Daniel Rice Thesis Defense (Desai Lab)

Date: 

Friday, January 27, 2017, 10:00am

Location: 

Harvard Herbaria Seminar Room

Title:  Statistics and Dynamics of Adaptation from de novo Mutations

Abstract:  All living organisms are shaped by natural selection. Therefore, by studying natural selection as a generic process, we may gain insights into broad questions regarding the diversity of life, the fit of organisms to their environments, and the rate of evolutionary change. The process of natural selection consists of two steps. First, genetic mutations generate distinct genotypes within a population, which vary in their rates of survival and reproduction. Second, competition among genotypes determines which mutations go extinct and which persist and contribute to the long-term evolution of the population. This dissertation uses theoretical and experimental techniques to study: (1) the statistical properties of how mutations effect organismal fitness, and (2) how the dynamics of natural selection determine the fate of these mutations.

In Chapter 2, we measure the fitness gains and genome sequence changes of a large number of populations of Saccharomyces cerevisiae yeast adapting to experimentally-controlled conditions. Our results suggest that interactions among different beneficial mutations generate predictable patterns of fitness gain.

Chapter 3 is a theoretical investigation of how natural selection shapes the distribution of fitness effects of mutations available to a population. We show that selection may induce certain regularities on this distribution that are independent of many biological details.

Chapters 4 and 5 use DNA sequence data from experimentally-evolved populations of S. cerevisiae to study the dynamics of natural selection in rapidly-adapting populations. Chapter 4 demonstrates that competition between rival lineages, known as clonal interference, dominates the dynamics of asexuallyreproducing populations. Chapter 5 shows how sexual reproduction can alleviate the effects of clonal interference, making selection more efficient and increasing the rate of adaptation.