Chapter 2
A Quick Tour of Delta2D 3.4

This quick tour will give you a first impression of Delta2D's many features by showing how to use and analyze the example gel images and data that come with Delta2D. To keep this tour short, we will have to skip many interesting features. However, some of the left-out features will be listed at the end of each section.

________________________________________________________________________________

Note:

If Delta2D is not already installed on your computer, please refer to chapter 3 for installation instructions.

2.1 The Example Project

Let us start our quick tour by opening Delta2D. The first shown window is the Project Manager together with a dialog that invites you to open a project. During the start Delta2D provides assistance with a Tip of the Day. You might wish to review the different tips, otherwise simply close this window. Delta2D includes an example project which is named Demonstration. The example project contains six gel images plus one fused image generated by Delta2D. Please select it and press Open. Now you see the example project as a tabular view for gel images arranged in groups.

Figure 2.1: The project manager with the example project that comes with Delta2D

You can see the full name of a gel image plus additional information in a tooltip when you point on its thumbnails. A small icon in the header indicates whether there is a quantitation result available for the gel image. Another icon shows if there are labels attached to this gel image. As a rule, icons will appear only if data is available. Right clicking on a thumbnail opens a menu that shows the available operations for this gel image.

Each gel image is represented by a column and a row in the Project Manager's table. In a certain table cell, where the row of one gel image crosses the column of another gel image, Delta2D displays information for each gel image pair. The meaning of the icons is explained in detail in section 4.5. Point on a table cell to see information about the gel image pair. Right clicking on a table cell opens a menu that shows the available operations for this gel image pair.

Drag the border between two thumbnails to make all thumbnails larger or smaller. You can drag the gel images to change the order of the gel images.

The example project contains four groups whose thumbnails are colored in blue, dark orange, light orange and lemon respectively. The first group, the blue one, contains two images, from two control gels. The control gels are made from a sample of Bacillus subtilis under exponential growth. The four gel images in the orange groups are made from the same organism but at other points of time: two after one minute and two after 10 minutes imposition to stress. We have placed all replicates from the same sample into the same group, and you are free to put any number of gel replicates into the same group. The fourth group consists of one special image generated by Delta2D. This is a union fused image, thus condensing information about the spot signals from all other images. Additionally it serves as a simple proteome map since it contains labels with protein identifications.

2.2 Working with Gel Image Pairs

Figure 2.2: Project table details

At the bottom there is a black line indicating that all spots on the two gel images are perfectly matched (otherwise it would have blue or orange endings).

On top of this match status line you can see the icon if match vectors exist for this gel image pair.

The very left icon indicates the warp mode applied to this gel image pair and can look like this:

	Automatic Warp mode
	Exact Warp mode
	Global Warp mode
	Identical Warp mode and
	Implicit Warp mode.

For a more detailed description of warp modes please refer to section 5.7.

In the example project we have used the exact warp mode and the implicit warp mode only.

In the center of each cell you can see one of the following icons that provide feedback about the project's warping status:

	These two images can be warped according to the defined direct warp mode, since either no match map is needed (e.g. for warp mode identical if the images come from the same gel) or the match map contains approved match vectors.
	If the automatic warp mode is chosen, this icon means that you shall be aware of newly created non-approved match vectors. You should review these match vectors. Furthermore, the icon appears if the match map does not contain approved vectors but the warp mode (e.g. exact warp) needs a match map.
	You have specified the implicit warp mode for these two images. Instead of computing a direct warp for these two images, a warping chain of direct warpings via other images needs to be combined. This is not possible, since at least one of the warpings can not be computed.
	You have specified the implicit warp mode for these two images. In contrast to the previous icon description now you have defined too many direct warpings. A so called Warping Cycle occurrs, the matching may face conflicts. (For more details on warping cycle please refer to section 5.10)

In cells where you can't see one of the icons above, everything is fine, there is nothing to do.

As soon as you open a gel image pair in the Gel Image Pair View, the icon in the top left corner appears, symbolizing the false colors of the shown dual channel image.

