Microarray Data Analysis

Using Spot Image Analysis Software

This is a brief tutorial on using Spot in a Windows environment. For more details, see the official Spot User's Guide.

Spot is a program for extracting foreground and background intensities for individual spots on a microarray image. Spot is implemented as a library in the statistical program R.

Spot assumes that you have a series of arrays which you will want to process, all of which have a layout so similar that you can use the same template information for all of them. Spot calls this a batch of arrays. You will need to have some short name to identify this particular batch of arrays. In this tutorial, I will assume that the batch is called "mybatch".

To process your images using Spot, follow these steps:

Run R and load Spot

1. Put your TIFF images in a convenient directory. You will need to have separate TIFF images for the cy3 and cy5 channels of each array.
2. Run R. This will bring up a windows interface containing an R Console into which you can type commands.
3. Using the drop-down menu in R, change the current directory to the directory containing your TIFF images (File > Change Dir, then browse for your directory) .
4. Load Spot using the command library(Spot). The prompt in your R session will change to "Spot> ".

Create the images file

The images file is a tab-delimited text-file called "images.mybatch" giving the names of the files containing the TIFF images. The first line contains the column headings "R<tab>G" or "Red<tab>Green". Subsequent lines give the red (cy5) and green (cy3) TIFF file names for each array. The file might contain for example:

R                 G	
mybatchR1.tif     mybatchG1.tif
mybatchR2.tif     mybatchG2.tif
mybatchR3.tif     mybatchG3.tif

It is usually easiest to create this file directly yourself using a text editor, but you can also create it using the command SetImages("mybatch") in R. In the latter case, R will prompt you with a data editor spreadsheet into which you can type the file names.

Create the parameters file

The parameters file is a text file called "parameters.mybatch" containing an R command which specifies the layout of each array, i.e., the number of pin groups and the number of spots in each pin group. This is most easily created using the R command SetParameters("mybatch") which will prompt you for the required parameters. This will write the parameters file into your current directory containing something like the following:

list(nspot.r = 15,      # Number of rows of spots per grid
     nspot.c = 15,      # Number of columns of spots per grid
     ngrid.r = 8,       # Number of rows of grids per image
     ngrid.c = 4,       # Number of columns of grids per image
     tolerance.r = 50,  # Top/bottom translation tolerance
     tolerance.c = 50   # Left/right translation tolerance
)

The values given above for the two tolerance parameters are satisfactory for most arrays.

Create the template file

This is the most important step in using Spot, and the one which calls for most interaction from you.

The template file is a text file specifying the location of each grid or pin group on the TIFF image. It is assumed that the one template file will do all for the images in the batch.

1. Enter the command SetTemplate("mybatch",1). This will start an ImView window displaying your chosen image. (The number 1 specifies which of your images you would like to use to create the template for all the others. I have assumed that you will use the first image, but you can specify any one of the images listed in your images file. The number corresponds to the order of the images in the images file.)
2. Using the drop-down menu in ImView to switch to "add point mode" (Transform > Pointfile > Add point mode).
3. Left-click in the centre of the top left spot of each pin group, starting from the top left and going down by rows. You can use alt-D to undo your last click if you click somewhere by mistake.
4. Left-click in the centre of the bottom right spot of the last pin group. This click is especially important because it is used to estimate the height and width of all the pin groups.
5. Go back to the R Console window and press ENTER.

This writes the template file into your current directory. This file specifies the row and column positions of the top left of each pin group and the width and height of the last group.

Analyse each image

Now you're in a position to run Spot on each of the your images. The command

mybatch1 <- Spots("mybatch",1)

will create a matrix named "mybatch1" in your R session containing the red and green intensities for each spot on the first array. (The name which you choose for the matrix is arbitrary.) The ImView window will show the grid-lines for your image and then the segmented foreground regions for each spot. The columns of the matrix contain various information about each spot. See Spot: Description of Output for the meaning of each column.

You can continue with

mybatch2 <- Spots("mybatch",2)
mybatch3 <- Spots("mybatch",3)
...

until you have processed all your images. This will create a new matrix for each image.

Save your results

You will almost certainly want to save your results for future use. You can save all the R objects containing the results into an R data file for future use using save.image("mybatch.RData"). This will create the file "mybatch.RData" in your current directory.

It is also a good idea to create text files containing the Spot results for each your images. This can be done using

write.table(mybatch1,file="mybatch1.spot",quote=FALSE,sep="\t",row.names=FALSE)

and so on for each image in your batch. This will create files "mybatch1.spot" and so on in your current directory.