The relationship(s) between network connectivity and gene variability are not entirely clear. However, if the variance is “biological” and not technical or simply stochastic, it follows that variance and connectivity will increase in tandem; for the moderate sample sizes of most gene expression studies, gene–gene correlations and hence connectivity will be more easily detected. Colville et al. (2017) observed that selection for the High and Low preference lines was associated with a cluster of DV genes that were highly enriched in a single network module (greenyellow). The module was highly enriched in genes associated with receptor signaling (e.g., Chrna7, Grin2a, Htr2a, and Oprd1) but also included a large number of genes associated with cell adhesion. Cadherins and protocadherins were particularly enriched in the greenyellow module. Expanding the SH sample size from Colville et al. (2017) by ∼50% did not perceptually change the results. In the SH, the DV genes were highly enriched in a single module (magenta) that was similar to the greenyellow module (again remembering that module color has no meaning and is randomly assigned). The magenta and greenyellow
