How to split classes of a database in two ?

I have this code:

   %%Loading data
    load('wine.data');
    % first column stores the wine class according to wine.names file
    nClass=max(wine(:,1)); 
    %%Getting the mean of each class for the 13 parameters
    meanEachClass=arrayfun(@(x) mean( wine( wine(:,1)==x ,2:end) ), 1:nClass,'UniformOutput',false);
    %%Now checking the euclidean distance of a sample 
    % relative to the mean of each class
    nSampleToTest=10;
    truePositives = zeros(1, nSampleToTest); % Initialize
    for i=1:nSampleToTest
      % Randomly choosing a sample
      sampleNo=randi(size(wine,1));
      sample=wine(sampleNo,2:end);
        % calculate the Eudlidian distance to each class.
        distances=arrayfun(@(x) norm(sample-meanEachClass{x}), 1:nClass, 'UniformOutput',true);
        disp(sprintf('Sample #%d',sampleNo))
        disp(sprintf('Distance: \n   Class 1: %f \n   Class 2: %f \n   Class 3: %f \n',distances(1),distances(2),distances(3)));
        disp(sprintf('Based on distance, Sample seems to belong to class %d\n', find(distances==min(distances))))
        disp(sprintf('According to the database, sample belongs to class %d\n',wine(sampleNo,1)))
    [~, myClass] = find(distances==min(distances));
    trueClass = wine(sampleNo,1);
    if myClass == trueClass
         % Accurately determined the class.
         truePositives(sampleNo) = 1
    end
      end
    overallAccuracyPercentage = 100 * sum(truePositives) / nSampleToTest;

Here is the database: http://archive.ics.uci.edu/ml/machine-learning-databases/wine/ This code calculates the nearest prototype by using the euclidian distance. The problem is here that my train vectors are represented by the all vectors from database and my test vectors are randomly taken. I want to split in 2 each class and the first half to represent the train vectors and the other the test vectors. For example class 1 has 59 vectors: 30 train vectors and 29 test vectors and so on with the class 2 and 3. Then i want to be able to calculate the percentage of error. Can you provide me some code,please?