There are no facts, only interpretations.
Friedrich Wilhelm Nietzsche

Image Analysis

***The microarrays contain 29,000 70-mer oligonucleotide elements for Arabidopsis thaliana; these slides were produced at the University of Arizona. More information about the slides, including description of the gene loci, can be found at Platform GPL4570.***

Opening the Image

  1. Open GenePix Pro or GenePix Pro Demo.
  2. Click the Open/Save button on the right-hand side of the window.
  3. Select Open image and click on your image file from the desktop folder.

Loading the Array List

GenePix Pro automatically generates BLOCKS using information contained in a GenePix Array List (GAL) file: these files are used to link the information from the arraying process to the analysis; substance names and identifiers are assigned to each spot (feature).

  1. Click the Open/Save button on the right-hand side of the window.
  2. Select Load Array List.
  3. Go to the Bios 313 folder on the desktop.
  4. Open ATv3.5.2.Z_Locus.gal.

Limit yourself to ONE HOUR for aligning blocks and flagging features.

Aligning Blocks on an Image

View Blocks is automatically selected when you load an array list.
  1. Move ALL the blocks to the general area of the array.
  2. "Unselect" the blocks.
  3. Zoom in and MOVE ONE BLOCK AT A TIME.
    Using the mouse, drag and place each block over a block of spots.

    You will have to position EACH block (32 total); for FINE TUNING blocks, see p. 53 in the GenePix Pro User's Guide.

    It's OK if some of the BOXES overlap when you're positioning the blocks; what's more important is for the CIRCLES inside the boxes to line up with the SPOTS on the array.
  4. Select Feature Mode from the Tools group on the left-hand side of the Image tab; Align Features in ALL BLOCKS.
    *go to "option" ---> "alignment" ---> change to 100 minimum diameter (default = 33%)

Flagging Feature-Indicators

Spend some time looking at the entire slide to flag manually obvious problems, such as:

Some features should be ignored during the analysis because of imperfections on the array. Sometimes, the software marks or flags "Bad" features as "Good" (or vice versa). You may need to flag individual feature-indicators: just flag BAD features; leave all GOOD features unmarked.

Examining the features will be the most tedious part of getting your array image ready for analysis. You do NOT have to flag ALL the spots; just look at each block and make sure that the spots are marked correctly.

  1. To flag a feature, select Feature Mode from the Tools group on the left-hand side of the image.
  2. Click on the feature you want to flag.
  3. Hit "A" to flag the feature as "Bad."

    NOTE: You can also select a region to flag as bad by dragging a "square" around the features.
You may also need to increase (left cursor arrow)/decrease (right cursor arrow) the size of individual spots.

Saving the Positions of Blocks and Feature-Indicators

After you have finished aligning your array and flagging the feature-indicators, SAVE your configuration:
  1. Click Open/Save on the right-hand side of the window.
  2. Select Save Settings As.
  3. Enter a file name for the settings and click ok.

    Saving this information lets you come back later to perform the analysis.

IMPORTANT NOTE: GenePix Pro uses a local background subtraction method by default; this method is more responsive to background intensity variations across a slide. A different background is computed for each individual feature-indicator, and the median background value is subtracted from the feature intensity before any ratios are calculated (see GenePix Pro User's Guide, pp. 13-15).

Performing the Analysis and Viewing the Results

  1. Click on the Analyze button ("BCR").

    A JPEG file of the image is saved AUTOMATICALLY.

    To SHOW ONLY the UNFLAGGED ROWS, click the Display button in the Table group on the left-hand side of the Results tab and check the unflagged box.
  2. Switch to the Scatter Plot tab to generate a graph of the analysis measurements for all spots on the array.
  3. On the left-hand of the Scatter Plot tab, choose quantities to plot along the X and Y axes:
  4. Position the mouse over any spot on the scatter plot; the associated feature will be displayed in the Feature Viewer.
    The Feature Viewer reports the intensity of the pixels within the spot, its associated background intensity level, the precise location of the spot in the analysis array, and substance IDs or names that have been defined for each feature.

    The Scatter Plot and the Results spreadsheet are fully integrated: after choosing a spot on the scatter plot, select "Group Rows" in the Results spreadsheet; the details about the selected feature will appear at the TOP of the list.

Saving the Data


  1. Click the Open/Save button on the right-hand side of the window.
  2. Select Save Results As.
    NOTE: The default file format is *.gpr.
  3. Name the file and click OK.
  4. Next, save your customized view of your Results:
    a. Click the Open/Save button.
    b. Select Export Results.
    c. Name the file and click OK.
Exported Results are saved as PLAIN TEXT FILES (*.txt) and cannot be opened in GenePix Pro; open the file in Excel or Word (it's a tab-delimited text file).

Identifying the Genes

Your analysis will generate a vast amount of data; your analysis is not expected to be exhaustive. From the scatter plot of the median intensities, choose 5 GOOD spots (i.e., ones that are nicely rounded and look "real") that represent a several-fold increase or decrease in gene expression. Focus further analysis on 3 of these genes.

Describe the function of the 3 genes you selected for analysis: When are they normally expressed? How are they regulated? Are the genes related? Do mammals express these genes?

Think about the significance of your findings and comment on how they may explain the phenotype. Propose two experiments that will help answer your questions.

Check out VirtualPlant:

Copyright, Acknowledgements, and Intended Use
Created by B. Beason (bbeason@rice.edu), Rice University, 13 August 2002
Updated 9 March 2007