Fig. 2
From: On the relevance of technical variation due to building pools in microarray experiments
Estimates of blending error variance for simulated data. Log-estimates of the blending error variance \({\sigma _{2}^{2}}\) (left) and p-values (right) of RLRT (\(\mathbf {H}_{0}:{\sigma _{2}^{2}}=0\)) for simulated data. Top: perfectly blended individuals were simulated. The p-values of the interval [0,0.5) are uniformly distributed and nearly half of the transcripts have a p-value of 1. Middle: 3000 out of 9000 transcripts affected by imperfect blending of individuals. Bottom: all transcripts were simulated with imperfect blending