Revisiting regulatory decoherence and phenotypic integration: accounting for temporal bias in co-expression analyses

Abstract

Environment can alter the degree of phenotypic variation and covariation, potentially influencing evolutionary trajectories. However, environment-driven changes in phenotypic variation remain understudied. In an effort to exploit the abundance of RNASequencing data now available, an increasing number of ecological studies rely on population-level correlation to characterize the plastic response of the entire transcriptome and to identify environmentally responsive molecular pathways. These studies are fundamentally interested in identifying groups of genes that respond in concert to environmental shifts. We show that population-level differential co-expression exhibits biases when capturing changes of regulatory activity and strength in rice plants responding to elevated temperature. One possible cause of this bias is regulatory saturation, the observation that detectable co-variance between a regulator and its target may be low as their transcript abundances are induced. This phenomenon appears to be particularly acute for rapid-onset environmental stressors. However, our results suggest that temporal correlations may be a reliable means to detect transient regulatory activity following rapid onset environmental perturbations such as temperature stress. Such temporal bias is likely to confound the studies of phenotypic integration, where high-order organismal traits are hypothesized to be more integrated with strong correlation under stressful conditions, while recent transcriptome studies exhibited weaker coexpression between genes under stressful conditions. Collectively, our results point to the need to account for the nuances of molecular interactions and the possibly confounding effects that these can introduce into conventional approaches to study transcriptome datasets.