Prediction Analysis of Microarrays in Excel: Installation Guide and Manual

Robert J. Tibshirani, Trevor J. Hastie, Balasubramanian Narasimhan, and Gilbert Chu

$Revision: 1.2 $ of $Date: 2003/05/16 05:21:06 $.
PAM is a technique for sample classification from microarray data. This new package features an Excel interface, designed and written by Balasubramanian Narasimhan, with help from Trevor Hastie and Rob Tibshirani. It uses R functions written by Trevor Hastie and Rob Tibshirani.

1. Introduction

PAM implements the Nearest Shrunken Centroid method of Tibshirani, Hastie, Narasimhan and Chu (May 14, 2002) described in Diagnosis of multiple cancer types by shrunken centroids of gene expression.

Note that there are a number of steps in the installation but all are simple and steps 1-4 need to be done only once. They involve installation of the R package, and the pamr package, both of which are used by PAM for Excel.

After these steps, PAM can be used from within your version of Excel.

This document describes the installation and use of the PAM for R package. This document is always kept up-to-date at The PAM Page.

2. Installing PAM for Excel

  1. You first need to install a recent version of the R statistical package. This is free, and can be found at The R Project.

    Follow the instructions. Click on CRAN under Download. Pick a mirror closest to you. You probably want a pre-compiled version. For windows, choose the Windows (95 or later) link, select base, and download SetupR.exe. Installation takes 5-10 minutes.

    R is a great package, and is worth knowing about!

  2. Download the Windows version of pamr from the PAM for R Website. Note that PAM for Excel and PAM for R are matched packages and one should therefore make sure that same versions of both packages are installed, for example, version 1.12. When you download the R package, most browsers offer you a choice of opening the file or saving it. Elect to save it and remember the name of the folder where you saved it.

    Double click on the R icon on your desktop, pull down the Packages menu item and select Install packages from local zip file. A file chooser dialog will allow you to select the file pamr_1012.zip that you just saved in the previous step.

  3. If you have an older version of PAM for Excel already on your computer, remove it using Add/Remove Programs in your control panel, before installing the newer version.

    Go to the PAM for Excel Registration page and then follow instructions to download the pam_1012.zip file. Unzip it and click on the Setup icon to install it.

  4. Fire up Excel, go to the Tools menu and choose Addin. Browse to find the Addin file that the installation in the previous step installed. Unless you changed the default on installation, it is usually placed in the directory
    C:\Program files\PAMVB\Addin\pam.xla
    Click OK, and now you should see a PAM and PAM controller button on the top of your Excel window. You might have SAM installed already- that's OK! They can co-exist in your Excel.
The above concludes the installation process and needs to be done just once.

3. Using PAM for Excel

The data setup for PAM is much like that for SAM. There is one line per gene. There are additional rows at the top of the spreadsheet for class labels and (optional) sample and batch labels. Example files have been provided for you. For example the file

C:\Program Files\PAMVB\Examples\khan.xls

Load up this example, click on the PAM button and you should see a PAM Dialog. Specify the following inputs:

Click OK to start the PAM analysis.

4. Documentation

A detailed manual for PAM is available in the directory

C:\Program Files\PAMVB\doc\pam.pdf

It also contains the above installation instructions and a lot more detail. Any serious user of PAM should refer to it.