二维数据复杂体系分辨算法及其在中药研究中的应用
|
摘要
联用色谱仪综合色谱的分离能力与光谱的表征能力，提供了含有更多定性定量信息的二维甚至更高维的数据，是解决复杂体系的首选工具。中药研究已经成为一个世界性的热门课题话题，而中药恰好就是这样一个十分复杂的黑色分析体系。联用仪器与化学计量学的有机结合，为中药现代化研究提供了前所未有的机遇。本论文正是基于这些问题展开研究的。
论文绪论部分综述化学计量学的发展，尤其是二维数据分辨中的几个问题，如色谱流出模式的诊断，重叠峰中组分的纯色谱洗脱曲线和纯光谱解析等以及中药研究中所面临的机遇和挑战。然后阐述我们利用化学计量学方法解决这一问题的途径和看法。
在色谱分析中，流出模式的诊断是至关重要的一步，它是曲线分辨和拟合的基础。论文第二章提出“部分移动窗口因子分析”方法，可用于处理有悖于各流出组分“先进先出”假设的体系。算法能快速、有效，自动地用于分析联用数据。在对原油及中药等复杂实际体系的处理中，该法优越性明显，得到了很好的结果。
中药指纹谱图研究是中药现代化过程中一个非常关键的问题。而许多因素，比如中药的不同产地、制备工艺以及它们的收获季节等等都对中药的化学组分产生影响，并由此而导致它们不同的药理活性。论文第三章提出用一个修正的窗口目标检测因子分析对两种不同产地的大蒜和鱼腥草的地上与地下部分进行比较研究，研究结果清晰地显示了这二种中药由于它们的不同产地或者是用药部位导致的化学成分差异。所得到的结果为它们后续的药理药效及临床研究提供了强有力的依据。
通常情况下，使用联用色谱仪，比如HPLC.DAD和GC一MS
等，所测得的二维数据往往在开始的一段保留时间内存在溶剂峰的
影响。可能的待测组分被完全被包埋在溶剂峰组分中。现有的多元
分辨方法，对于这类问题的处理都变得非常棘手。论文第四章采用
一种同时分辨和背景校正的多元分辨方法来解决这一问题。它充分
利用了原始数据与组分零组分区、临时分辨色谱和光谱之间的关系
来确定背景与溶剂峰模型。然后，我们将溶剂峰也作为背景的一个
部分加以扣除。这样，被溶剂峰所包埋的各分析组分的纯色谱与光
谱信息便可获得。
中药的“道地性”研究是中药研究的一个重要方面。论文第五
章利用色谱联用技术(CC一s)对陕西兴平基地大蒜和湖南普通食用
大蒜中的挥发油成分进行测定，采用质谱相关色谱理论，比较它们
所含挥发油成分的异同。与此同时，基于直观推导式演进特征投影
法(H卫LP)对产生的二维色谱/质谱数据进行解析，根据得到的色谱保
留时间和纯质谱在质谱库中进行相似性检索，实现对组分的部分鉴
定。
Due to its advantages both in chromatographic separation and in spectral identification, hyphenated chromatography provides two-way or higher way data with more qualitative and quantitative information and becomes the chief tool for analyzing complex chemical systems. On the other hand, traditional Chinese medicines (TCMs) research has already became a worldwide topic, but it is such a complex "black system". With the flexible combination of hyphenated instruments and chemometrics, research opportunity to TCMs is unprecedented. This thesis is based on these problems.
In the introduction section, the development of chemometrics in this aspect is summarized, especially in some important problems of two-way data resolution, such as the diagnosis of elution chromatographic patterns, the resolution of pure chromatography and spectra of multi-component in embedded overlap system and the opportunities and challenges in TCMs research. And then, we set forth our ways and ideas to these problems by chemometrics.
In the chromatography analysis, the diagnosis of elution chromatographic patterns is a crucial step, because it is the fundament of curves resolution and fitting later. The method of sectional moving window factor analysis is proposed in chapter 2 for the treatment of the systems beyond of the hypothesis that "the eluting component appears first, will disappear first". In addition, the algorithm is fast, effective and automatic in analyzing hyphenated data. The results are very satisfying and its advantages are also apparent to some real crude oil and TCMs systems analysis.
