Biological aging is characterized by an age-dependent increase in the probability of death and by a decrease in the reproductive capacity. Individual age-dependent rates of survival and reproduction have a strong impact on population dynamics, and the genetic elements determining survival and reproduction are under different selective forces throughout an organism lifespan. Here we develop a highly versatile numerical model of genome evolution -both asexual and sexual -for a population of virtual individuals with overlapping generations, where the genetic elements affecting survival and reproduction rate at different life stages are free to evolve due to mutation and selection. Our model recapitulates several emerging properties of natural populations, developing longer reproductive lifespan under stable conditions and shorter survival and reproduction in unstable environments. Faster aging results as the consequence of the reduced strength of purifying selection in more unstable populations, which have large portions of the genome that accumulate detrimental mutations. Unlike sexually reproducing populations under constant resources, asexually reproducing populations fail to develop an age-dependent increase in death rates and decrease in reproduction rates, therefore escaping senescence. Our model provides a powerful in silico framework to simulate how populations and genomes change in the context of biological aging and opens a novel analytical opportunity to characterize how real populations evolve their specific aging dynamics.

9AEGIS (Ageing of Evolving Genomes In Silico) is a versatile population-genetics numerical-simulation tool 10 that enables the evolution of life history trajectories under sexual and asexual reproduction and a wide variety 11 of evolutionary constraints. By encoding age-specific survival and reproduction probabilities as discrete ge-12 nomic elements, AEGIS allows these probabilities to evolve freely and independently over time. Simulation 13 of population evolution with AEGIS demonstrates that ageing-like phenotypes evolve in stable environments 14 under a wide range of conditions, that life history trajectories depend heavily on mutation rates, and that sexual 15 populations are better able to accumulate high levels of beneficial mutations affecting early-life survival and 16 reproduction. AEGIS is free and open-source, and aims to become a standard reference tool in the study of 17 life-history evolution and the evolutionary biology of ageing. 18Species in nature vastly differ in life histories, with dramatic variation in maturation rate, lifespan, and fecun-20 dity. In general, age-dependent mortality increases as a function of age while age-dependent fecundity declines, 21 a phenomenon known as ageing or senescence. However, in some organisms mortality decreases or remains 22 constant through life, while fecundity remains constant or increases (Jones et al., 2014). These difference in 23 demography can have important effects on fitness, giving rise to dramatic differences in lifetime reproductive 24 output between species. 25The evolution of age-dependent changes in mortality and reproduction has been an important object of the-26 oretical investigation since the dawn of population genetics, giving rise to a number of theories to explain the 27 widespread occurrence of senescence in nature. Work from Haldane, Medawar, Hamilton and others predicts 28 that the declining force of natural selection after reproductive maturation should inevitably lead to the accumu-29 lation of deleterious gene variants, resulting in increased mortality later in life (Haldane, 1941; Medawar, 1952; 30 Hamilton, 1966; Charlesworth, 2000). While these mutation-accumuation theories of ageing explain ageing as 31 a fundamentally non-adaptive process, other evolutionary theories of ageing suggest senescence could evolve 32 1 as an antagonistic side-effect of positively-selected traits (Williams, 1957), or even as a kin-or group-selected 33 adaptation in its own right (Longo et al., 2005; Lohr et al., 2019). 34Up to now, the evolution of life-history traits, including age-dependent changes in survival and reproduc-35 tion, has primarily been performed using analytical approaches (Hamilton, 1966; Charlesworth, 1994; Fisher, 36 1930); while some simple numerical models exploring the evolution of ageing have been proposed (Penna, 37 1995; Dzwinel et al., 2005; Werfel et al., 2015), there remains a need for a flexible simulation tool to model the 38 evolution of ageing. In particular, a model which permits independent evolut...

No abstract