Causes and Consequences of the Evolution of Mutation Rate

As the ultimate source of all genetic variation, mutation is required for evolution. The mutation rate measures the rate at which mutations occur over time. How the mutation rate evolves and how it interacts with other evolutionary processes are still far from clearly understood. In this thesis, I employ individual-based simulations and aim to understand how the evolution of mutation rate interacts with other evolutionary forces such as mutation biases, selection for evolvability, genetic drift and the evolution of recombination rate. In Chapter 3, I studied the role of mutation bias (bias towards high mutation rate) in the evolution of mutation rate, which has been long ignored in the literature. I found that the effect of mutation bias on the evolution of mutation rate is significant when compared to that of other broadly promoted evolutionary forces such as natural selection, mutator hitchhiking, and genetic drift. Even in sexual populations, I found that mutation bias can still operate and drive the evolution of mutation rate. In Chapter 4, when the mutation rate (U) and recombination rate (R) are not allowed to evolve, I found that there exists an optimal mutation rate (Uopt), at which a population can achieve their maximal evolvability (E). Popula- tions displayed negative evolvability if U was above a critical value (Ucrit). Asexual and sexual populations showed similar relationships between E and U. Moreover, increasing R also increased E, Uopt, and Ucrit. In Chapter 5, I found that selection for evolvability cannot optimize U when it is allowed to evolve because U increased without bound — a phenomenon known as mutation rate catastrophe. In addition, the effect of mutation bias is much stronger than the selection for evolvability in sexual populations, indicating the selection for evolvability is not the only force that can affect the evolution of U and is a weak selection. However, U cannot be opti- mized when both forces are operating. Lastly, I found that selection can optimize the recombination rate ...