Figure 1

Relationship between densities and ROC curves considering equal variances on both groups. Probability density functions of gene expression values of two groups and their corresponding empirical ROC curves, where Y is the random variable which represents the expression values under the experimental condition and X the random variable which represents the expression values for the control group. The same classification rule was considered for all ROC plots, namely, high values of the decision variable correspond to positive regulation. Density plots were obtained using kernel density estimation from two samples of size 100 simulated from normal distributions. A) X ∼ N(20, 4),Y ∼ N(30, 4); B) X ∼ N(20, 4),Y ∼ N(22, 4); C) X ∼ 0.5N(−20, 2) + 0.5N(20, 2),Y ∼ N(0, 11); D) X ∼ N(0, 11),Y ∼ 0.5N(−20, 2) + 0.5N(20, 2); E) X ∼ N(30, 4),Y ∼ N(20, 4).