The Evolution of Ageing

Ageing, or senescence, is the phenomenon of mortality rates of organisms increasing through the course of their lives. Despite senescence having obvious negative fitness consequences (such as death), it is nevertheless a fact of life that most organisms age. Why doesn't natural selection lead to ever-increasing lifespans? The 'developmental theory of aging' (DTA) suggests that aging evolves due to evolution prioritizing optimization of early-life fitness at the cost of fitness later in life (For example, see Lemaître et al. 2024). In other words, if the optimal value of a trait varies with age, the DTA posits that evolution favors optimality early in life even if this comes with a cost later in life. Aging in this framework is thus a consequence of failure to adapt to a changing optimum leading to either hypofunction or hyperfunction in gene regulation, which in turn leads to sub-optimal, and eventually catastrophic, organismal function. I am using tools from mathematical optimization theory and the calculus of variations to formalize this idea and examine when age-dependent optimal trait expression can lead to senescence. For instance, studies of aging routinely assume that selection is weaker later in life (the famous 'selection shadow'). Is a selection shadow really necessary for aging as envivionsed by the DTA? If so, how strong must this shadow be? If not, can aging evolve simply due to limited plasticity in trait expression?


Any sufficiently complex organism could conceivably be conceptualized as a system of interacting sub-systems. These sub-systems could be genes, organs, tissues, or something more abstract, such as cellular pathways, metabolic functions, or foraging ability. Crucially, in any such conceptualization, the functioning of each sub-part depends on the well-being of other sub-parts. For instance, a faulty heart places additional stresses on lung functioning. Organismal functioning and mortality then depend on the fraction of currently operational sub-parts. For my PhD, I am using mathematical models rooted in graph theory and network percolation to study whether such interdependencies are sufficient to reproduce patterns of demographic senescence without invoking a selection shadow. I am also trying to establish connections with more standard control theoretic ideas based on the developmental theory of ageing and outline how trade-offs between maintenance of different sub-systems could potentially lead to interesting eco-evolutionary insights.