Chapter 6
Spot Quantitation and Matching

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. Basically, the data are obtained in three steps:
spot detection
identification of image segments that are occupied by spots
spot quantitation
summing up and calibration curve dependent recalculating of 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
on individual detection: attempt to find one or more corresponding spots on the sample gel for each master spot (or vice versa). When transferring spots from another gel image (e.g. a proteome map), matching is inherent to the process of transferring and thus comprises 100% of all detected spots.

Spot detection is done automatically, controlled by a few parameters that are set by the user. In Delta2D, any "spot painting" or "trimming" by hand is obsolete.

Once spots have been detected, quantitation is also done automatically – with only one parameter that is necessary for background subtraction.

Since Delta2D includes image warping (introduced into 2DE gel image analysis by DECODON in the year 2000), spot matching is very reliable, even with individual spot detection for each gel image. In traditional packages for the analysis of 2D gel images, where spot matching is based on spot patterns rather then spot positions, these steps are error-prone and require extensive manual corrections. In Delta2D spot matching is done automatically since after warping corresponding spots already have the same position throughout the whole experiment.

With its intuitive and modern approach, Delta2D tries to automate the analysis as far as possible, leading faster to results you can rely on.

________________________________________________________________________________

Note:
Based on image warping and image fusion, DECODON has introduced full expression profiles to avoid missing values in the quantitation table and the resulting problems during statistical analysis. To receive full expression profiles, create a Proteome Map make an union fused image out of the whole set of images in your experiment(at least out of one representative image of every group of replicates). Then do the spot detection on the resulting Proteome Map only and transfer the spots to the original images. Please refer to sections 4.8 ff for details.

Read more about the benefits of 100% Spot Matching at www.decodon.com/Solutions/Delta2D/100_Percent_Spot_Matching.html.

Starting the spot detection is easy: In the Project Manager right click on the thumbnail of the respective image (probably the fused image), select Quantify Gel Image . . . from the context menu. Delta2D presents the Quantitation Dialog to set or confirm the settings before Quantitation itself is done.

6.1 The Quantitation Dialog

The Quantitation dialog comes with a proposal for three numerical parameters. The numbers are derived directly from the images and should lead to reasonable results. However, you can change the parameters according to your individual preferences. Having changed the parameters proposed by Delta2D, you can restore the proposal again by using the feature Parameter estimation. Simply click on icons/Wizard16. Please refer to section 6.3 for a detailed description of each of the parameters.

The appearance of the Quantitation dialog is shown in figure 6.1. Note that when you have chosen to quantify only one gel, only the parameters corresponding to that gel will be editable.


ui/dialog_quantitation_scaled
Figure 6.1: The quantitation dialog.

Quantifying One Single Gel

This dialog will appear when you have chosen to perform spot detection and quantitation for one single gel image (probably the proteome map).

Besides from the Project Manager, you can invoke this dialog from the Gel Image View from the Spots menu, by selecting the Spots |\ Quantify |\ Gel name of first gel. . . or Spots |\ Quantify |\ Gel name of second gel. . . menu item , as appropriate.

The meaning of a parameter is explained in the panel on the right-hand side. Please refer to section 6.3 for a detailed description of each parameter.

Parameter sets can be saved and loaded using the buttons in the top right panel.

The set of Parameters, i.e. the list of parameters that were used for quantitation, normally is saved within the *.qnt files which contain the quantitation information of a 2D gel image. You can load a parameter set from a previously exported quantitation file by loading the corresponding *.qnt file in this dialog.

Clicking on the OK button will start the spot quantitation process, Cancel will discard your settings.

Quantitation is always done using the original images while warping and histogram adjustment have no effect on the results. The background for a spot is computed and subtracted automatically, it is the very same background that is switched on and off using the layer panel.

When the quantitation process is finished, spot boundaries will be shown in the main window of the Gel Image View. The spot boundaries for the respective gel image are overlaid on the image, placed on a separate layer (one spot layer per gel image). These layers can be switched on and off, just like the image layers, using the overlays rollup. Select Rollups |\ Overlays to open the rollup.

