diff --git a/create_cvsplit.m b/create_cvsplit.m
deleted file mode 100644
index 11588b4..0000000
--- a/create_cvsplit.m
+++ /dev/null
@@ -1,74 +0,0 @@
-%  [tr_split,te_split] = create_cvsplit(Y,nFolds)
-%
-% Creates a Cross Validation split for the labels Y
-% In: 
-%  'Y' - labels
-%  'nFoldsK' - how many splits to create
-%
-% out: 
-%  'tr_split','te_split' - cell array of size [nFolds,1] with indices of
-%  points in training and test splits
-%
-% Peter Gehler, August 27th 2008
-% pgehler@tuebingen.mpg.de
-function [tr_split,te_split] = create_cvsplit(Y,nFolds)
-
-
-for f=1:nFolds
-    tr_split{f} = [];
-    te_split{f} = [];
-end
-
-classes = sort(unique(Y));
-
-for c=1:numel(classes)
-    nExamples = sum(Y==classes(c));
-    
-    if nExamples < nFolds
-	error('less examples than folds');
-    end
-    
-    nPerFold = floor(nExamples/nFolds);
-    
-    ind = find(Y==classes(c));
-    ind = ind(randperm(numel(ind)));
-    
-    for f=1:nFolds-1
-	te_ind = ind( (f-1)*nPerFold + (1:nPerFold))';
-	tr_ind = setdiff(ind,te_ind);
-	
-	if size(te_ind,1) ==1, te_ind = te_ind'; end
-	if size(tr_ind,1) ==1, tr_ind = tr_ind'; end
-
-	tr_split{f} = [tr_split{f};tr_ind];
-	te_split{f} = [te_split{f};te_ind];
-    end
-    
-    te_ind = ind( (nFolds-1)*nPerFold+1:end)';
-    tr_ind = setdiff(ind,te_ind);
-
-    if size(te_ind,1) ==1, te_ind = te_ind'; end
-    if size(tr_ind,1) ==1, tr_ind = tr_ind'; end
-    
-    tr_split{nFolds} = [tr_split{nFolds};tr_ind];
-    te_split{nFolds} = [te_split{nFolds};te_ind];
-
-    if (all(size(tr_split{nFolds})~=1))
-	error('dimension error');
-    end	
-    if (all(size(te_split{nFolds})~=1))
-	error('dimension error');
-    end	
-end
-
-for i=1:nFolds
-    if size(tr_split{i},1)==1
-	tr_split{i} = tr_split{i}';
-	assert(numel(tr_split{i})<numel(Y));
-    end
-    if size(te_split{i},1)==1
-	te_split{i} = te_split{i}';
-	assert(numel(te_split{i})<numel(Y));
-    end
-end
-
diff --git a/plot_heatmap.R b/plot_heatmap.R
index 9519d28..8256b90 100644
--- a/plot_heatmap.R
+++ b/plot_heatmap.R
@@ -1,4 +1,6 @@
 
+#Reusing this R script from an earlier project.
+
 args <- commandArgs(TRUE) #send cellline, region#, oligoLen
 #send inputFilename outputFolder rank oligoLen
 inputFilename <- args[1]
diff --git a/predictTestExamples.m b/predictTestExamples.m
deleted file mode 100644
index 1b7c27d..0000000
--- a/predictTestExamples.m
+++ /dev/null
@@ -1,31 +0,0 @@
-function yhat = predictTestExamples(allSeqsAsBagsNonShifted_Test, allSeqsNonShiftedTransformationKernel_Test, allSeqsAsBagsShifted_Test, allSeqsShiftedTransformationKernel_Test, nBagsForTest, nClusters, weightVectors, biasValue)
-    
-    yhatTemp = 0.0;
-
-    % bagPhi for Test is nfeatures-by-nBagForTest-by-nClusters
-    bagPhiNonShifted_Test = zeros(  size(allSeqsAsBagsNonShifted_Test{1},1), nBagsForTest, nClusters);
-    bagPhiShifted_Test = zeros( size(allSeqsAsBagsShifted_Test{1},1), nBagsForTest, nClusters);
-    %
-    % For normalization of test samples, we need to obtain the normalization factor for them.
-    % 
-    for i=1:length(allSeqsAsBagsNonShifted_Test)% this is nBagsForTest
-        for j=1:nClusters
-            %allSeqs NonShifted/Shifted TransformationKernel already has nInstances x nClusters 
-            % ALSO NEED NORMALIZATION
-            term = sum(allSeqsAsBagsNonShifted_Test{i} .* repmat(allSeqsNonShiftedTransformationKernel_Test{i}(:,j)', size(allSeqsAsBagsNonShifted_Test{i},1), 1), 2);
-            bagPhiNonShifted_Test(:, i, j) = term ./ norm(term, 2);
-            term = sum(allSeqsAsBagsShifted_Test{i} .* repmat(allSeqsShiftedTransformationKernel_Test{i}(:,j)', size(allSeqsAsBagsShifted_Test{i},1), 1), 2);
-            bagPhiShifted_Test(:, i, j) = term ./ norm(term, 2);
-        end
-    end
-    fprintf('Bag Phi for test samples done.\n');
-    for i=1:nClusters
-        fprintf('Samarth yhat...\n');
-        yhatTemp = yhatTemp + (weightVectors(:, i)' * bagPhiNonShifted_Test(:, :, i));
-    end
-    for i=(nClusters+1):2*nClusters
-        yhatTemp = yhatTemp + (weightVectors(:, i)' * bagPhiShifted_Test(:, :, (i-nClusters)));
-    end
-    fprintf('Bias value: %.4f\n', biasValue);
-    yhat = yhatTemp + biasValue;
-end % function ends