Let us now open a gel image pair view. To do this, place the mouse cursor for example over the table cell in the first row and second column. The tool tip will show the names control_01 and control_02. Click on the button View or double click on the table cell to open a new window that contains a dual channel image of these two gel images.

The Gel Image Pair View

Figure 2.3: The Gel Image Pair View.

At the top you see the menu bar and the tool bar (see figure 2.4). On the left hand side there is a vertical panel with five tool buttons, they are used for activating the Match Vectors, Spots, Edit Spots, Zoom, and Labels tool, respectively. For now, we will only look at the available data, so we don't need to use any of the tools.

Comparing 2D Gel Images

Delta2D's approach to 2D gel images is very innovative compared to traditional software packages. In traditional packages spot detection has to be done before differential analysis of protein patterns, thus carrying spot detection and matching errors into the comparison. This requires time consuming manual correction before you can use the results of the "automatic" spot detection and matching advertised by these packages.

In contrast, Delta2D will find corresponding spots based on the complete image information of the whole experiment, using every pixel from all gel images. Additionally, you can guide this process by fixing a few pairs of corresponding spots by hand. Based on that information, Delta2D removes the variation in spot positions on different gel images, making spot matching much easier and reliable.

The process of matching regions of one image to those on another is done by warping the image. Delta2D's warping algorithms help you to generate dual channel images in which corresponding spots are perfectly overlaid. In the dual channel image, differences in protein expression levels can then be easily recognized. The same technology is applied in the subsequent quantitation step to obtain accurate and reliable spot matching information.

Figure 2.6: A region of the dual channel image, before and after exact warp. To illustrate the process of warping a grid is overlaid on each image before warping: as you can see, before warping the grids are perfectly overlaid. After warping, the blue grid, belonging to the master image, is still unchanged, whereas the orange grid shows the distortions applied to the sample image. Corresponding spots are now overlaid exactly, allowing easy identification of spots with changed expression level. Black spots are equally present on both gel images; blue means that a spot is present or much stronger on the master gel image, orange indicates spots that are present or much stronger on the sample gel image.

Having opened the dual image view on control_01 and control_02 you see match vectors that connect corresponding spots. The warp mode has been set to Exact by choosing Matches Warp mode Exact from the menu or the pull-down menu almost in the center of the tool bar. Now warp the images by choosing Matches Warp. This will create a dual channel image similar to figure 2.3.

Interpretation of the Dual Channel Image

For better interpretation, you may wish to hide match vectors using the overlay rollup as described in section 5.2 (use Rollups Overlays to open the overlays rollup). Clicking one of the small control buttons will toggle the visibility of the respective objects between three modes: visible, non-visible, and context controlled visibility.

In the dual channel image orange means that a spot is present only (or much stronger) in the sample gel image, while blue spots are present only (or much stronger) in the master gel image. This enables you to easily identify whole sets of spots with similar or varying expression levels.

Open the Colors rollup using Rollups Colors. The colored square shows how colors can be interpreted in the dual channel image (see Figure 2.7). From left to right the intensity of the master spots is increasing, while the sample spot intensity is increasing from the bottom to the top. The color code for common background pixels is in the bottom left corner, common pixel of maximum possible grey intensity (often in saturated spots) have the color of the top right corner. The diagram includes the color of any possible combination of spot intensities on the master and sample image, as displayed by Delta2D.

Figure 2.7: How colors are combined in the dual channel image

Navigating in Images

In Delta2D you have three possibilities to access details in your gel images:

Instant Magnification

Clicking and dragging with the right mouse button will open a magnifying glass you can move around all over your image.

The zoom tool bar

For quickly increasing, decreasing, and adjusting the magnification factor, use the buttons of the Zoom Tool Bar . Their function is identical with the commands in the View menu. For example, clicking on the zoom out button will zoom out the image view.

