Balasubramanian Narasimhan
Prediction Analysis of Microarrays in Excel
*******************************************

Prediction Analysis of Microarrays (PAM) is a statistical technique
for class prediction using gene expression data using shrunken
centroids. It is described in Tibshirani, Hastie, Narasimhan, and
Chu (2002).  The method of nearest shrunken centroids identifies
subsets of genes that best characterize each class. The technique
is general and can be used in many other classification problems.

PAM Software for the R has been available for some time now from the
http://www-stat.stanford.edu/~tibs/PAM. Using the software, one can
train the classifier, perform cross validation to get an idea of the
value to use for thresholding, and do predictions. There has been
some demand for a GUI version for PAM.

Our experience with SAM (Significance Analysis of Microarrays, Tusher,
Tibshirani, and Chu, 2001) software led us to use Excel as a GUI for
the package. SAM was written in Visual Basic, using a Java COM library
at the core. However, it was clear that this approach would not be an
efficient one in the long run for several reasons.
Among them:
  o Many statistical tools would have to be rewritten in Java, or
    Visual Basic wasting valuable development time rather than
    leveraging already available tools.
  o The outcome of the market place battle, Microsoft Java versus
    Sun Java, is hardly clear. Besides, programming in Microsoft Java
    is like programming with one hand tied behind your back.
  o While Microsoft's commitment to DCOM seems quite firm, its
    commitment to Java COM seems to have stalled.

At DSC 2001, Neuwirth and Baier (2001) demonstrated an R DCOM server and
client. We decided to exploit this technology for PAM to produce an
environment much like that of the SAM except with R as the computation
engine.