Research articles
 

By Mr. Yu He , Mr. Xiao-Ru Li
Corresponding Author Mr. Xiao-Ru Li
College of Chemistry and Chemical Engineering, Central South University, Changsha, - China
Submitting Author Mr. Yu He
Other Authors Mr. Yu He
chemistry institute, - China

CHINESE MEDICINE

Heuristic Evolving Latent Projections (HELP), Volatile Constituents, GC-MS, Temperature-Programmed Retention Indices (PTRIs)

He Y, Li X. Analysis of Volatile Omponents In Rhizome Zingibers, Zingiber Officinale Roscoe And Ginger Peel By Gas Chromatography-Mass Spectrometry And Chemometric Resolution. WebmedCentral CHINESE MEDICINE 2010;1(9):WMC00662
doi: 10.9754/journal.wmc.2010.00662
No
Submitted on: 20 Sep 2010 08:21:11 AM GMT
Published on: 20 Sep 2010 04:38:09 PM GMT

Abstract


The similarities and differences of essential oil components in rhizome zingibers(RZ)?zingiber officinale roscoe(ZOR) and ginger pee(GP) were investigated by GC-MS combined with a chemometric method named heuristic evolving latent projections(HELP).And temperature-programmed retention indices(PTRIs) was used together with mass spectra for identification of the essential oil components. For essential oils of RZ, ZOR and GP, 85, 81 and 80 volatile components were determined representing 81.43%, 86.38% and 84.79% of the total relative content, respectively. Also, the essential oils significantly differed both qualitatively and quantitatively. There were total 58 common compounds existing in RZ and ZOR, 63 common components between RZ and GP, 60 common components between ZOR and GP, and 52 common components existing among each of the three systems. The results obtained may be helpful to the further study of pharmacological activity for their potential utilization as therapeutical agents.

Introduction


1. Introduction
Tradition Chinese medicines have gained more and more popularity in our modern society because of their pharmacological activity and low toxicity, which makes them widely used in food and medicine regions. Low toxicity, rare complications and nature are their advantages. Only after further research can we utilize full of their values, then the development of hyphenated chromatographic device combined with Chemometric methods give us the opportunity to acknowledge their essential components and exactly qualify and quantify our targets.
The herbal materials rhizome zingibers(RZ) zingiber officinale roscoe(ZOR) and ginger peel(GP) are widely used in the region of food industry, which could improve appetite and aid digestion combined with the functions of Refreshing and Antibacterial anti-inflammatory. In traditional Chinese medicine research, they also play an important role, with the pharmacological activities of relieving rheumatism and cold, keeping warm and antiemetic. The RZ, ZOR and GP are contained in the medicine against Cold, Stomach pain and Diarrhea etc.
The RZ comes from the dry rhizome of ginger, ZOR is the fresh rhizome, and GP is the scarfskin of fresh rhizome of ginger. Previous investigations and researches have showed that volatile components are their main medicinal elements which contains Zingiberene, Gingerol and Citral and so on.
In this paper, the volatile components extracted were detected with gas chromatography-mass spectrometry (GC-MS). However, the GC-MS data from volatile oil involves a great number of overlapped and even embedded peaks. These overlapped and embedded peaks may bring about many difficulties when carrying out quantitative and qualitative analysis correctly. In order to resolve the problems, chemometric methods as a very useful assistant tool, use comprehensive chromatographic and spectral information to make it possible to resolve one complex system clearly and accurately. The resolution of the two-dimensional data by chemometric method heuristic evolving latent projections (HELP) is applied in this study. Then, qualitative identification of these constituents was performed using temperature-programmed retention indices (PTRIs) and mass spectra. Finally, every component was quantized with the total volume integral method. Through the procedures above, we could compare the same and distinction of volatile components in RZ, ZOR and GP and supply foundation for their further application in regions of food and medicine.


