Evolution in a flat fitness landscape (Q1175540)

A simple model of a population of fixed size \(M\) of asexually reproducing individuals is defined. Each individual has a genome characterized by \(N\) binary units. To each individual \(\alpha\) at generation \(t\) is assigned a parent \(G_ t(\alpha)\), which is, with equal probability, any of the \(M\) individuals of generation \(t-1\) ( flat fitness landscape or absence of selection); the parent has the same genome except for point mutations, which occur randomly at a fixed rate \(\mu\). It is shown that the model is equivalent to a stochastic dynamical system (with reversed time), the `` annealed random map model'' introduced by \textit{B. Derrida} and \textit{D. Bessis} [J. Phys. A 21, No. 9, L509--L515 (1988)], making it possible to solve exactly for genealogy statistics and genetic variability. The average relatedness and the variability fluctuate over time, even as \(M\to+\infty\). The model may also be considered a generalization to \(N>1\) of the neutral Wright model with genetic drift and mutation for a fixed population, but has the advantage of allowing computation of genealogy statistics.