AEM
Home Help [Feedback] [For Subscribers] [Archive] [Search] [Contents]
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Huber, F.
Right arrow Articles by Peduzzi, P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Huber, F.
Right arrow Articles by Peduzzi, P.
Agricola
Right arrow Articles by Huber, F.
Right arrow Articles by Peduzzi, P.

 Previous Article  |  Next Article 

Applied and Environmental Microbiology, July 2004, p. 4390-4392, Vol. 70, No. 7
0099-2240/04/$08.00+0     DOI: 10.1128/AEM.70.7.4390-4392.2004
Copyright © 2004, American Society for Microbiology. All Rights Reserved.

SHORT REPORT

Online Tool for Analysis of Denaturing Gradient Gel Electrophoresis Profiles

Florian Huber and Peter Peduzzi*

Institute of Ecology and Conservation Biology, University of Vienna, Vienna, Austria

Received 28 August 2003/ Accepted 12 March 2004

ABSTRACT

We present an online tool (EquiBands, http://www.univie.ac.at/IECB/limno/equibands/EquiBands.html) that quantifies the matching of two bands considered to be the same in different samples, even when samples are applied to different denaturing gradient gel electrophoresis gels. With an environmental example we demonstrate the procedure for the classification of two bands of different samples with the help of EquiBands.

In denaturing gradient gel electrophoresis, 16S rRNA gene amplicons are separated in a linear gradient of a denaturant, producing characteristic band patterns ("fingerprints") of complex mixtures of microorganisms (3, 4). A standardized classification procedure for bands that appear in different samples is vital for the interpretation of band profiles. Visual analysis alone cannot provide a quantitative basis for the decision of whether two bands are equivalent or not. Owing to nonuniform migration distance (the "smile effect") (see Fig. 2), bands are sometimes spatially distorted at the edges of a gel, leading to patterns that are not horizontally aligned. Estimation by eye can therefore lead to incorrect assignments of pairs of bands, especially when complex band profiles in distant lanes or even different gels have to be compared.



View larger version (63K):
[in this window]
[in a new window]
  		FIG. 2. Denaturing gradient gel electrophoresis band patterns of four different bacterioplankton samples. Bands between > and < symbols were treated as internal standards. Tested bands are indicated by asterisks.

 
To improve the method of band classification, we used simple regression analysis for comparing band profiles of gels with linear gradients. Theoretically, a regression analysis of two identical samples, where the x and the y axes correspond to the migration distances of the bands that appeared within the fingerprints, reveals a correlation coefficient of 1.

With an environmental sample consisting of 12 bands that was applied repeatedly both to one and to two different gels, we confirmed that a linear model can be used to compare band profiles (Table 1). Pairs of bands with the same migration distance define the regression line and are designated "internal standards." In this test we used the same sample repeatedly; therefore, all 12 bands were treated as internal standards. When the fingerprints generated were compared, the migration distances of bands were different in absolute values but the patterns of bands were equal. To evaluate the quality of the regression the correlation coefficient is presented as a decimal value in the upper left corner of the applet. The correlation coefficients were never below 0.99994 (Fig. 1; Table 1), indicating a strong linear relationship between bands.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Migration distances for one sample applied to the same gel and two different gels

 


View larger version (27K):
[in this window]
[in a new window]
  		FIG. 1. Graphic output window of the Java applet EquiBands. Migration distances, in pixels, of pairs of bands of samples from one gel (see Fig. 2 and Table 2) were plotted and calculated with EquiBands. It was determined whether the pair of band at positions 499 and 492 are equivalent.

 
In the following example we show the procedure forclassifying two bands in distant lanes of one gel (Fig. 2). It was tested if a band in sample 1 is equivalent to a band in sample 4 (both marked with an asterisk). Each band detected in the fingerprint patterns of bacterioplankton samples was referred to as an operational taxonomic unit (OTU). Bands that occurred at the same height in gels with the same linear gradients were regarded as equivalent OTUs. However, it has to be kept in mind that bands that appear at the same height in the gel could also be derived from different sequences with the same melting behavior. At the beginning of the identification process the comparison of bands was restricted to neighboring lanes. In this phase of analysis it was possible to determine visually if bands had equal migration distances. Figure 2 shows that some bands form rows of bands on the gels. These bands, called internal standards, were used as controls for the equity of migration distances of bands that appeared in distant lanes. Even though sequence identity was not confirmed by sequencing, our assumptions regarding the equity of fragments are supported by the fact that bands appeared persistently within samples taken from the same environment in relatively short intervals. Therefore this method is applicable to samples that have a spatio-temporal connection with each other, e.g., sequential samples collected during a monitoring program or a bioassay experiment (1, 2, 5). Replacing bands with their migration distances produced a numerical transformation of all fingerprint patterns (Tables 1 and 2).


