Figure 9From: A prediction-based resampling method for estimating the number of clusters in a datasetCorrelation matrix, NCI60 dataset. Images of the correlation matrix for the 61 cell line mRNA samples based on expression profiles for (a) all p = 5,244 genes and (b) the p = 200 genes with the largest variance. The mRNA samples are ordered by class: 7 + 2 breast, 6 CNS, 7 colon, 6 + 2 leukemia, 8 melanoma, 9 NSCLC, 6 ovarian, 8 renal.Back to article page