Fig. 1From: Variable selection for binary classification using error rate p-values applied to metabolomics dataThe null cumulative distribution functions. The graphs show the log 10 transformed null CDFs of \( {\widehat{er}}_{up}^{*} \) (black lines) and \( {\widehat{er}}_{min}^{*} \) (red line), for group sizes N 0 = 21 and N 1 = 12 using weight sets 1 (solid lines) and 2 (dashed lines). The dark (α = 0.001) and light (α = 0.05) blue lines represent points of reference discussed in the textBack to article page