Fingerprints research in TCMs is a key to the pharmacological, clinical research and quality control etc. But there are many factors, such as different planting places, preparation techniques and their harvest seasons etc. all have effects on the chemical components hi TCMs and lead their different pharmacological activities. In chapter 3, a modified window target-testing factor analysis is employed to study the chemical ingredients in two different places garlic and overground and underground Houttuynia cordata Thunb for comparison. The results obviously show their similarity and difference of chemical components in these two kinds of TCMs. With these powerful results at hand, pharmacological and clinical research of these TCMs is our work in the future.
In common state, the effect of solvent ingredient always exists in the first part of retention time of two-way data, such as HPLC-DAD and GC-MS etc. Possible analytes are completely embedded overlap in solvent peaks. To this kind of intractable problem, the present resolution methods are willing to help but unable to do so. In chapter 4, simultaneous resolution and background correction is proposed to solve this problem. By making full use of the relationship among original data, zero-component region, provisional resolution chromatography and spectra, the background and solvent peak models are determined. Then, the solvent peak is deducted as a part of background. Finally, the embedded overlap components in solvent peak can be easily obtained.
The "phytoequivalence" in TCMs is an important aspect in Chinese medicines research. In chapter 5, the volatile oil in Shanxi Xingping garlic base and Hunan common edible garlic is extracted and comparatively analyzed by mass spectra correlation chromatography. Then, the different chemical ingredients between the two places garlic are
obtained. At the same time, the resolution results by heuristic evolving latent projections (HELP) are searched in mass spectra database and parts components are identified by their mass spectra and retention times.
论文绪论部分综述化学计量学的发展，尤其是二维数据分辨中的几个问题，如色谱流出模式的诊断，重叠峰中组分的纯色谱洗脱曲线和纯光谱解析等以及中药研究中所面临的机遇和挑战。然后阐述我们利用化学计量学方法解决这一问题的途径和看法。
在色谱分析中，流出模式的诊断是至关重要的一步，它是曲线分辨和拟合的基础。论文第二章提出“部分移动窗口因子分析”方法，可用于处理有悖于各流出组分“先进先出”假设的体系。算法能快速、有效，自动地用于分析联用数据。在对原油及中药等复杂实际体系的处理中，该法优越性明显，得到了很好的结果。
中药指纹谱图研究是中药现代化过程中一个非常关键的问题。而许多因素，比如中药的不同产地、制备工艺以及它们的收获季节等等都对中药的化学组分产生影响，并由此而导致它们不同的药理活性。论文第三章提出用一个修正的窗口目标检测因子分析对两种不同产地的大蒜和鱼腥草的地上与地下部分进行比较研究，研究结果清晰地显示了这二种中药由于它们的不同产地或者是用药部位导致的化学成分差异。所得到的结果为它们后续的药理药效及临床研究提供了强有力的依据。
通常情况下，使用联用色谱仪，比如HPLC.DAD和GC一MS
等，所测得的二维数据往往在开始的一段保留时间内存在溶剂峰的
影响。可能的待测组分被完全被包埋在溶剂峰组分中。现有的多元
分辨方法，对于这类问题的处理都变得非常棘手。论文第四章采用
一种同时分辨和背景校正的多元分辨方法来解决这一问题。它充分
利用了原始数据与组分零组分区、临时分辨色谱和光谱之间的关系
来确定背景与溶剂峰模型。然后，我们将溶剂峰也作为背景的一个
部分加以扣除。这样，被溶剂峰所包埋的各分析组分的纯色谱与光
谱信息便可获得。
中药的“道地性”研究是中药研究的一个重要方面。论文第五
章利用色谱联用技术(CC一s)对陕西兴平基地大蒜和湖南普通食用
大蒜中的挥发油成分进行测定，采用质谱相关色谱理论，比较它们
所含挥发油成分的异同。与此同时，基于直观推导式演进特征投影
法(H卫LP)对产生的二维色谱/质谱数据进行解析，根据得到的色谱保
留时间和纯质谱在质谱库中进行相似性检索，实现对组分的部分鉴
定。
Due to its advantages both in chromatographic separation and in spectral identification, hyphenated chromatography provides two-way or higher way data with more qualitative and quantitative information and becomes the chief tool for analyzing complex chemical systems. On the other hand, traditional Chinese medicines (TCMs) research has already became a worldwide topic, but it is such a complex "black system". With the flexible combination of hyphenated instruments and chemometrics, research opportunity to TCMs is unprecedented. This thesis is based on these problems.
