen, C. X.; Chen, Y. Chem. Phys. Lett. 2004, 389, 230. doi: 10.1016/j.cplett.2004.03.080
(33) Huang, C. S.; Li, Z. X.; Zhao, D. F.; Xin, Y.; Pei, L. S.; Chen, C. X.; Chen, Y. Chin. Sci. Bull. 2004, 49, 438.
(34) Huang, C. S.; Zhu, Z. Q.; Wang, H. L.; Pei, L. S.; Chen, Y. J. Phys. Chem. A 2005, 109, 3921.
(35) Daugey, N.; Caubet, P.; Bergeat, A.; Costes, M.; Hickson, K. M. Phys. Chem. Chem. Phys. 2008, 10, 729.
(36) Páramo, A.; Canosa, A.; Le Picard, S. D.; Sims, I. R. J. Phys. Chem. A 2006, 110, 3121. doi: 10.1021/jp0544787
(37) Páramo, A.; Canosa, A.; Le Picard, S. D.; Sims, I. R. J. Phys. Chem. A 2008, 112, 9591. doi: 10.1021/jp8025336
(38) Hu, R. Z.; Zhang, Q.; Chen, Y. J. Chem. Phys. 2010, 132, 164312. doi: 10.1063/1.3400070
(39) Hu, R. Z.; Zhang, Q.; Chen, Y. J. Chem. Phys. 2010, 133, 114306. doi: 10.1063/1.3480395
(40) Hu, R. Z.; Zhang, Q.; Chen, Y. Acta Phys. -Chim. Sin. 2010, 26, 2619. [胡仁志, 张群, 陈旸. 物理化学学报, 2010, 26, 2619.] doi: 10.3866/PKU.WHXB20100938
(41) Matsugi, A.; Kohsuke, S.; Miyoshi, A. J. Phys. Chem. A 2010, 114, 4580. doi: 10.1021/jp1012494
(42) Li, N.; Huo, R. P.; Zhang, X.; Huang, X. R.; Li, J. L.; Sun, C. C. Chem. Phys. Lett. 2011, 503, 210. doi: 10.1016/j.cplett.2011.01.029
(43) Kolbanovskii, Y. A.; Borisov, Y. A. Mendeleev Commun. 2011, 21, 305. doi: 10.1016/j.mencom.2011.11.003
(44) Huo, R. P.; Zhang, X.; Huang, X. R.; Li, J. L.; Sun, C. C. Mol. Phys. 2012, 110, 2205. doi: 10.1080/00268976.2012.668969
(45) Dangi, B. B.; Maity, S.; Kaiser, R. I.; Mebel, A. M. J. Phys. Chem. A 2013, 117, 11783. doi: 10.1021/jp402700j
(46) Hu, R. Z.; Xie, P. H.; Zhang, Q.; Si, F. Q.; Chen, Y. Acta Phys. -Chim. Sin. 2013, 29, 683. [胡仁志, 谢品华, 张群, 司福祺, 陈旸. 物理化学学报, 2013, 29, 683.] doi: 10.3866/PKU.WHXB201302046
(47) Huo, R. P.; Zhang, X.; Huang, X. R.; Li, J. L.; Sun, C. C. J. Mol. Model. 2013, 19, 1009. doi: 10.1007/s00894-012-1616-8
(48) Yang, X. L.; Xu, G.; Chen, Y.; Chen, C. X. Acta Chim. Sin. 2006, 64, 47. [杨学良, 徐刚, 陈旸, 陈从香. 化学学报, 2006, 64, 47.]
(49) Herzberg, G. Molecular Spectra and Molecular Structure I. Spectra of Diatomic Molecules. Van Nostrand: Princeton, 1950.
(50) Wang, J. H.; Han, K. L.; He, G. Z.; Li, Z. J. Chem. Phys. Lett. 2003, 368, 139. doi: 10.1016/S0009-2614(02)01810-9
(51) Leu, M. T.; Smith, R. H. J. Phys. Chem. 1982, 86, 73. doi: 10.1021/j100390a015
(52) Lin, Y. L.; Wang, N. S.; Lee, Y. P. Int. J. Chem. Kinet. 1985, 17, 1201.
