Abstract

The reason for the prevalence of sexual reproduction among multicellular eukaryotes is a long-standing unanswered question in evolutionary biology. It is widely believed that sexual reproduction and the resulting genetic recombination provide a selective advantage by increasing a population’s genotypic variance. However, recombination will only do so when the population is in negative linkage disequilibrium. It has been proposed that if alleles do not contribute multiplicatively to fitness, but instead display a negative epistatic fitness curve, then the population will be in such a state following selection, and sex will be advantageous. However, there is no a priori reason to believe that fitness values should generally be negatively epistatic, as opposed to positive or zero.

In this study, we explore the relationship between contest competition and the maintenance of sexual reproduction. In Chapter 2, we develop a two-locus bi-allelic haploid model to examine the relationship between phenotype, competitive selection, and realized fitness. We assume that competitive ability is directly proportional to the phenotypic value and that the outcome of pairwise competition is dependent on the ratio of competitive abilities of the competing individuals. The stronger competitor does not always win, but it wins more frequently than the weaker competitor does. Using this very simple model, we find that intraspecies competition can lead to frequency dependent changes in genotypic fitness. In addition, competition can result in negative epistasis at the level of realized fitness. This leads to the build-up of negative linkage disequilibrium among the alleles that affect fitness following selection.

In Chapter 3, we explore the effects of competition when the selection is repeated over several generations. We find that the amount of negative linkage disequilibrium builds up continuously in the absence of recombination, and this hinders the progress of selection. This accumulation of negative linkage disequilibrium is alleviated when recombination is present. We show, using both a numerical modeling approach and through individual-based simulations, that sexual individuals will increase in frequency in a mixed population of sexual and asexual individuals. The selective advantage of recombination is strongest when there is a large difference in competitive ability between genotypes, and when the selected alleles are initially rare.

Finally in Chapter 4, we consider a different mapping of phenotype onto fitness to ensure that our findings are robust. In this case, the outcome of pairwise competition is not determined by the ratio of competitive abilities; instead the stronger competitor always wins. In this case, as expected, there is even stronger negative epistasis at the fitness level. As a result, we find that advantageous alleles go to fixation at a faster rate when recombining, and sexual individuals increase in frequency when competing against asexual individuals in a mixed population.

Our results indicate that competition can play a significant role in the maintenance of sexual reproduction, and that the advantage of sexual reproduction may not lie in fertility selection, but in viability selection. In a competitive situation, it is better to produce one winner rather than two losers.