% 
% Written by:
% -- 
% John L. Weatherwax                2005-08-14
% 
% email: wax@alum.mit.edu
% 
% Please send comments and especially bug reports to the
% above email address.
% 
%-----

close all; drawnow; 
rehash; 
clc; 

clear; addpath( '../../Code/eda_data' ); addpath( '../../Code/eda_toolbox' ); addpath( '../Chapter1' ); 

% Example 8.1

% First load the data.
load leukemia
[n,p] = size(leukemia);
x = zeros(n,p);
% Standardize each row (gene) to be mean
% zero and standard deviation 1.
for i = 1:n
    sig = std(leukemia(i,:));
    mu = mean(leukemia(i,:));
    x(i,:) = (leukemia(i,:) - mu)/sig;
end
% Do hierarchical clustering.
Y = pdist(x);
Z = linkage(Y,'complete');
[H, T] = dendrogram(Z,15);
title('Leukemia Data')

% Example 8.3
% Use same leukemia data set 
% and matrix Z from Example 8.1.
% The second argument is the dendrogram distance. The
% third argument is a string
% specifying what information
% the second argument provides.
%%%%%rectplot(Z,15,'nclus')
rectplot(Z,13.0,'dis');

% We now see how this rectplot shows the original cancer type.
% 
labs = zeros(length(cancertype),1);
% Now find all of the AML cancers.
% Set them equal to 1.
inds = strmatch('AML',cancertype);
labs(inds) = 1;
% Now do the rectangle plot.
%%%%%rectplot(Z,15,'nclus',labs)
rectplot(Z,13.0,'dis',labs)
saveas( gcf, '../../WriteUp/Graphics/Chapter8/prob_8_4_rectplot', 'epsc' );