Spot centers are marked by points. The center is located where a spot cutter would obtain the maximal protein amount.

Quantitation data can be saved as a complete set of data (spot boundaries and quantities for the whole gel image) using the Spots |\ Export |\ menu.

Individual Quantitation for Both Gel Images

________________________________________________________________________________

Note:
For the sake of completeness we describe the process of quantitative analysis for one or few gel pair(s) with individual detection. In some cases it might might be necessary to proceed this way. Nevertheless we recommend in most cases to use a fused image for detection and to transfer the spot boundaries to all (other) images in the project for obtaining a 100% spot matching throughout the complete project.

ui/dialog_quantitation_dual_scaled
Figure 6.2: The quantitation dialog for both gel images loaded in the Gel Image View.

This dialog will appear when you have chosen to perform spot detection and quantitation for both gel images. It allows you to specify parameters for the detection and quantitation process corresponding to both gel images.

You can invoke this dialog only from the Gel Image View, from the Spots menu, by selecting the Spots |\ Quantify |\ Both Gel Images. . . menu item, by using the keyboard shortcut sequence Ctrl+Alt-Q, or, if there are no quantitations available, by clicking on the Segments tool icon icons/spots-tool16 in the tool panel.

Clicking on the OK button will start the spot quantitation process, resulting in the display of the quantitation table. It will take a few moments for Delta2D to find all the spots in your gel images and prepare the table: if you realize you have made a mistake with your parameters or match vectors, you can stop the computation at any time by clicking on the Cancel button in the progress window that appears. You can then go back and make corrections before starting the quantitation process again.

6.2 Spot Matching on Individual Quantitation

Spot matching is integrated into the quantitation process and happens mostly unnoticed by the user. Delta2D just checks if two spots overlap sufficiently, taking overlap quantities into account. It may happen that a spot has more than one corresponding spot on the other gel, or no corresponding spot at all.

________________________________________________________________________________

Note:
Delta2D's spot matching relies on well-aligned gel images, so - depending on the warping strategy - you might need to have set match vectors before.
When quantitation data is available for both gel images, Delta2D will automatically match spots according to the current match map. Similarly, you will be asked if Delta2D should update the spot matching when you have changed the match map and then request the table view for spot pairs. Note that the spot matching is computed from the same match map that is used for producing dual channel images, making it easy to control the quality of the matching.

6.3 Spot Detection and Quantitation Parameters

Configuring the Local Background Region

The local background region refers to the radius (in pixels) of the region used for local background determination in the quantitation step. This option controls the computation of background quantities (influencing spot quantities and ratios). Lower values result subtraction of more background from the spots volume, especially for large spots.

Reasonable values should be 1.5 to 2.0 times the diameter of the largest spot in the image.

________________________________________________________________________________

Note:
This option also effects the snap to spot feature. If this parameter is set to a value that is much too high, it happens that snap to spot gets difficulties to differentiate between adjacent spots.

Configuring the Average Spot Size

Specifying the average size of spots in your gel images enables Delta2D to separate overlapping spots more accurately , as well as to distinguish spots from the image background. The value specified refers to the radius of an average spot, in pixels.

Higher values will decrease noise sensitivity. Use lower values to separate spot clusters better and to detect very small spots.

Note that you can get an idea of the size of the spots in your image by looking at Delta2D's status bar, at the bottom of the main frame. The status bar displays information about the current pixel position of the mouse pointer within a gel image, so you can see how large your spots are by moving the mouse over the spots and observing the number of pixels covered directly.


ui/statusbar_pixelcount
Figure 6.3: The cursor position in pixel count in the Status bar

Configuring the Weak Spot Sensitivity

This parameter allows you to control how strictly Delta2D discriminates between spots and the small signals. Being able to configure this sensitivity is useful when you have gel images containing very weak spots.

Just type a percentage value directly into the text entry field corresponding to either the master or sample gel image.