Methods


2. Theory
A two-dimensional data Am×n produced by GC–MS can be represented as follows:

Am×n= CSt+E

=∑CiSit+E (i=1, 2… N)

Where Am×n denotes an absorbance matrix expressing N components of m chromatographic scan points at n atom mass units or wavelength points. C is the pure chromatographic matrix and S is the pure mass spectral matrix. E represents the noise. t is the transform of matrix S. The unique resolution of a two-dimensional data into chromatograms and spectra of the pure chemical constituents is carried out with local full rank analysis in the HELP method. Only concise theoretical explanation is showed here for the sake of brevity, while detail description could be found in Refs. [1–3].
1. Confirm the background and correct a drifting base line.
2. Determine the number of components, the selective region and zero-component region of each component by the use of the evolving latent projective graph and rankmap on the basis of the eigenstructure tracking analysis.
3. With the help of the selective information and zero-component region, conduct a unique resolution of two-dimensional data into pure chromatographic profiles and mass spectra by means of local full rank analysis.
4. Verify the reliability of the resolved result.
After the pure chromatogram and spectrum of the ith component have been resolved, this component can be determined qualitatively by comprehensive use of the chromatographic retention time and mass spectrum. Next, the term in Eq. (1) is taken as the overall volume integration value. Similar to the general chromatographic quantitative method with peak area, is directly proportional to the mass of the ith component and so it is quantified.
3. Experimental
3.1 Materials
All single herbals were purchased from the market and identified by Institute of Materia Medica, Hunan Academy of Traditional Chinese Medicine and Materia Medica (Changsha, Hunan, China). n-Alkane standard solutions of C8–C20 (mixture no. 04070) and C21–C40 (mixture no. 04071) were purchased from Fluka Chemika (Buchs,Switzerland).
3.2 Extraction of volatile oil
Extraction of volatile oil of the herbal: 50g of each dried single herbal, RZ, ZOR and GP, were weighted exactly and mixed with 350ml distilled water and then processed according to the standard extracting method for the volatile oil described in Chinese Pharmacopoeia (2005 version).
3.3 Instruments
The GC-MS instrument was QP2010 with a QP-5000 mass spectrometer (both from Shimadzu) employed in this study.
3.4 Detection of volatile oil
In the gas chromatographic system, a DB-1 capillary column (30m×0.25mm I.D.) was applied. The original column temperature was maintained at 50 degrees for 3min, then programmed from 50 degrees to 150 degrees at the rate 5 degrees/min, and from 150 degrees to 190 degrees at the rate 2 degrees/min, and from 190 degrees to 250 degrees at the rate 10 degrees/min, finally maintained for 3min. The inlet temperature was kept at 270 degrees and interface temperature at 250 degrees. The carrier gas was Helium with a constant flow-rate of 1.0ml/min. To the experimental conditions of the mass spectrometer, the electron impact (EI+) mass spectra was recorded at 70 eV. Splitting ratio was 10:1. Scan at 5scan/s from m/z 35 to 500 amu. The ionization source temperature was set at 200 degrees.
3.5 Retention indices
Van den Dool and Kratz[19] proposed a quasi-linear equation for temperature- programmed retention indices as follows:
ITX = 100n+ 100 [ ( tRx - tRn ) / ( tR ( n + 1) -tRn ) ]
Where ITX is the temperature-programmed retention index of the interest, tRn, tR (n + 1), tRx are the retention time in minute of the two standard n-alkanes containing n and n+1 carbons and the interest, respectively. This equation was used to calculate retention indices in the research work. It could improve distinguishing components, especially those very similar.
3.6 Data analysis
All data analysis was performed on Celeron®2.66GHz(Intel) personal computer, and all programs of chemometric resolution methods were coded in MATLAB 6.5 for windows. The library searches and spectral matching of the resolved pure components were conducted on the National Institute of Standard and Technology (NIST) 107 MS database containing 107886 compounds. The exactly qualitative results were obtained with the help of PTRIs.


