Frequently Asked (or Anticipated) Questions

Revised 2 August 2006

Is a user manual available?

Sorry, there's no user manual. We suggest you read this FAQ page and the results section of the GelBuddy paper Nucleic Acids Res. 33(9):2806-2812.

What are the basic steps required to analyze a gel?

	Generate lane tracks by selecting the show lanes tool.
	Adjust the automatically generated lane tracks if necessary by using the Edit Lanes Mode tool.
	Use the Show Calibration Information tool to view the calibration ladder and de-smiling curves.
	Adjust the calibration ladder by dragging the 0%, 700bp, 200bp, and 100% markers at the left margin to match image features.
	To start recording signals, use the Record Signals Mode tool.

Record Signals mode offers several options for recording and deleting signals:

Click on a band to record a single signal.
Click and drag to record co-migrating signals in consecutive lanes.
Click and drag an existing signal to record a co-migrating copy in another lane.
Shift-click and drag an existing signal to record co-migrating copies in consecutive lanes.
Alt-click and drag an existing signal to delete co-migrating copies in consecutive lanes.
Alt-click an existing signal to delete it.

You can also delete a signal by pressing the DELETE key while the cursor is located over the signal.
Use the Control key in combination with the above options to copy or delete all signals from one lane. For example:
Use the ESC key while dragging a signal to cancel the current operation.
Use the right mouse button to alternate between 700 and 800 channel images.

Can GelBuddy automatically detect and record image bands?

Yes. The paper Till, B. J. et al. (2006) High-throughput discovery of rare human nucleotide polymorphisms by Ecotilling. Nucleic Acids Res (in press) describes an algorithm developed for this purpose. An overview of this algorithm is presented in GelBuddy's Automatic Signal Detection Algorithm. A description of parameter adjustments (including an interactive threshold adjustment, not used in the paper) is provided in Using Automatic Signal Detection in GelBuddy.

Automated band detection is most appropriate for analysis of gel images containing a large number of cleavage product bands. For images containing only a few bands (such as TILLING images used to screen a mutagenized population) manual band selection by an experienced operator is more efficient. The automated band detection algorithm performs best if fragment length calibration is accurate and each cleavage product is visible in both channels.

Calibration doesn't behave as described in the GelBuddy paper Nucleic Acids Res. 33(9):2806-2812. What's changed?

By default, the current version of GelBuddy constructs three independent de-smiling curves from image data, at the location of the lower (200bp) marker, the location of the upper (700bp) marker, and the location of the 0% marker (full length product). You can change this behavior on a per-markup basis by using the menu commands Options/Calibrate Lower Size Standard Using Image Data, Options/Calibrate Upper Size Standard Using Image Data, and Options/Calibrate Full Length Product Using Image Data, respectively.

In many cases, it is possible to obtain better calibration results by using the IRDye 800 image for calibration by using the menu command Options/Use 800 Channel Image Data For Calibration.

What information should I enter into the Gel Information window?

The Gel Information window prompts the user for information needed for analysis and reporting.

[GelBuddy Gel Info Dialog]

Run Name
This name is used when you save a markup (.XML) file or post results over the web.
Created by
This is used for annotation of markup (.XML) files and posted results only. You may enter any text you wish.
Size of full length product
GelBuddy uses this number to match IRDye 700-labeled fragments with complementary IRDye 800-labeled fragments. Note that this number need not be the same as the full-length product length identified by the 0% marker at the left margin of the Gel Image.
Signal grouping
Selecting the "Every 16 lanes" option restricts grouping of co-migrating signals to sets of 16 adjacent lanes. For example, a signal in lane 2 may be grouped with a lane in lane 14 but cannot be grouped with a signal in lane 17. The "All Lanes" option is appropriate for most applications.

How can I improve the performance of GelBuddy on my images?

You will achieve the best results if you create images similar to those on the Reference Images page. The High-througput TILLING paper Plant Physiol. 126: 480–484, Figure 2, describes the running conditions and comb-loading procedures used. However, GelBuddy has several settings you can change to optimize analysis of your images. A few settings, such as lane count and straight/segmented lane track selection, are selected by the user whenever GelBuddy generates new lane tracks. A number of parameters affecting lane tracking and size calibration can be configured in the Global Configuration window, accessible by menu command Edit/Alter Global Configuration. Most of these parameters take effect after loading a new set of gel images, and don't affect the currently loaded image or existing markup (.XML) files. GelBuddy has not been tested for all possible values of these parameters: alter these settings with caution.

[GelBuddy Global Configuration Dialog]

Lane finder comb spacing filter constant
This number, together with the lane count, determines the minimum expected spacing between lanes. Decrease this number if you get many double lane tracks; increase it if GelBuddy seems to be skipping lanes.
Lane finder image zone count
This is the number of horizontal "slices" that GelBuddy uses when generating segmented lane tracks. Increasing this number allows more horizontal deviation in lane tracks but increases the number of errors caused by band detection failure.
Lane finder missing lane insertion threshold
GelBuddy inserts lane tracks to compensate for "gaps" caused by loading and reaction failure. Decreasing this number increases the likelihood of inserting such tracks.
Upper size standard
This specifies the size of the upper calibration standard. For best results, this should be less than or equal to the size of the full length product. Users TILLING in short amplicons will probably need to adjust this setting.
Lower size standard
This specifies the size of the lower calibration standard.
Upper marker de-smiling limit
This specifies the amount by which the vertical position of the 100% and upper marker de-smiling curves may vary between adjacent lanes. Increasing this number allows for correction of more distorted gels, but is also more likely to produce incorrect results if strong signals are present.
Bottom marker de-smiling limit
Similar to the upper marker de-smiling limit, but for the lower marker de-smiling curve. Because the bottom of most TILLING gel images contains a large number of sporadic signals and relatively little distortion, the default value for this parameter is smaller than the default value for the upper marker de-smiling limit.

