The evolution of sex: empirical insights into the roles of epistasis and drift

Key Points The origin and maintenance of sexual reproduction is one of the oldest and most intriguing problems in evolutionary biology. Most models assume that the advantage of sex relies on the increased production of genetic variation, which can be used by natural selection. For this to occur, there must be forces that systematically diminish this variation (that is, that cause 'negative linkage disequilibrium'), such as genetic drift or negative epistasis. Recent laboratory-based experiments that use microbes or comparative genomic analyses of microcrustaceans and nematodes have shown the ability of sex to facilitate adaptation and slow down maladaptation, but have not revealed whether drift or epistasis is the main responsible force. Direct and indirect empirical studies of epistasis have been numerous, but have not provided decisive support for negative epistasis. Frustration with the equivocal evidence for negative epistasis has stimulated two recent developments: studies of genetic drift as a cause of low genetic variation, and studies that combine the effects of more than one mechanism (the so-called 'pluralist approaches'). Recent in silico studies point at a possible two-way evolutionary relationship between recombination and negative epistasis. By selecting for more robust and modular genomes, sex might promote negative epistasis as well as long-term evolvability. So, sex modulates genome architecture, by which it might forge its own evolution. Theories of how sex evolved are now being explored experimentally, particularly regarding the roles of epistasis and drift. Although a generalizable theory remains elusive, new models, including theories that involve genetic architecture and robustness, are helping to understand the available evidence. Despite many years of theoretical and experimental work, the explanation for why sex is so common as a reproductive strategy continues to resist understanding. Recent empirical work has addressed key questions in this field, especially regarding rates of mutation accumulation in sexual and asexual organisms, and the roles of negative epistasis and drift as sources of adaptive constraint in asexually reproducing organisms. At the same time, new ideas about the evolution of sexual recombination are being tested, including intriguing suggestions of an important interplay between sex and genetic architecture, which indicate that sex and recombination could have affected their own evolution.