In the introduction section, the development of chemometrics in this aspect is summarized, especially in some important problems of two-way data resolution, such as the diagnosis of elution chromatographic patterns, the resolution of pure chromatography and spectra of multi-component in embedded overlap system and the opportunities and challenges in TCMs research. And then, we set forth our ways and ideas to these problems by chemometrics.
In the chromatography analysis, the diagnosis of elution chromatographic patterns is a crucial step, because it is the fundament of curves resolution and fitting later. The method of sectional moving window factor analysis is proposed in chapter 2 for the treatment of the systems beyond of the hypothesis that "the eluting component appears first, will disappear first". In addition, the algorithm is fast, effective and automatic in analyzing hyphenated data. The results are very satisfying and its advantages are also apparent to some real crude oil and TCMs systems analysis.
Fingerprints research in TCMs is a key to the pharmacological, clinical research and quality control etc. But there are many factors, such as different planting places, preparation techniques and their harvest seasons etc. all have effects on the chemical components hi TCMs and lead their different pharmacological activities. In chapter 3, a modified window target-testing factor analysis is employed to study the chemical ingredients in two different places garlic and overground and underground Houttuynia cordata Thunb for comparison. The results obviously show their similarity and difference of chemical components in these two kinds of TCMs. With these powerful results at hand, pharmacological and clinical research of these TCMs is our work in the future.
In common state, the effect of solvent ingredient always exists in the first part of retention time of two-way data, such as HPLC-DAD and GC-MS etc. Possible analytes are completely embedded overlap in solvent peaks. To this kind of intractable problem, the present resolution methods are willing to help but unable to do so. In chapter 4, simultaneous resolution and background correction is proposed to solve this problem. By making full use of the relationship among original data, zero-component region, provisional resolution chromatography and spectra, the background and solvent peak models are determined. Then, the solvent peak is deducted as a part of background. Finally, the embedded overlap components in solvent peak can be easily obtained.
The "phytoequivalence" in TCMs is an important aspect in Chinese medicines research. In chapter 5, the volatile oil in Shanxi Xingping garlic base and Hunan common edible garlic is extracted and comparatively analyzed by mass spectra correlation chromatography. Then, the different chemical ingredients between the two places garlic are
obtained. At the same time, the resolution results by heuristic evolving latent projections (HELP) are searched in mass spectra database and parts components are identified by their mass spectra and retention times.
