10 Fold Cross Validation Accuracy On Sample-by-Sample Transformation using <i>k</i>-Nearest Neighbours.

The results of 10-fold cross-validation on the dataset using sample-by-sample affinity matrices for PCA and Isomap. The a priori manifold learning method (which operates using a gene-by-gene affinity matrix) still provides comparable results with the other methods, while outperforming them in some of the cases. We have emphasised in bold the cases which a priori manifold learning outperforms the rest of the methods.