Results


4.Results and discussion
4.1 Resolution of overlapped peaks by HELP
The Fig.1 shows the real total ionic chromatogram (TIC) of the volatile oil of RZ (a) ZOR (b) and GP (c).
We can see each of the TICs is a very complex analytical system. Although many chromatographic peaks are separated, here still exist some overlapped peaks. Due to these, if you directly search with the MS database, definitely fail you will get, because the mass spectrum of mixtures measured can never get good matching index with that of a pure component in the NIST MS database. Furthermore, for the component with low content, it is also very difficult to be identified correctly with the NIST MS database, since a two-dimensional data obtained by mass spectral measurement unavoidably contains peaks associated with column background and residual gases. Without background correction, both the resolution of the overlapped peaks and the identification of the components with low content are impossible. For commercial GC–MS systems, background subtraction is usually performed as follows. First, a scan point, which only contains the background mass spectrum, is subjectively found. Next, the intensities of the same inter mass numbers appearing in the target and background spectra are subtracted, and so the practical target mass spectrum is obtained. Obviously, the practical target mass spectrum strongly depends on the selection of the background point. If this selection is wrong, different target mass spectra may be obtained. The genuine mass spectrum to be searched is surely confused with the subtracted spectrum. As for the HELP method, the local rank analysis of the zero-component regions, which contain no components eluting, before elution of the first chemical component starts and after the last chemical constituent has eluted, can together provide sufficient information for accurately correcting a drifting baseline [1–3]. Hence, a much better background subtraction could be obtained. After background subtraction, the resolution of the overlapping peaks becomes possible. To illustrate how to extract the pure mass spectra efficiently by HELP resolution, the peak clusters marked A,B and C are chosen as examples and then processed by HELP.
Fig.2 shows the TIC curve of the peak cluster A which looks like a pure peak only containing one constituent. Likewise, only one compound named beta.-Phellandrene (C10H16) can be searched in the NIST MS library. Therefore, peak cluster A is generally regarded as a one-compound system in a classic analytical way. However, if HELP resolution method is applied to the two-dimensional data matrix of peak cluster A, three distinct components named Eucalyptol (C10H18O), beta.-Phellandrene (C10H16) and D-Limonene(C10H16)can be resolved even they have quite similar mass spectra. Detail procedures would be introduced in the following part.
Peak purity can be identified through a fixed size moving window evolving factor analysis (FSWMEFA) [7] or so-called eigenstructure tracking analysis [3]. In the fixed size window method (FSWM) plot, the noise level is characterized by eigenvalue curves which have similar numerical values and appears together at the bottom. Eigenalue curves higher than the noise level represent the appearance of new components. If a studied system contains only one species, there is only one eigenalue curve higher than the noise level in its FSWM plot. From the FSWM plot of peak cluster A shown in Fig. 3, there are four eigenvalue curves higher than the noise level within the peak region. We can conclude that the peak A may not be a pure one. Furthermore, after a special pretreatment described in Ref. [8] is conducted, the new result are shown in Fig.4 from which one could conclude that the region of i is the pure area of the first component, the region of ii is the overlapping region of the first and second components, the same to the region iii overlapped by the second and third components, lastly the region of IV is the pure region of the third component.
The stepwise eluting information of chemical components in peak cluster A can be further confirmed by evolving latent projection graph (ELPG) [1–3]. This technique is based on the use of ordered nature of hyphenated data and that the selective regions appear as straight-line segments in bivariate score plots of principal component analysis. Thus, the ELPG is essentially a principal component projective curve from chromatographic or spectral spaces. There are a few advantages to use the ELPG: (i) In a bivariate score plot a straight line segment pointing to the origin suggests selective information in the retention time direction, while in a bivariate loadings plot a straight line segment pointing to the origin suggests selective information in the spectral direction. The concept of "straight line" here is, of course, under sense of least squares; (ii) The evolving information of the appearance and disappearance of the chemical components in retention time direction can be also provided in ELPG. In the ELPG from the chromatographic space, the straight line section represents the pure selective region of one component while the curving section denotes the overlapping region of at least two constituents; (iii) Information enabling the detection of shifts of the chromatographic base line and instrumental background is also provided in ELPG. If there is an offset in chromatography the points cannot concentrate on the origin in the plot even if one includes the zero-component regions in data; (iv) ELPG is also a very good diagnostic tool to identifying the embedded peaks in chromatogram. This information is very important for resolution of concentration profiles of embedded peaks. The ELPG likes a data scope to see the insight of data structure of the two-way data. Fig. 5 shows the ELPG of peak cluster A. From this plot, one can see that this peak cluster is a three-component system. The marks, say 1, 1+2, 2+3 and 3 in the plot, indicate respectively the pure region of the first component, the overlapping region of the first and second components and the overlapping region of the second and third components and the pure region of the third component in the chromatographic direction. This is consistent with the results obtained from the FSWM plot after subtracting the heteroscedastic noise.