引文
[1] R.EW. Scott. Techniques and Practice of Chromatography. New York:Marcel Dekker, Inc., 1995
[2] 梁逸曾．白灰黑复杂多组份分析体系及其化学计量学算法．长沙：湖南科学技术出版社，1997
[3] E. Sanchez, B. R. Kowalski. Anal.Chem.,1986, 58, 499-501
[4] R.A. Harshman, UCLA Working Papers in Phonetics, 1970, 16, 1-84
[5] H.L. Wu, M. Shibukawa and K. Ogttma. J. Chemom., 1998, 12, 1-26
[6] J.H. Jiang, H.L. Wu, Y.Li and R.Q. Yu. J. Chemom., 1999, 13, 557-578
[7] R. Bro, C.A. Anderson and H.A.L.Kiers. J. Chemom., 1999, 13,295-312
[8] 蒋健晖．[博士学位论文]．长沙：湖南大学，1999
[9] E J. Gempedine. J. Chem. Inf. Comp. Sci., 1984, 24, 206-220
[10] B.G. M.Vanteginste, T.Esser, T.Bosman, et al. Anal.Chem., 1985, 57, 971-986
[11] M. Maeder, Anal. Chem., 1987,59, 527-530
[12] M. Maeder, A. Zilian, Chemom. Intell. Lab. Syst., 1988,3,205-213
[13] E.R. Malinowski, J. Chemom., 1992,6, 29-40
[14] O.M. Kvalheim, Y. Z.Liang, Anal. Chem., 1992,64, 936-946
[15] Y.Z. Liang, O. M. Kvalheim, H. R. Keller, Anal. Chem., 1992,64, 946-953
[16] R.Manne, H.-L.Shen, Y.-Z. Liang, Chemom. Intell. Lab. Syst., 1999,45, 171-176
[17] H.L.Shen, R.Manne, Q.S. Xu, et al. Chemom. Intell.Lab. Syst., 1999, 45, 335-351
[18] R. Manne. Chemom. Intell.Lab. Syst.,1995, 27, 89-94
[19] H.R. Keller, D. L. Massart, Anal. Chim. Acta, 1991,246, 379-390
[20] J. Toft and O. M. Kvalheim, Chemom. Intell.Lab. Syst., 1993, 19, 65-78
[21]沈海林，梁逸曾，俞汝勤，等．中国科学(B辑)，1998，41，21-34
[22] R.G.Brereton, A.Rahmani, Y.Z. Liang et al. Photochem.Boichem., 1994, 59, 99-112
[23] X.G. He. J Chromatogr. A, 2000,880, 203-232
[24] P.G. Pietta, C. Gardana, Fitoterapia 2002,in press
[25] P. Mauri, P. Pietta.. J. Pharm. Biomed. Anal., 2000,23, 61-68
[26] F.M.Areias, P.Valentao, P.B.Andrade, F.Ferreres, et al. Food Chemistry, 2001, 73, 307-311
[27] B.B. Borse, L. Jagan Mohan Rao, S. Nagalakshrni et al. Food Chemistry, 2002, 79,419-424
[28] J.M. Davis and J.C. Giddings, Anal. Chem., 1983,55, 418-424
[29] Z.D. Wang, M. Fingas, M. Landriault, et al. J. Chromatogr. A,1997,75, 251-265
[30] L.Z. Lin, X. G. He, L.Z. Lian, W.Y. King, J. Elliott, J. Chromatogr. A, 1998,810, 71-79
[31] M.T. Lohnes, R. D. Guy and P. D. Wentzell, Anal. China. Aeta, 1999, 389 95-113
[32] C.D. Brown, P. D. Wentzell, Chemom. Intell. Lab. Syst., 2000,51, 3-7
[33] C.J. Xu, J.H. Jiang, Y.Z. Liang, Analyst, 1999,124,1471-1476
[34] A.C. Andrew, Whitson, M. Maeder, J. Chemom., 2001,15, 475-484
[35] G.H. Golub, F.V. Loan, Matrix Computations. Baltimore:Johns Hopkins University Press, MD, 1983
[36] V.E. Tyler. J. Nat. Prod., 1999, 62, 1589-4592
[37] Y.Z. Liang and O. M. Kvalheim. Anal. Chim. Aeta, 1994,292,5-15
[38] Mitchell T. Lohnes, Robert D. Guy and Peter D. Wentzell. Anal. Chim. Aeta 1999,389,95-113
[39] Christopher D. Brown, Peter D. Wentzell. Chemom. Intell. Lab. Syst., 2000,51,3-7
[40] K.C.Wen, C.Y.Huang and F.L.L. J Chromatogr A, 1993,631,241-250
[41] E. R. Malinowski. Factor analysis in chemistry. 2nd ed. New York:Wiley, 1991