The zoom tool

To focus on a certain part of your image, you can use the Zoom Tool. It is activated by using the tool panel in the top left area of the main window (Figure 2.8). Press the zoom tool button to activate zoom mode. The mouse cursor will be changed to a magnifying glass. Click somewhere inside the image to enlarge it. Click and drag to specify a region that should be zoomed in.

Figure 2.8: The tool panel.

SmartVectors Technology for automatic warping

Delta2D's new SmartVectors Technology can align the gel images automatically. By pressing Find Match Vectors Delta2D will be initiated to automatically analyze the gel image pair for corresponding patterns and then display the resulting matchmap. To see what the SmartVectors Technology can do for gel images in the example project, you will have to remove all match vectors using Matches Delete.

Use the menu item Matches Warp Mode Automatic to set the warp mode to automatic and Matches Warp to start the warping process. The computing time will vary depending on your computer. If you are not satisfied with the result (especially at the gel borders sometimes minor corrections may be necessary) you can easily add or delete vectors manually when the Match Tool is activated and then use Matches  Warp again to see the effect. The match vectors are just proposals until you approve them. You can approve match vectors by selecting them (one-by-one or dragging a rectangle) and then use the menu item Matches Approve Selected . When you press the Find Match Vectors button now for asking Delta2D to automatically find match vectors again, you will see a dialog for non-approved match vectors where you have to decide what Delta2D should do with the non-approved match vectors. Having read the explanation in this dialog you will understand that you have two options to improve the matchmap iteratively:

1. Delete the smart match vectors that you do not want to include in the matchmap and let Delta2D automatically approve the non-approved match vectors.
2. Manually approve the smart match vectors to include them in the matchmap and let Delta2D automatically delete the remaining non-approved ones.

________________________________________________________________________________

Note:

SmartVectors Technology has been developed to accelerate the search for match vectors connecting corresponding spots on different gel images.

Read more at www.decodon.com/Support/Howto/SmartVectors.html

There's More. . .

Ratio mode

is a unique visualization tool that shows expression ratios directly on the gel images by color-coding pixels according to the sample / master intensity ratio. (see section 5.13)

Color schemes

You can use any combination of colors to produce the dual channel images. (see section 5.13)

Background subtraction

Locally varying and strong background can be eliminated from the images (see section 5.4)

Spots in 3D

See spots and regions of your gel images in a 3-dimensional representation. (see section 5.2)

Match vector snaps

speeds up setting match vectors by letting them snap to spot maxima. (see section 5.8)

Keyboard shortcuts

let you switch quickly between gel images, tools, and windows. (see chapter 10)

2.3 Quantitative Data

Using dual channel images one can quickly gain valuable information about changed spots. However, for many applications, the quantitation of spots is required to enable quantitative analysis. The data are obtained in three steps:

spot detection

– identification of image segments that are occupied by spots

spot quantitation

– summing up the gray values of the pixels belonging to each spot. Background is subtracted, and quantities are normalized before comparing quantities between gel images.

spot matching

– attempt to find one or more corresponding spots on the sample gel for each master spot (or vice versa). If you decide to use a fusion gel based consensus spot pattern spot matching is no longer necessary.

In traditional packages for the analysis of 2D gel images, these steps are error-prone and require extensive manual corrections. Delta2D tries to automate the analysis as far as possible, leading faster to results with higher reproducibility.

Spot detection is done automatically, controlled by a few parameters. Any "spot painting" or "trimming" by hand is obsolete. Quantitation is also done automatically – with only one parameter which is necessary for background subtraction.

The Quantitation Tables

Figure 2.9: The quantitation window

The tables may seem a bit overwhelming at first because they display a lot of information. However, you can easily customize them, e.g. by rearranging or hiding columns, or by sorting and filtering rows. To continue please select the all gel images tab on the buttom of the table.

Sorting Spots

Now let us sort the table by the relative volume of the master spots. Just click onto the

of the column header. A small arrow indicates the sort order, click again to sort in reverse order.