View this table:
[in this window]
[in a new window]
  		TABLE 2. Migration distances of selected bands from two samples (see Fig. 2)a

 
To decide if migration distances of bands were equivalent we used the EquiBands applet. First, the regression line was defined by entering at least three internal standards into the corresponding input field. A correlation coefficient near 1 indicates that a linear model is applicable to the data set. A high number of internal standards is recommended for the analytical procedure presented; the regression calculated by the applet then becomes more robust. Furthermore the reliability of the regression is improved when bands are used as standards that are distributed along a wide range of the lanes. This means that test bands should lie within the area of the standard bands.

When the input field on the right of the applet window is used, EquiBands switches to test mode and the color of the output changes from red to black. After the data of the tested band pair are entered, EquiBands calculates a band pair that perfectly fits the linear model, which is based on the internal standard. This hypothetical pair of bands consists of the first test band and a band that is computed on the basis of the parameters of the regression line. The difference, in pixels, between the second test band and the band that corresponds optimally to the first tested band is called the "divergence from optimum" and is displayed together with the hypothetical pair of bands ("optimal matching") on the output screen of the applet. A small divergence indicates that the band pair tested corresponds to the linear model and therefore that the bands are likely equivalent.

A limit for the matching of two bands should be set depending on the sharpness of the bands in the gel, the quality of the gel, and the resolution of the image of the gel. In our example (Fig. 2) we used a tolerance of 6 pixels, the observed thickness of most bands in the image. That means a band that was within the range of 3 pixels of the position that was given by the optimal matching was considered equivalent. The results from our example show that the tested band pair missed the optimal match by only 1 pixel (Fig. 1; Table 2).

Especially when profiles from different gels are compared and bands are not sequenced, special attention should be paid to the adequate identification of OTUs. The tool presented here allows quantification of the matching of bands, allowing comparison of samples in distant lanes and even in different gels. Furthermore, analysis with EquiBands can help to decide which bands should be sequenced if a taxonomical analysis is to be performed.

(This work was submitted in partial fulfillment of the requirements for an M.S. degree from the University of Vienna by F.H.)

ACKNOWLEDGMENTS

We thank the National Park Authority and the Austrian River Authority for enabling our research in the Danube Alluvial Zone National Park. Our research was funded by the Austrian Science Foundation (grants P11720 BIO and P14721 BOT).


   FOOTNOTES
 
* Corresponding author. Mailing address: Department of Limnology, IECB, University of Vienna, Althanstr. 14, 1090 Vienna, Austria. Phone: (01) 4277-54342. Fax: (01) 4277-9542. E-mail: peter.peduzzi{at}univie.ac.at. Back

REFERENCES

  1. Ferrari, V. C., and J. T. Hollibaugh. 1999. Distribution of microbial assemblages in the central Arctic Ocean basin studied by PCR/DGGE: analysis of a large data set.Hydrobiologia 401:55-68.[CrossRef]
  2. Lindström, E. S. 1998. Bacterioplankton community in a boreal forest lake. FEMS Microbiol. Ecol. 27:163-174.[CrossRef]
  3. Muyzer, G., T. Brinkhoff, U. Nübel, C. Santegoeds, H. Schäfer, and C. Wawer. 1998. Denaturing gradient gel electrophoresis (DGGE) in microbial ecology, p.1 -27. In A. D. L. Akkermans, J. D. van Elsas, and F. J. de Bruijn (ed.), Molecular microbial ecology manual. Kluwer Academic Publishers, Dordrecht, The Netherlands.
  4. Sekiguchi, H., M. Watanabe, T. Nakahara, B. Xu, and H. Uchiyama.2002 . Succession of bacterial community structure along the Changjiang River determined by denaturing gradient gel electrophoresis and clone library analysis. Appl. Environ. Microbiol. 68:5142-5150.[Abstract/Free Full Text]
  5. van Hannen, E. J., G. Zwart, M. P. van Agterveld, H. J. Gons, J. Ebert, and H. J. Laanbroek.1999 . Changes in bacterial and eukaryotic community structure after mass lysis of filamentous cyanobacteria associated with viruses. Appl. Environ. Microbiol. 65:795-801.[Abstract/Free Full Text]

Applied and Environmental Microbiology, July 2004, p. 4390-4392, Vol. 70, No. 7
0099-2240/04/$08.00+0     DOI: 10.1128/AEM.70.7.4390-4392.2004
Copyright © 2004, American Society for Microbiology. All Rights Reserved.




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Huber, F.
Right arrow Articles by Peduzzi, P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Huber, F.
Right arrow Articles by Peduzzi, P.
Agricola
Right arrow Articles by Huber, F.
Right arrow Articles by Peduzzi, P.

Home Help [Feedback] [For Subscribers] [Archive] [Search] [Contents]
J. Bacteriol. Microbiol. Mol. Biol. Rev. Eukaryot. Cell All ASM Journals