From the discussion above, the chromatographic eluting order can be determined and so the number of components in the system, the selective regions and zero-concentration regions of all the constituents. Because of all the acknowledged information, the two-dimensional data matrix can be uniquely resolved into pure chromatographic profiles and mass spectra of all components.
The qualitative of the chemical composition of the sample determination can be directly performed by means of similarity searches in the NIST mass library now, as the pure chromatographic curve and mass spectrum of each component have been resolved. The result shows that these three components in peak cluster A are Eucalyptol(C10H18O), beta.-Phellandrene(C10H16) and D-Limonene(C10H16). Their corresponding chromatographic curves are shown in Fig. 6, and the resolved mass spectra together with the standard spectrum of each component from the NIST MS library are also given in Figs. 7, 8 and 9.
From the resolved pure chromatographic peaks, it is obvious that the first and the third component both have a relatively small quantity in the whole system, and the three peaks overlap seriously, additionally the mass spectrums of the second and the third component is a little similar. Because of these, if you search directly in the mass-library, only the component containing more would be found out, or a wrong result appear. With the help of chemometric resolution method called heuristic evolving latent projections (HELP), the reliable and accurate results could be obtained.
Likewise, Fig.10 shows the TIC curve of peak cluster B, appears to be a mixed system of only two constituents. Only two components named Bicyclo[2.2.1]heptane-2,5-diol,1,7,7-trimethyl-,(2-endo,5-exo)-(C10H18O2) and 2-Cyclohexen-1-ol, 1-methyl-4-(1-methylethyl)-, trans-(C10H18O)can be directly matched in the NIST MS database. However, four isomeric components which are 1-Methyl-3-(1-methycyclopropyl)cy clopentene(C10H16), Bicyclo[2.2.1]heptan-2-ol, 1,7,7-trimethyl-,exo-(C10H18O), 2-Cyclohexen-1-ol, 1-methyl-4-(1-methylethyl)-, trans-(C10H18O), 6-Octen-1-ol, 7-methyl-3-methylene-(C10H18O)can be resolved by means of the HELP resolution method with the procedure described above.
After the background is removed, the ELPG is plotted in Fig.11, which suggests that the peak cluster B is much complex than the peak cluster A. The marks, say 1, 2, 4 in the plot, indicate respectively the pure region of the first, the second and the third component, and the marks 2+3,3+4 in the plot, indicate the overlapping region of component 2 and 3 and the overlapping region of component 3 and 4, respectively.
The FSWM plots after correction of the heteroscedastic noises for peak cluster B are shown in Figs. 12. The regions marks by i,ii,iii,IV,V and VI, indicate the region of the pure component 1, the overlapping region of components 1 and 2, the region of the pure component 2, the overlapping region of components 2 and 3, the overlapping region of components 3 and 4 and the region of the pure component 4, respectively.
The FSWM plot obtained after correcting the heteroscedastic noise clearly shows that there are four components in peak cluster B. After the eluting regions of all components are determined the unique resolution into chromatograms and mass spectra can be then conducted on the two-dimensional data. Figs.13—17 show the resolved results. These figures definitely support the presence of four isomeric components in peak cluster B.
And the TIC curve of peak cluster C is show in Fig.18, relatively more complex systems containing 4 constituents named Bicyclo[2.2.1]heptan-2-ol, 1,3,3-trimethyl-(C10H18O), cis-2-Pinanol(C10H18O), Bicyclo[2.2.1]heptane,2-methoxy-1,7,7-trimethyl-(C11H20O) and 6-Octen-1-ol, 7-methyl-3-methylene-(C10H18O) could obtained if you search in the NIST database directly. However, six isomeric components which are Bicyclo[2.2.1]heptan-2-ol, 1,3,3-trimethyl-(C10H18O), Isopulegol(C10H18O), cis-2-Pinanol(C10H18O), Bicyclo[2.2.1]heptan-2-ol, 1,7,7-trimethyl-, exo-(C10H18O), 2-Cyclohexen-1-ol,1-methyl-4-(1-methylethyl)-trans-(C10H18O)and 6-Octen-1-ol, 7-methyl-3-methylene-(C10H18O)can be resolved by means of the HELP resolution method with the procedure described before.
Just like the sample shown above, the FSWM plots after correction of the heteroscedastic noises for peak cluster C are shown in Figs. 19. The regions marks by i, ii, iii, IV, V, VI, ? and indicate the region of the pure component 1, the overlapping region of components 1 and 2, the region of the pure component 2, the overlapping region of components 2 and 3, the overlapping region of components 3 and 4, the overlapping region of components 4 and 5, the overlapping region of components 5 and 6 and the region of the pure component 6, respectively.
After the background is removed, the ELPG is plotted in Fig.20, which suggests that the peak cluster C is much complex than the formers. The marks, say 1, 2, 3, 6 in the plot, indicate respectively the pure region of the first, the second, the third and the sixth component, and the marks 3+4, 4+5, 5+6 in the plot, indicate the overlapping region of component 3 and 4, the overlapping region of component 4 and 5 and the overlapping region of component 5 and 6, respectively.
Figs.21—27 show the resolved results about the unique resolution into chromatograms and mass spectra.
4.2 Qualitative analysis
Other peaks in the sample at other chromatographic scan point were also qualified in the same way as described above. The tentatively qualitative results of constituents from RZ, ZOR and GP are displayed in Table 1.
4.3 Quantitative analysis
Using the total volume integral method, the quantitative results were calculated and also shown in Table 1. In total, 85, 81 and 80 volatile components in volatile oil of RZ, ZOR and GP were respectively determined qualitatively and quantitatively, accounting for 81.43%, 86.38% and 84.79% total contents of volatile oil of RZ, ZOR and GP respectively.
4.4 Comparison of volatile components among RZ, ZOR and GP
From the Table 1, there are total 58 common compounds existing in RZ and ZOR, 63 common components between RZ and GP, 60 common components between ZOR and GP, and 52 common components existing among each of the three systems.
It is obvious that the volatile oils in RZ, ZOR and GP are nearly the same, only a little different. Because they come from a same herb, chemical properties and active effects are very similar. However, the quantities of common volatile oils in different systems are different, it is useful to accurately qualify and quantify them in food or medicine industries. Just like this paper did.