(53) Lafage, C.; Pauwels, J. F.; Carlier, M.; Devolder, P. J. Chem. Soc. Faraday Trans. 2 1987, 83, 731. doi: 10.1039/f29878300731
(54) Wilson, C.; Hirst, D. M. J. Chem. Soc. Faraday Trans. 1994, 90, 3051. doi: 10.1039/ft9949003051
(55) Swinnen, S.; Elsamra, R. M. I.; Nguyen, V. S.; Peeters, J.; Carl, S. A.; Nguyen, M. T. Chem. Phys. Lett. 2011, 513, 201. doi: 10.1016/j.cplett.2011.07.098

[1]	安保林,杨富方,杨震,段远源,于养信. 在确定的热力学状态下测量气体导热系数与黏度[J]. 物理化学学报, 2016, 32(5): 1129 -1133 .
[2]	何冰,罗勇,李秉轲,薛英,余洛汀,邱小龙,杨登贵. 基于分子描述符和机器学习方法预测和虚拟筛选乳腺癌靶向蛋白HEC1抑制剂[J]. 物理化学学报, 2015, 31(9): 1795 -1802 .
[3]	李秉轲, 丛湧, 田之悦, 薛英. 基于分子描述符和机器学习方法预测和虚拟筛选MMP-13对MMP-1的选择性抑制剂[J]. 物理化学学报, 2014, 30(1): 171 -182 .
[4]	丛湧, 薛英. 基于机器学习方法的丙型肝炎病毒聚合酶NS5B非核苷抑制剂的定量构效关系研究[J]. 物理化学学报, 2013, 29(08): 1639 -1647 .
[5]	王志明, 韩娜, 袁哲明, 伍朝华. 基于岭回归和SVM的高维特征选择与肽QSAR建模[J]. 物理化学学报, 2013, 29(03): 498 -507 .
[6]	吕巍, 薛英, 孟庆伟. 基于机器学习方法的H1N1神经氨酸苷酶抑制剂的分类预测[J]. 物理化学学报, 2013, 29(01): 217 -223 .
[7]	时静洁, 陈利平, 陈网桦, 石宁, 杨惠, 徐伟. 基于启发式方法和支持向量机方法预测有机物的热导率[J]. 物理化学学报, 2012, 28(12): 2790 -2796 .
[8]	王秀军, 龙汨. 统计方法校正O3LYP方法计算的生成热[J]. 物理化学学报, 2012, 28(11): 2581 -2588 .
[9]	代志军, 周玮, 袁哲明, . 基于支持向量机的高维特征非线性快速筛选与肽QSAR建模[J]. 物理化学学报, 2011, 27(07): 1654 -1660 .
[10]	吕巍, 薛英. 基于机器学习方法的丙型肝炎病毒非结构蛋白5B聚合酶抑制剂活性预测[J]. 物理化学学报, 2011, 27(06): 1407 -1416 .
[11]	王丹, 张燕玲, 乔延江. 支持向量机与KStar模型预测细胞色素P450酶催化的氧脱烃反应[J]. 物理化学学报, 2011, 27(02): 343 -351 .
[12]	潘善飞,胡桂香,吕杨,邹建卫, 俞庆森. 离子液体中有机物溶解度的QSPR模型分析[J]. 物理化学学报, 2010, 26(09): 2494 -2502 .
[13]	吕巍, 薛英. 基于机器学习方法的激素敏感脂肪酶抑制剂活性预测[J]. 物理化学学报, 2010, 26(02): 471 -477 .
[14]	李平, 谈宁馨, 饶含兵, 李泽荣, 陈宇综. 基于支持向量机方法的HERG钾离子通道抑制剂分类模型[J]. 物理化学学报, 2009, 25(08): 1581 -1586 .
[15]	黄钦;庄艳;乔学斌;徐筱杰. 用支持向量机建立中药有效成分聚集体的预测模型[J]. 物理化学学报, 2007, 23(08): 1141 -1144 .