Figure 1From: Robust assignment of cancer subtypes from expression data using a uni-variate gene expression average as classifier Distribution of AUC values for all genes in a dataset. AUC values in respect to MI-Status are plotted in solid lines for SanchezC data in A) and Stransky data in B). AUC values in respect to high grade (G3) are plotted in solid lines for SanchezC data in C) and the Stransky data in D). Distributions of AUC values for randomized versions are plotted in dashed lines and the intervals that contain 99% of randomized data are indicated with vertical dashed lines.Back to article page