Figure 2.10: Quantitation table, sorted on relative volume in descending sequence

________________________________________________________________________________

Note:

Note that all spots you select in this table are also selected in all corresponding gels. If you want to handle only spots of one gel you have to switch to the corresponding one gel table by using the <gelname> tab.

Select one of the rows by clicking on it. Notice how the corresponding spot segments on the master and sample gel images are highlighted. By dragging the mouse over consecutive rows, you can select them as well. It is also possible to select a spot in the Gel Image Pair View by clicking on its boundary or center. The corresponding row in the table will be selected automatically.

Filtering Spots

As an example, we want to see expression profiles only where the spots of the gel image group "1 min" have an expression ratio of less than 0.5 or more than 2 compared to the "control" group. First make sure that the statistic table is visible (Refer the tabs on the buttom and select statistics). Choose the filter from the menu: Filter Ratios ratio '1 min' / 'control'.

Figure 2.11: Editing a ratio filter. Here, Delta2D will show only spots that have expression ratios below 0.5 and above 2.

The button in the header of the filtered column has changed from Filter... to a short description of the filter. Leave your mouse pointer over the button for a while for a tool tip containing the filter definition. Click on a column's filter button to access the filter diaolog.

Filters can be combined to implement more complex criteria, such as "show all expression ratios of well reproduced spots (relative standard deviation rsd within a group of less then 30%), with expression ratios (based on the mean of the groups' normalized volumes) below 0.5 and above 2, and with spots that contain a normalized volume (%V) of at least 0.02."

Scatter Plots

Figure 2.12: The scatter plot

There's More. . .

Complete synchronization between tables and gel image views

What you see in the table is what you see in the gel image view; what you select in the table will be selected in the gel image view.

Working with spots in the gel image view

You can select, cancel, mark, or hide spots and spot groups in the gel image view. (see section 8.2)

Normalization sets

let you define a set of spots that can be used to normalize for relative spot quantities. (see section 8.1)

2.4 Labels

Delta2D allows you to place labels anywhere on a gel image – independently from spot locations. You can use them to annotate spots and to control spot picking. Also labels can be used for pI and MW determination of defined positions within a gel. Labels can be transferred from one gel to the another, Delta2D will place them in the right position automatically, using the same alignment that produces the dual channel image. Managing master gel images containing spot identifications is easy – a single click will transfer labels from and to the master gel image. You can change label formats according to your preferences. Label data and formats are saved in XML files that can be easily processed by other applications.

Labels in the Example Project

The example project includes some labels for the proteome map basing on the fusion gel (lemon group). They are displayed on the gel image view as well as in the quantitation tables. To see which labels exist, open the quantitation table for the fusion gel and click on the label column header to sort spots by label name. The spots with labels will now head the table.

Creating Labels

You can place labels on either of the gel images shown in the Gel Image Pair View. Usually, labels for both gel images will be displayed together on the dual view, but with a different look.

To start creating labels, select the label tool in the tool bar of the Gel Image Pair View (see figure 2.13).

Figure 2.13: Delta2D tool panel with activated label tool.

Changing the Text of a Label

To change the text of the label, click once inside the label and start typing. Press Enter to stop editing. By pressing Escape, the changes you have made will be discarded.

Moving a Label and its Target

You can move a label's text around by dragging it with the mouse. See how the line is placed automatically into the corners or in the middle of the label's border. When the label is sufficiently near the target, the line will be hidden in order to make the display simpler.

When you want to change the label's target you can drag the line to move the whole label to the desired place. By default, the label will snap to spot centers when you drag it around. Hold down the Ctrl key to switch off the snapping.

The Label Context Menu

Right click on a label to get its context menu (see figure 2.15).

Figure 2.15: The context menu for a label

Moving a Label to Another Image (Label Transfer)

In the context menu you see whether the label is in the master or in the sample image. Newly created labels are always placed on the master image. Select Move to Sample to move a label to the sample gel. The label's format will be usually changed when you move it to another gel.

