Abstract
Microbial populations undergo multiple phases of growth, including a lag phase, an exponential growth phase, and a stationary phase. Therefore mutations can improve the frequency of a genotype not only by increasing its growth rate, but also by decreasing the lag time or adjusting the yield (resource efficiency). Furthermore, many mutations will be pleiotropic, affecting multiple phases simultaneously. The contribution of multiple life-history traits to selection is a critical question for evolutionary biology as we seek to predict the evolutionary fates of mutations. Here we use a simple model of microbial growth to quantify how these multiple phases contribute to selection. We find that there are two distinct components of selection corresponding to the growth and lag phases, while the yield modulates their relative importance. Despite its simplicity, the model predicts nontrivial population dynamics when mutations induce tradeoffs between phases. Multiple strains can coexist over long times due to frequency-dependent selection, and strains can engage in rock-paper-scissors interactions due to non-transitive selection. We characterize the environmental conditions and patterns of traits necessary to realize these phenomena, which we show to be readily accessible to experiments. Our results provide a theoretical framework for analyzing high-throughput measurements of microbial growth traits, especially interpreting the pleiotropy and correlations between traits across mutants. This work also highlights the need for more comprehensive measurements of selection in simple microbial systems, where the concept of an ordinary fitness landscape breaks down.