This study integrates the methods from both gene association studies and gene expression studies to study association trajectories. The study by Stead et al. (2006) of brain development in the male rat provides an exemplary model for comparison. The primary correspondence is between the gene expression levels in Stead et al. (2006) and the effect sizes in this study. Although effect sizes are signed and gene expression levels always positive, in gene expression studies measures of the difference from the mean are often the feature of interest. Trajectories of gene expression are shown in Stead et al. (2006, Figure 5) and PCA plots in Stead et al. (2006, Figure 2). While a single PCA in Stead et al. (2006) was sufficient to represent the variation in gene expression in both age (along the x-axis) and region (along the y-axis), in this study the complexity of the association relations necessitated the use of age specific PCAs, as can be observed in Figure 6 of the main document. The specific statistical methods described here are meant to deal with the situation of
