MMEE2024

Mathematical Models in Ecology and Evolution

July 15-18, 2024
Vienna, AUSTRIA

"Global Epistasis in Multiple Environments"

Brooks, Caelan

Empirical studies have shown that "global" patterns of diminishing returns and increasing costs epistasis are typical in microbial evolution. These patterns mean that beneficial mutations typically become systematically less beneficial (and can even become deleterious) in more-fit backgrounds, while deleterious mutations become more deleterious. Recent theoretical work has found that these patterns are a natural property of high-fitness genotypes in a model in which there are widespread random, "idiosyncratic" epistatic interactions between many loci. The essential idea is that higher-fitness genotypes tend to be more "tuned" to take advantage of these random interactions, so they are correspondingly more susceptible to disruption. However, selection pressures in natural populations often fluctuate across space and time, so organisms are often tuning fitness across multiple environments. In these cases, patterns of pleiotropy (i.e. the effects of a given mutation across different environments) are also critical to evolution. Here, we analyze these patterns of pleiotropy and epistasis in a more general class of models of widespread random epistatic and pleiotropic effects among many loci. We show that systematic global patterns of both epistasis and pleiotropy emerge naturally from these models, as a result of a tension between how finely tuned fitness can be in different environments. These patterns have general implications for evolution in fluctuating conditions, and also make empirically testable predictions, which we confirm using experimental data describing the effects of a large set of gene disruption mutations across many genetic backgrounds and environmental conditions.

« back