Conclusion(s)


With the use of HELP resolution method upon two-dimensional data combined with the abundant mass spectral database, one can exactly analyze such complex systems, like the traditional Chinese medicine and natural herbs, further the accurate and reasonable results could be obtained, and more constituents in a special system can be qualified and quantified better. This demonstrated that the combination of hyphenated instruments and relevant chemometric methods opens a new way for quick and accurate analysis of real unknown complex samples. The method developed in this paper can improve controlling quality of the materials and help examining the value of products in food industries and other regions.

Acknowledgement(s)


This work is financially supported by the National Nature Foundation Committee of PR China (Grant Nos. 20235020 and 20475066) and the Cultivation Fund of Key Scientific and Technical Innovation Project, Ministry of Education of China (No. 704036).

Authors Contribution(s)


Yu He, Xiao-Ru Li*, Shao-Yin Liu, Shi-Rong Wang, Bing-Xin Li, Yi-Zeng Liang
(Research Center of Modernization of Chinese Herbal Medicine, College of Chemistry
and Chemical Engineering, Central South University, Changsha 410083, P. R. China)
*Corresponding Author:
Dr. Xiao-Ru Li, Professor, College of Chemistry and Chemical Engineering,
Central South University, Changsha 410083, P. R. China
E-mail: xrli@mail.csu.edu.cn
Submitting Author:
Yu He, graduate student, College of Chemistry and Chemical Engineering, Central South
University, Changsha 410083, P. R. China.