[42] R.A. Johnson, D.W. Wichern. Applied Multivariate Statistical Analysis. New Jersey:Prentice-Hall, Inc., Englewood Cliffs. 1982
[43] F.C. Sanchez, J.Toft, B. van den Bogaert, et al. Anal. Chem., 1996,68, 79-85
[44] W. Winding, J. Guilment. Anal. Chem., 1991,63,1425-1438
[45] B.V. Grande, R. Manne. Chemom. Intell. Lab. Syst., 2000,50, 19-33
[46] R.Tauler, E.J. Casassas. J. Chemom., 1988,3 (Supp. A), 151-162
[47] R.Tauler. Chemom. Intell. Lab. Syst., 1995,30, 133-148
[48] C.J. Xu, Y. Z. Liang, R.Q.Yu. J. Chemom., 1999,10, 63-72
[49] R. Manne. Chemom. Intell. Lab. Syst., 1995,27, 89-94
[50] R.P.W. Scott. Techniques and Practice of Chromatography, New York:Marcel Dekker, Inc. 1995
[51] P.J. Gemperline, J.H. Cho, B. Archer. J Chemom. 1999,13, 153-165
[52] 沈海林．[博士学位论文]．长沙：湖南大学，2001
[53] R. Manne, B.V. Grande. Chemom. Intell. Lab. Syst., 2000,50, 35-46
[54] Y.Z. Liang, O.M. Kvalheim. A. Rahmani, R.G. Brereton, Chemom. Intell. Lab.Syst., 1993,18, 265-279
[55] H.L. Shen, X.N. Li, Y.Z. Liang. Chin. Sci. Bull., 2000,45,587-592
[56] Huang T K. HANDBOOK OF COMPOSITION AND PHARMACOLOGICAL ACTION OF COMMONLY-USED TRADITIONAL CHINESE MEDICINES. Chinese Medical Publishing House. 1994
[57] Wang J H, Wang Z G, Guo H M. HeiLongJiang Medical Journal., 1996,9(6),340-341
[58] B. H. Wang, K. A. Zuzel, K. Rahman et al. Toxicology. 1999,132(2-3),215-225
[59] G. W. Chen, J. G. Chung, C. L. Hsieh et al. Food and Chem. Toxi., 1998, 36(9-10),761-770
[60] Yu X R, Cong YZ.. Chin Tradit Herb Drugs., 1994,25(3),158-161
[61] Gong F, Peng Y G, Cui H et al. Chem. J. Chinese Universities., 1999,20(2),199-203
[2] 梁逸曾．白灰黑复杂多组份分析体系及其化学计量学算法．长沙：湖南科学技术出版社，1997
[3] E. Sanchez, B. R. Kowalski. Anal.Chem.,1986, 58, 499-501
[4] R.A. Harshman, UCLA Working Papers in Phonetics, 1970, 16, 1-84
[5] H.L. Wu, M. Shibukawa and K. Ogttma. J. Chemom., 1998, 12, 1-26
[6] J.H. Jiang, H.L. Wu, Y.Li and R.Q. Yu. J. Chemom., 1999, 13, 557-578
[7] R. Bro, C.A. Anderson and H.A.L.Kiers. J. Chemom., 1999, 13,295-312
[8] 蒋健晖．[博士学位论文]．长沙：湖南大学，1999
[9] E J. Gempedine. J. Chem. Inf. Comp. Sci., 1984, 24, 206-220
[10] B.G. M.Vanteginste, T.Esser, T.Bosman, et al. Anal.Chem., 1985, 57, 971-986
[11] M. Maeder, Anal. Chem., 1987,59, 527-530
[12] M. Maeder, A. Zilian, Chemom. Intell. Lab. Syst., 1988,3,205-213
[13] E.R. Malinowski, J. Chemom., 1992,6, 29-40
[14] O.M. Kvalheim, Y. Z.Liang, Anal. Chem., 1992,64, 936-946
[15] Y.Z. Liang, O. M. Kvalheim, H. R. Keller, Anal. Chem., 1992,64, 946-953
[16] R.Manne, H.-L.Shen, Y.-Z. Liang, Chemom. Intell. Lab. Syst., 1999,45, 171-176
[17] H.L.Shen, R.Manne, Q.S. Xu, et al. Chemom. Intell.Lab. Syst., 1999, 45, 335-351
[18] R. Manne. Chemom. Intell.Lab. Syst.,1995, 27, 89-94
[19] H.R. Keller, D. L. Massart, Anal. Chim. Acta, 1991,246, 379-390
[20] J. Toft and O. M. Kvalheim, Chemom. Intell.Lab. Syst., 1993, 19, 65-78
[21]沈海林，梁逸曾，俞汝勤，等．中国科学(B辑)，1998，41，21-34
[22] R.G.Brereton, A.Rahmani, Y.Z. Liang et al. Photochem.Boichem., 1994, 59, 99-112
[23] X.G. He. J Chromatogr. A, 2000,880, 203-232