When both gel images are connected by a match map, the label will be moved according to that match map. Say, your label is placed on the master image and you have loaded a match map for master and sample image. Then you place the label inside a master spot. When you move the label to the sample gel, Delta2D will move it to the point that corresponds to the master spot. This behavior allows you to collect labels from many different gel images, which is especially useful when you want to produce a proteome map containing protein identifications. See 7.5 for details. For collecting labels it is very convenient to use the fusion gel.

Copying a Label to Another Image

Deleting a Label

There's More. . .

Complete synchronization between labels and spots

Moving a label from one spot to another will be immediately reflected in the quantitation table.

Greek letters

can be freely used in label names.

Multiple labels per spot

You can add any number of labels to a spot.

Label formats

let you change color and visibility of the labels' arrow, border, text etc..

Spot picking

You can have your labels numbered automatically and let Delta2D export labels and spot centers to a variety of formats for controlling spot picking robots. (see section 8.6)

Scouts

Query web based databases for protein data for single or multiple spots and have them all attached to labels. (see section 7.7)

2.5 Analysing Projects

Delta2D keeps your projects organized by putting all relevant data in one location – the gel pool, a central repository for all your gel images and associated data such as labels, quantitation results and match maps. You can use the Project Manager to access and organize data in the pool. Behind the scenes, Delta2D updates the pool data every time you make a change to the project.

Expression Profiles

Figure 2.16: The Expression Profiles window. Each expression profile of marked spots creates a corresponding bar chart.

Gel Image Regions

Figure 2.17: The same region of four different gel images

When you have opened a gel image view window, its displayed region will determine the first region shown in the Gel Image Regions window. Within the gel image regions window you can navigate the regions synchronously, zoom in or out, and control the visibilty of spot boundaries, background, and labels as well as the application of the histogram settings as they are defined in the gel image views.

Furthermore, you can balance the gel image regions using the button for equilization to have better comparable view on the image regions. Remove a gel from the Gel Regions View by hiding it from the All gel images table in the Quantitation Table(see section 8.1).

There's More. . .

All data is kept in one location

making it easy to do backups or move data around (see section 4.1).

Multiple pools

You can have more than one pool.

Background jobs

Let Delta2D execute quantitation and automatic warping for the whole project in the background, while you work. (See section 4.10)

Open data format

All project data is stored in XML, an industry standard, open data format. This makes your data easy to access for further processing.

Statistics table

The statistics table (fig.2.18) displays aggregate data on gel groups, such as minimum and maximum values, means and standard deviations. Note: since the statistics table refers to groups of gel images, some of the data does not contain relevant information if groups do not contain replicates.

---

Figure 2.18: Statistics on the gel groups

---

Color Coded Spots

See the distribution of spots over complete sets of gel images on one single image, controlled by your own criteria. (see section 8.4)

Image Fusion

Create your own proteome maps by performing Image Fusion. Image Fusion is one of Delta2D's outstanding features which combines multiple gel images by using intelligent algorithms to one artificial but nevertheless authentical image. (see section 4.8)

2.6 Instant MS Excel Reports

Delta2D lets you generate reports about your analysis project with a single click (see Figure 2.19). Open a quantitation table and choose File Generate report in Excel to produce an Excel worksheet that contains the currently visible data in the table, plus an extensive set of diagrams and statistics. You need to have Excel, versions 2000 (9), XP (10), or 2003 (11) installed to use this feature.

________________________________________________________________________________

Note:

The creation of a report makes sense only for the purpose of comparing several gel images with each other. Thus this feature is only available in the view of the table showing the data for all gel images.

Figure 2.19: A page from the MS Excel report for the example project

2.7 Instant Export to MS PowerPoint

You can create a PowerPoint slide that includes everything you see in the gel image view: images, spots, and labels (Figure 2.20). In the Gel Image Pair view, use View Export To PowerPoint to produce a PowerPoint slide that contains the currently visible objects in the gel image view. These objects are fully editable inside PowerPoint. You need to have PowerPoint version 2000, XP, or 2003 installed to use this feature.

Figure 2.20: A PowerPoint slide created from a gel pair