Fig. 3: True positive detection rates under the “GxE” scenario.
From: Assessment of two statistical approaches for variance genome-wide association studies in plants

True positive detection rates for the simulated traits under the “GxE” scenario settings at a false discovery rate of 0.05 for A Arabidopsis at a sample size of N = 500, B maize at a sample size of N = 500, C Arabidopsis at a sample size of N = 1087, and D maize at a sample size of N = 2815. On each figure, the X-axis represents the QTN, and a description of the which environment(s) in which they were simulated are detailed in the corresponding X-coordinate label. The Y-axis is the proportion of replications where a true positive is detected at least once. The error bars depict 95% confidence intervals, calculated using the method from Clopper and Pearson (1934). BFT Brown–Forsythe test, DGLM double generalized linear model, MLM unified mixed linear model.