Specifying a higher value will result in Delta2D detecting spots with weak intensity more reliably. Specifying a lower value will mean that more noisy background artifacts in the images will be suppressed successfully.

Generally, a value between 5 and 20 % is suitable.

Saving and Loading Sets of Options

Delta2D saves the parameters used for detection together with the detected spots for each gel image individually. Additionally, Delta2D lets you export and import your options to and from files, e.g. for exchange with other Delta2D users, or if you want to try out different settings but want to keep a special one.

To save the current set of options to a file, click on the Save icon icons/Save16. A dialog will appear, enabling you to specify a file to save your options to.

To load previously saved options, click on the Load icon icons/Open16. A dialog will appear, enabling you to choose a file from which to load a set of options.

6.4 Pixel Based or Modeled Spot Boundaries

Pixel based spot boundaries directly reflect the raw gray value distribution within the scanned gel image. Since the gel images usually include noise and spots divided into pixels, this kind of spot boundaries regularly look erratic.

For different reasons, be it that, due to a low resolution of your image, the spot outlines look to rough, be it for purposes of printing or presenting results, or simply for a better overview, a smoother appearance of the spots its often preferred. Delta2D includes the option to model spot boundaries within the process of the spot detection. Simply check the box Create Modeled Spots when defining the parameters for spot detection and quantitation. (Fig.6.4).


imageprocessing/spots_square  imageprocessing/spots_round
Figure 6.4: The same region with pixel based and model based spots.

_________________________________________________________________________________________________________________________________________________________
Note:
The spot boundaries define the relevant area for spot quantitation. Therefore, whether you decide for modeled spots or not, the spot boundaries you face determine the quantities for the spots on your gel images. To achieve comparable spot quantities for your analysis, we strongly recommend to decide for one type of spot shapes for the entire experiment.

You have access to the option in the dialog for the spot detection parameters only. The option will be used for the next spot detection, existing spots on other gel images will not be affected. For changing the spot shape a re detection by using the altered parameter is necessary. "Keep atttributes" preserves already done classifications like hidden, canceled, exclude from normalization, etc..

6.5 Edit Spots

You can correct the results of Delta2D's automatic spot detection by setting "markers". Using markers you can control where a spot should be detected; Delta2D will then compute the new boundary accordingly. There are two basic operations for spot editing: creating a new spot, and joining two or more spots. In any case, Delta2D will compute spot boundaries automatically, using your input. Delta2D's approach to spot editing maximizes reproducibility while giving you a lot of control over which spots are detected.


ui/dialog_edit_spots_scaled
Figure 6.5: Edit spots

Adding a Spot

To add a spot, click on the Spot Editing Tool icons/spotedit-tool in the tool bar of the Gel Image Pair View. Now click on the position in your gel image where the spot should be detected, trying to hit the darkest point of the aspired spot. Where you clicked, a "+" will appear and according to this manually added marker a new spot will be detected instantly. If the result is not satisfying, you can either move the marker by dragging it, or you can remove the marker by right-clicking on it and set a new marker somewhere else. Your manually added spots will look and behave exactly like the automatically detected spots but still keep their marker as manually added, visible when switched to the Spot editing tool. Thus reproducibility is granted and at any time you have the possibility to edit them again.

In some cases it can be necessary that you drag a spot marker instead of setting it by a click. The dragged line markers have some influence on the spot shape and help for the correct detection of by gel breaks separated by spots.

Splitting a Spot in Two

To split a detected spot in two parts, simply override the detected spot with two manually set detection markers: select the edit spots tool as described above. Now click on the two sections of the spot you want to divide, trying to hit the centers of the aspired spots.

Joining two Spots

If a spot mistakingly was detected as two spots, you can easily join the two spots: switch to the edit spots tool, and drag a line from one to the other half of the aspired spot. Delta2D will join the two spots touched by the line.

Removing a Manually Edited Spot

To remove an edited spot, choose the edit spots tool as described above. Now right click on the marker you want to remove.