What types of images can GelBuddy process?

GelBuddy can process two 8-bit or 16-bit grayscale images, in TIFF or JPEG format, subject to the following conditions:

The two source images must be equal in width and bit depth.
The two source images must be approximately equal in height. There is no explicit check for this, but many functions including size calibration will behave erratically if the source images differ significantly in height.
16-bit images (typically TIFFs) will be negated and inverted when loaded. 8-bit images (typically JPEGs) will not.
GelBuddy is typically used with images with an approximate size of about 8 megapixels. Significantly larger image files may take a very long time to load. If your gel scanner is producing excessively large images, you may need to change its configuration.

Loading corrupt images will lead to unpredictable results.

How can I convert TIFFs to JPEGs?

We recommend the convert utility provided by the ImageMagick package (available at http://www.imagemagick.org). Our image archiving system uses the command:


convert -flip -negate -quality 80 name.tiff name.jpg

GelBuddy automatically flips and negates all TIFF (16-bit) images, and does not flip or negate any JPEG (8-bit) images.

How can I capture or print the GelBuddy window?

The most convienient way to capture or print the GelBuddy window is to use the screen capture facility of your computer's operating system to copy a screen image to the clipboard and then paste, crop, and save the image using Photoshop or a similar image editing program. Here's how we created the screen shots for the NAR paper:

Start GelBuddy and click the "maximize window" box at the upper right corner (next to the "X")
Load the XML file you wish to view. Turn off anything you don't want to appear in the markup; for example:
- click the "show lanes" button to turn off lane tracks
- click on the "show calibration information" button to turn of de-smiling curves
- select Options/show lane number to turn off the lane numbers at the top of the screen
- select Options/show status bar to turn off the status bar at the bottom of the screen
Zoom to the region of interest; The "Zoom/Zoom to coordinates" function is handy if you wish to do this reproducibly.
Press the Control, Alt, and Printscreen keys simultaneously to place your image on the clipboard
Open Photoshop
Select File/New; a window with the "clipboard" preset should appear.
Select "CMYK Color" and "8 bits" for color mode -- journals don't like RGB images
Click "OK"
Select Edit/Paste; your screenshot will appear in the newly created window
Select View/Actual pixels; this will show the uncompressed image
Make sure the View/Snap is turned off (this makes it easier to use the crop tool)

select Window/Actions to reveal the "Actions" window
select the "Actions" tab of the "Actions" window
click on the "New action" button (it's the icon on the bottom of the "Actions" window next to the trashcan)
enter a name for your new action (such as "crop") and click "record"
click on the crop tool (on the main image menu) and drag out a rectangle on the image to select the region you wish to show
press "Enter" if you are satisfied with the cropped region, or "ESC" if you want to try again
to stop recording, click on the square "stop" button on the bottom left of the "Actions" window

To save the file:
- File/Save As. . .
- save in TIFF format
- enter an image name
- uncheck the "layers" checkbox
- click "save"
- make sure "Image Compression:NONE" is selected in the TIFF options window and click "OK".

I've installed a previous version of GelBuddy. Do I need to remove it before installing a new version?

No. You can overwrite your existing copy of GelBuddy (recommended) or simply install the new version into a different location. Multiple GelBuddies can peacefully coexist on a single machine. Preference settings are stored on a per-user basis but will be shared between all instances of GelBuddy run by that user.
If you decide to keep multiple versions on one machine, you should check the version number on the startup screen to make sure you're running the most recent version.

What's the difference between GelBuddy for Macintosh OS X and GelBuddy for Microsoft Windows?

Not much. The code and Java Archive (JAR) used by both versions are identical. The Windows distribution is a Zip archive containing the raw .JAR file and a batch file to specify the proper VM settings. The Macintosh distribution is a disk image (.dmg) file containing an application bundle, which in turn contains VM settings, the application icon, and other data needed to run GelBuddy.

When GelBuddy detects that it is being run under Mac OS X, it invokes platform-specific code from AppleJavaExtensions.jar (published by Apple) to implement Macintosh-style menu commands and other user interface elements.

Can I run GelBuddy under Solaris or Linux?

Possibly. We run GelBuddy under Windows and Mac OS X, but it should run under any platform that supports a sufficiently recent version of Java and the Java Advanced Imaging library. You can obtain the necessary Java Archive (JAR) file and command-line parameters from the GelBuddy for Windows download page. Imaging performance will be highly dependent on platform-specific optimization of the Java Advanced Imaging library.

How can I store and analyze data from a marked-up image?

GelBuddy only allows analysis of one gel at a time. Markup information is stored as an .XML file, and analysis of this information is available from the Report menu. To consolidate data from multiple sample sets or primer pairs, you'll need to create code to parse the .XML file to extract the necessary data. Do not attempt to automatically extract information from the text report, as the format of this report will change in subsequent versions of GelBuddy.

How does GelBuddy post results over the web?

GelBuddy simply sends username, password, and markup (.XML) data via an HTTP POST command to the server of your choice. The posted .XML data is exactly the same as that saved in a file.

The GelBuddy Project Home Page