References


1. O.M.Kvalheim, Y.Z. Liang, Heuristic evolving latent projections: resolving two-way multicomponent data.1. Selectivity, latent-projective graph, datascope, local rank, and unique resolution. [J]?Anal. Chem.  64 (1992) 936.
2. Y.Z. Liang, O.M.Kvalheim, H.R. Keller, D.L. Massart, P.Kiechle, F.Emi. Heuristic evolving latent projections: Resolving two-way multicomponent data. Part 2: Detection and resolution of minor constituents [J]?Anal. Chem. 64 (1992) 946.
3. Y.Z. Liang, O.M.Kvalheim, A. Rahmani, R.G. Brereton. Resolution of strongly overlapping two-way multicomponent data by means of Heuristic Evolving Latent projections. J.Chemon. 7 (1993) 15.
4. Pietta P G, Gardana C, Pietta A M. Analytical methods for quality control propolis[J]. Fitoerapia?2002, 73: 7-20.
5. Sticher O. Quality of Ginkgo preparations [J]. J. Planta Medica, 1993, 59(1): 2-11.
6. The Chemical Society. J. Chem. Inf. Comp. Sci., 1975, 15: 201.
7. H.R. Keller, D.L. Massart. Evolving factor analysis in the presence of heteroscedastic noise[J].  Anal. Chem. Acta , 246 (1991) 379.
8. H.R. Keller, D.L. Massart, Y.Z. Liang, O.M.Kvalheim. Evolving factor analysis in presence of heteroscedastic noise. Analytica Chimica Acta, 263 (1992) 125.
9. Liang Y Z?Kvalheim O M?Manne R. A two-way procedure for background correction of chromatographic/spectroscopic data by congruence analysis and leastsquares fit of the zero-component regions:Comparison with doublecentering[J]. Chemom. Intell. Lab. Syst., 1993, 18(3): 265-279.
10. Manne R, Shen H L, Liang Y Z. Subwindow factor analysis [J]. Chemom. Intell. Lab. Syst. 1999, 45(1-2): 171-176.
11. Shen H L, Manne R, Xu Q S, Chen D Z, Liang Y Z. Local resolution of hyphenated chromatographic data [J]. Chemom. Intell. Lab. Syst. 1999, 45(1-2): 323-330.
12. Gong F, Liang Y Z, Cui H, et al. Gas chromatography–mass spectrometry and chemometric resolution applied to the determination of essential oils in Cortex Cinnamomi.[J]. J. Chromatogr. A, 2001, 905(1-2): 193-205.
13. Gong F, Liang Y Z, Xu Q S, et al. Determination of volatile components in peptic powder by gas chromatography–mass spectrometry and chemometric resolution[J]. J. Chromatogr. A, 2001, 909 (2): 237-247.
14. Li X N, Liang Y Z. Analysis of the Volatile Fractions of Schisandra Chinensis (Turcz.) Baill. with GC/MS and Chemometric Resolution[J]. Phytochemical Analysis, 2003, 14: 23-33.
15. Gong F, Liang Y Z. Combination of GC-MS with local resolution for determining volatile components from si-wu decoction[J]. Journal of Separation Science, 2003, 26: 112-122.
16. Li B Y, Liang Y Z, Hu Y, et al. Evaluation of gas chromatography–mass spectrometry in conjunction with chemometric resolution for identification of nitrogen compounds in crude oil [J]. Talanta, 2003, 61(6): 803-809.
17. Zhang T M, Liang Y Z, Li B Y, et al. Identification of structures of nitrogen-containing compounds in crude oils in conjunction with chemometric resolution[J]. Annali di Chimica, 2004, 94(11): 783-791.
18. Zhang T M, Liang Y Z, Li B Y, et al. Systemic analysis of structures and contents of heteroatom-containing compounds in organic complex samples in conjunction with chemometric resolution technique[J]. Anal. Sci., 2004, 20(4): 717-724.
19. Zeng Z D, Liang Y Z, Li X R, et al. Alternative moving window factor analysis for comparison analysis between complex chromatographic data [J]. J. Chromatogr. A, 2006, 1107: 273-285.
20. Li X R, Liang Y Z, Zhou T, et al. Comparative analysis of volatile constituents between recipe Jingfangsan and its single herbs by GC-MS combined with AMWFA method [J]. J. Sep. Sci., 2009, 32(2): 258-266.
