Chapter 1
Analyzing Gel Images with Delta2D
This chapter provides a short overview about what can be done with Delta2D, based on example data
that comes with the installation of Delta2D.
1.1 Setting up Projects
________________________________________________________________________________
-
Note:
- In 2003 DECODON has introduced full expression profiles to avoid missing values in
the quantitation table and the resulting problems during statistical analysis.
Read more about the benefits of 100% per cent Spot Matching at
www.decodon.com/Solutions/Delta2D/100_Percent_Spot_Matching.html.
In a nutshell, the typical workflow for creating complete expression profiles contains the following
steps:
-
Create a New Pool
- To keep our experimental data handy, we recommend to create a new
pool for every new experimental context. (See section 4.1 for details on how to do this)
-
Import Your Gel Images Into Delta2D
- See section 4.1
-
Load the images into your project
- Create a new Project (see 4.2), include the relevant
images and distribute them on groups, so that replicate gel images from the same sample
are all in the same group.
-
Assign the Group Warping Strategy
- Use the Strategy Manager (See section 5.10) to
assign the Group Warping Strategy to the complete project. It might be necessary to
change the warp mode for single pairs, but using the warping manager you make sure
to obtain persistent warping chains and don't produce warping cycles. Additionally you
reduce the number of gel pairs you have to handle to the minimum necessary explicit
warps.
-
Edit the explicit warps
- Examine the gel image pairs with explicit (any warp mode but the
implicit one) warp mode one by one and correct them where necessary by setting match
vectors and / or even changing the warp mode to a more suitable if necessary. For details
please refer to section 5.7.
-
Compensate for Experimental Variations
- Depending on the number of replicates per
group, you might want to condense the replicate groups to a single image per group,
compensating for all experimental variations within each group. To achieve this, create an
Fused Image of each group, using average fusion. In section 4.8 you can learn everything
about image fusion and how to perform it.
-
Create a Fused Image of your complete project
- Irrespective whether you produced
Average Fused Images from your groups, it is a good idea to create a fused image over
the complete project, using either Max Intensity or Union as fusion type. Please refer to
section 4.8 for details, or simply produce both image fusions and compare them, if you
don't know which one to prefer. When doing so, please be careful not to include the first
fused image when producing the second.
-
Detect Spots
- on the image fusion to include every spot existing on any gel image in your
project. Thus you get a central set of spots, which can be edited on a central place and and
transferred to the other gel images afterwards. If necessary you can adjust parameters to
reduce the number of false positive or increase the number of detected weak spots. Details
on detection parameters can be found in section 6.1.
-
Edit Spots
- to separate two spots detected as one or to join two parts of a spot detected
accidentally as separate spots if necessary. Spot editing needs to be done on your "master"
set of spots, if you plan to transfer spots.
You can find more on editing spots in section 6.5.
-
Transfer Spots
- to all other gel images like described in section 4.9.
-
Open Quantitation Tables
- to view the quantitation of all spots. Sort the tables after your
needs to find the interesting spots or refer to the next section on how to identify the
interesting spots.
-
Perform Statistical Analysis
- to employ a variety of advaced methods for finding patterns
in your data, or for clustering your expression profiles: read on in section 8.6.
________________________________________________________________________________
-
Note:
- Delta2D is able to completey and reliably analyse DIGE projects. Even for DIGE
projects Delta2D's core technologies are quite useful: If you want to analyse a DIGE
project with several gels e.g. image warping will provide significant time savings and
quality improvements.
For analysing DIGE projects you will have to consider that the projects needs to be
defined as a DIGE project and that for every image the corresponding gel needs to be
defined and which of the images of a certain gel contains the internal standard.
Read more about how to set up DIGE projects with Delta2D in section 4.7
.
1.2 Finding Interesting Spots
Identifying Interesting Spots by their Appearance
Open or switch to the Dual View of the gel image pair
you want to examine with a double click on the respective table cell in the Project Manager. In the
Dual View click on the Equalize Image button in the tool bar to make sure that the histogram settings
of both images are balanced. Now you can see at a glance the tendence of every spot: black spots have
about equal volume in both images, which means that there is no mentionable change in expression.
Orange1
spots are increased in expression and blue spots decreased. The more their color tends to one or the
other, the bigger the difference in expression.
|
| Figure 1.1: | Quantified Spots, one unchanged, one upregulated |
|
Finding Interesting Spots in a Scatter Plot
Another approach to finding interesting spots is to
compare their expression level. In the spots menu of the Dual View please click on Spots
Show
Scatter Plot. This opens a graphical representation of the spots seen in these two gel images. The
position of a spot is determined by its relative quantity on each image: the x-position by its quantity in
Control A and the y-position by its quantity in the Sample 1 A. Thus, a spot having an unchanged
volume on both images, appears on the diagonal of this graph, whereas spots whose expression is
increased are found in the upper left part and spots whose expression decreased in the lower right part
of this graph.
| Figure 1.2: | Scatter plot: illustrates the expression differences between two gel images |
|
Now just click on one of the extreme spots in the top or bottom of the Scatter Plot and keep an eye
on the Dual View. The selected spot shape will be highlighted in the Dual View, and, if the image is
zoomed in, the view will scroll to the respective spot.
Identify Interesting Spots by Their Expression Profile
You can even see an expression profile of a spot over the complete project: in the menu of the Dual
View, please click on Rollups
Expression Profiles. A new window will open, keeping its position in
front of its parent window: a so-called Rollup. Make sure that the Segments Tool is selected and
simply point to any spot. You don't need to click or select it in any way, just keep the mouse cursor
above it. Now move the mouse pointer to the next spot and watch the Expression Profiles Rollup.
It changes its display, showing the expression of the respective spot across the complete
project.
| Figure 1.3: | Expression profil rollup indicating project wide spot intensity |
|
Filtering for Interesting Expression Profiles
Switch to the Quantitation Table by choosing from any menu Window
Quantitation Table. What
you see now is a set of tables full of values. You have a lot of data but still don't feel informed. How we
can find relevant spots now? Delta2D offers powerful tools to obtain your results: arbitrarily
combinable filters. With an adequate combination of filters, you can obtain the set of spots
corresponding to nearly every question. Let's try it out: We are looking for spots matching the
following criteria:
- increased intensity by a factor of at least 22
- increase should be observable under experimental conditions used in all images compared
to the control image
To see only the spots matching these criteria, just take the following steps:
- Switch to the quantitation table and select table for allimages.
- Make sure that the control image is selected as the Ratio Master in the Table Properties
window. Use the menu item Column
Table Properties to find this setting.
- Find the first column containing Ratio data. To see the name of a column, either drag the
line between two column headers to make it wider, or simply point to its header and wait
a second. Its name will appear as a tool tip (do not click). Click on the top part of the
column header, labeled Filter to open the respective filter.
- Find the Label in the section above and click on the field below this label.
- Type 2into the field Show values from:, or drag the left slider below the histogram in
this dialog to the right until the value of this field gets as close to 2 as possible.
- The check box Filter active will be checked automatically since you have changed the
filter settings.
- Apply the settings with the button Apply.
- Click now on the next ratio column in the table and repeat the above steps 4-7.
- Repeat the procedure with each ratio column in the statistics table.
Switch back to the Dual View to see the spots matching these criteria.