The role of sexual selection, particularly through recombination, in shaping the rate of adaptation has been the subject of many theoretical and empirical studies.In asexual populations, beneficial mutations occurring in distinct individuals compete for fixation, slowing down adaptation.The situation is even worse in a structured metapopulation (of the same size): a beneficial mutation needs to fix in each subpopulation (aka deme), where genetic drift is stronger.Recombination, however, creates individuals with high fitness carrying multiple beneficial mutations, preventing their loss (Hill and Robertson, 1966).In the presence of recombination, however, the effect of population structure becomes more complex.Demes provide some degree of independent evolution, which increases genetic diversity at the global population level and thereby enhances recombination's effect.Studying this interaction brings another fundamental question about the timing of the underlying events: migration and recombination.This is particularly relevant, as in many species the two are connected: dispersal structures are the product of sexual reproduction (e.g., in many plants).Liu and Weismann (2026) provide a first attempt to study this question.They developed a theoretical model in which the timing of dispersion and sexual reproduction can be varied, enabling testing of different hypotheses about their synchronization.This allows them to distinguish two types of synchronization: individual-level synchronization, in which dispersal and sexual reproduction occur simultaneously within individuals but 1 independently between individuals, and population-level synchronization, in which dispersal and sexual reproduction do not occur simultaneously within individuals but are synced between individuals.Individual synchronization corresponds to a physiological constraint, whereas in natural populations, population synchronization is typically induced by an environmental cue.Of course, the two types of synchronization can be combined.The model is heavily inspired by the yeast experimental evolution setup of Kryazhimskiy et al (2012).This work first confirmed that when clonal interference is strong, population structure does not slow down adaptation and can even increase it, because the rate of adaptation (within demes) depends little on population size and recombination is more efficient due to increased genetic diversity across the global population.The authors then demonstrate that synchronizing dispersal and sexual reproduction can accelerate adaptation.This effect is highly synergetic, as synchronization of sexual reproduction alone in a well-mixed population leads to a slower rate of adaptation.The authors further show that individual synchronization also increases the rate of adaptation, and that this effect can add up to that of the population, but is comparatively smaller.The authors reveal that this is because the effect on adaptation rate is mostly driven by recombination between migrants, which is enhanced when synchronization occurs at the population level, and when both sexual reproduction and dispersal occur simultaneously at that level.This work provides an innovative model and new insights into the impact of sexual reproduction on the rate of adaptation.One important point here is that the reported effects are in the absence of epistasis, showing that some of the well-documented effects of recombination also hold on smooth fitness landscapes.New results also bring new questions.As highlighted by one of the reviewers, the timescales of within-and between-deme processes shape the global dynamics, as demonstrated by the effect of the period of synchronized events on the resulting rate of adaptation.A deeper understanding of this effect requires additional investigations.Finally, this work potentially sheds light on the evolution of life history traits, including the individual synchronization of reproduction and dispersal, as well as on the evolution of environment-sensing systems that ultimately lead to population synchronization.As envisioned by the authors, an extension of the proposed framework with modifier loci is likely to be enlightening.
Julien Yann Dutheil (Mon,) studied this question.