21. Leung A K, Gong F, Liang Y Z, et al?Analysis of water soluble constituents of Cordyceps Sinensis with Heutiatic Evolving Latent Projections[J]. Anal. Lett., 2000, 33(15): 3195-3211.
22. Li B Y, Liang Y Z, Du Y P, et al. Resolution and identification of the acidic fraction of a petroleum ether extract of Radix Rehemaniae preparata by an evolving chemometric approach[J]. J. Chromatogr. A, 2003, 57(3-4): 235-243.
23. Guo F Q, Liang Y Z?Xu C J, et al. Determination of the volatile chemical constituents of Notoptergium Incium by gas chromatography-mass spectrometry and iterative or non-iterative chemometrics resolution methods [J]. J. Chromatogr. A, 2003, 1016(1): 99-110.
24. Van Den Dool, H. Dce. Kratz, P.  A generalization of the retention index system including linear temperature programmed gas-liquid partition chromatography[J]. J Chromatogr, 1963, 11: 463~471
25. Richmond R. Calibrated salvage of gas chromatography capillary column retention indices [J].  J. Chromatogr. A, 1996, 742: 131-134.
26. Richmond R., Pombo-Villar E. Use of persistent trace gas chromatography artifacts for the calculation of Pseudo-Sadtler retention indexes [J]. J. Chromatogr. A, 1998, 811: 241~245
27. Cavaleiro C, Salgueiro L R. Analysis by gas chromatography-mass spectrometry of the volatile components of Teucrium lusitanicum and Teucrium algarbiensis[J]. J. Chromatogr. A, 2004, 1033: 187-190.
28. Hudaib M, Speroni E. GC/MS evaluation of thyme (Thymus Vulgaris L.) oil composition and variations during the vegetative cycle[J]. J. Pharm. Biomed. Anal., 2002, 29: 691-700.
29. Li Xiao-Ning, Liang Yi-Zeng, Chau Foo-Tim, et al. Smoothing methods applied to dealing with heteroscedastic noise in GC/MS[J]. Chemometrics and Intelligent Laboratory Systems,2002,63(2):139-153.
30. Chen ZengPing, Morris Julian, Martin Elaine, et al. Recursive evolving spectral projection for revealing the concentration windows of overlapping peaks in two-way chromatographic experiments[J]. Chemometrics and Intelligent Laboratory, 2004, 72(1): 9-19.
31. Xu Cheng-Jian, Liang Yi-Zeng, Chau  Foo-Tim, et al. Identification of essential components of Houttuynia cordata by gas chromatography/mass spectrumetry and the integrated[J]. chemometric approach Talanta,2005, 68(1): 108-115.
32. Shen Hailin, Liang Yizeng, Kvalheim OlavM, et al. Rolf Determination of chemical rank of two-way data from mixtures using subspace comparisons[J]. Chemometrics and Intelligent Laboratory Systems, 2000, 51(1): 49-59.
33. Liang Yi-Zeng. Chemometrics applied to Chinese medicine[J]. Chemical Journal of Chinese Universities, 1999, 20(5): 454-456.
34. Liang Yi-Zeng, Xie Peishan, Chan Kelvin, et al. Quality control of herbal medicines[J]. Journal of Chromatography B, Analytical Technologies in the Biomedical and Life Sciences, 2004, 812(1-2): 53-70.
35. Li Bo-Yan, Hu Yun, Liang Yi-Zeng, et al. Quality evaluation of fingerprints of herbal medicine with chromatographic data[J]. Analytica Chimica Acta, 2004, 514 (1): 69-77.
36. Golub G H, Loan F V. Matrix Computations [M]. Baltimore: Johns Hopkins University Press. MD, 1983.
37. Xu C J,  Liang Y Z, Li Y. Chemical rank estimation by noise perturbation in functional principle component analysis [J]. The Analyst, 2003, 128: 75-81.
38. Shen H L, Wang J H, Liang Y Z, et al. Chemical rank estimation by multiresolution analysis for two-way data in the presence of background[J].. Chemom. Intell. Lab. Syst., 1997, 37:  261-269.

