Network-level selection constrains the adaptive evolution of gene expression noise in regulatory genes in changing environments

In simulated model gene regulatory networks, expression noise is only transiently favored under directional selection but persistently maintained under fluctuating selection, depending on network position.

• Why it matters: Expression noise can help populations adapt in changing environments, yet it spreads through regulatory networks, making gene evolution interdependent and hard to predict from single-gene views.
• What they did: In silico evolution experiments simulated populations of model gene regulatory networks under directional and fluctuating selection on gene expression level, allowing basal expression and expression noise to mutate.
• The result: Target genes were more likely than regulator genes to increase gene-specific expression noise under selection, implying network-dependent evolution of both mean and variance that may support escape from extinction via phenotypic heterogeneity.

AI summary of the abstract · may contain errors · not medical advice · read the paper.