[24] P.G. Pietta, C. Gardana, Fitoterapia 2002,in press
[25] P. Mauri, P. Pietta.. J. Pharm. Biomed. Anal., 2000,23, 61-68
[26] F.M.Areias, P.Valentao, P.B.Andrade, F.Ferreres, et al. Food Chemistry, 2001, 73, 307-311
[27] B.B. Borse, L. Jagan Mohan Rao, S. Nagalakshrni et al. Food Chemistry, 2002, 79,419-424
[28] J.M. Davis and J.C. Giddings, Anal. Chem., 1983,55, 418-424
[29] Z.D. Wang, M. Fingas, M. Landriault, et al. J. Chromatogr. A,1997,75, 251-265
[30] L.Z. Lin, X. G. He, L.Z. Lian, W.Y. King, J. Elliott, J. Chromatogr. A, 1998,810, 71-79
[31] M.T. Lohnes, R. D. Guy and P. D. Wentzell, Anal. China. Aeta, 1999, 389 95-113
[32] C.D. Brown, P. D. Wentzell, Chemom. Intell. Lab. Syst., 2000,51, 3-7
[33] C.J. Xu, J.H. Jiang, Y.Z. Liang, Analyst, 1999,124,1471-1476
[34] A.C. Andrew, Whitson, M. Maeder, J. Chemom., 2001,15, 475-484
[35] G.H. Golub, F.V. Loan, Matrix Computations. Baltimore:Johns Hopkins University Press, MD, 1983
[36] V.E. Tyler. J. Nat. Prod., 1999, 62, 1589-4592
[37] Y.Z. Liang and O. M. Kvalheim. Anal. Chim. Aeta, 1994,292,5-15
[38] Mitchell T. Lohnes, Robert D. Guy and Peter D. Wentzell. Anal. Chim. Aeta 1999,389,95-113
[39] Christopher D. Brown, Peter D. Wentzell. Chemom. Intell. Lab. Syst., 2000,51,3-7
[40] K.C.Wen, C.Y.Huang and F.L.L. J Chromatogr A, 1993,631,241-250
[41] E. R. Malinowski. Factor analysis in chemistry. 2nd ed. New York:Wiley, 1991
[42] R.A. Johnson, D.W. Wichern. Applied Multivariate Statistical Analysis. New Jersey:Prentice-Hall, Inc., Englewood Cliffs. 1982
[43] F.C. Sanchez, J.Toft, B. van den Bogaert, et al. Anal. Chem., 1996,68, 79-85
[44] W. Winding, J. Guilment. Anal. Chem., 1991,63,1425-1438
[45] B.V. Grande, R. Manne. Chemom. Intell. Lab. Syst., 2000,50, 19-33
[46] R.Tauler, E.J. Casassas. J. Chemom., 1988,3 (Supp. A), 151-162
[47] R.Tauler. Chemom. Intell. Lab. Syst., 1995,30, 133-148
[48] C.J. Xu, Y. Z. Liang, R.Q.Yu. J. Chemom., 1999,10, 63-72
[49] R. Manne. Chemom. Intell. Lab. Syst., 1995,27, 89-94
[50] R.P.W. Scott. Techniques and Practice of Chromatography, New York:Marcel Dekker, Inc. 1995
[51] P.J. Gemperline, J.H. Cho, B. Archer. J Chemom. 1999,13, 153-165
[52] 沈海林．[博士学位论文]．长沙：湖南大学，2001
[53] R. Manne, B.V. Grande. Chemom. Intell. Lab. Syst., 2000,50, 35-46
[54] Y.Z. Liang, O.M. Kvalheim. A. Rahmani, R.G. Brereton, Chemom. Intell. Lab.Syst., 1993,18, 265-279
[55] H.L. Shen, X.N. Li, Y.Z. Liang. Chin. Sci. Bull., 2000,45,587-592
[56] Huang T K. HANDBOOK OF COMPOSITION AND PHARMACOLOGICAL ACTION OF COMMONLY-USED TRADITIONAL CHINESE MEDICINES. Chinese Medical Publishing House. 1994
[57] Wang J H, Wang Z G, Guo H M. HeiLongJiang Medical Journal., 1996,9(6),340-341
[58] B. H. Wang, K. A. Zuzel, K. Rahman et al. Toxicology. 1999,132(2-3),215-225
[59] G. W. Chen, J. G. Chung, C. L. Hsieh et al. Food and Chem. Toxi., 1998, 36(9-10),761-770
[60] Yu X R, Cong YZ.. Chin Tradit Herb Drugs., 1994,25(3),158-161
[61] Gong F, Peng Y G, Cui H et al. Chem. J. Chinese Universities., 1999,20(2),199-203