Source(s) of Funding


This work is financially supported by the National Nature Foundation Committee of PR China (Grant Nos. 20235020 and 20475066) and the Cultivation Fund of Key Scientific and Technical Innovation Project, Ministry of Education of China (No. 704036).

Competing Interests


None

Disclaimer


This article has been downloaded from WebmedCentral. With our unique author driven post publication peer review, contents posted on this web portal do not undergo any prepublication peer or editorial review. It is completely the responsibility of the authors to ensure not only scientific and ethical standards of the manuscript but also its grammatical accuracy. Authors must ensure that they obtain all the necessary permissions before submitting any information that requires obtaining a consent or approval from a third party. Authors should also ensure not to submit any information which they do not have the copyright of or of which they have transferred the copyrights to a third party.
Contents on WebmedCentral are purely for biomedical researchers and scientists. They are not meant to cater to the needs of an individual patient. The web portal or any content(s) therein is neither designed to support, nor replace, the relationship that exists between a patient/site visitor and his/her physician. Your use of the WebmedCentral site and its contents is entirely at your own risk. We do not take any responsibility for any harm that you may suffer or inflict on a third person by following the contents of this website.

Comments
0 comments posted so far

Please use this functionality to flag objectionable, inappropriate, inaccurate, and offensive content to WebmedCentral Team and the authors.

 

Author Comments
0 comments posted so far

 

What is article Popularity?

Article popularity is calculated by considering the scores: age of the article
Popularity = (P - 1) / (T + 2)^1.5
Where
P : points is the sum of individual scores, which includes article Views, Downloads, Reviews, Comments and their weightage

Scores   Weightage
Views Points X 1
Download Points X 2
Comment Points X 5
Review Points X 10
Points= sum(Views Points + Download Points + Comment Points + Review Points)
T : time since submission in hours.
P is subtracted by 1 to negate submitter's vote.
Age factor is (time since submission in hours plus two) to the power of 1.5.factor.

How Article Quality Works?

For each article Authors/Readers, Reviewers and WMC Editors can review/rate the articles. These ratings are used to determine Feedback Scores.

In most cases, article receive ratings in the range of 0 to 10. We calculate average of all the ratings and consider it as article quality.

Quality=Average(Authors/Readers Ratings + Reviewers Ratings + WMC Editor Ratings)