A minimal derivation of convergence stability measures (Q2177159)

The author considers only a population divided into a finite number of demos, with the same demographic and selective forces acting in all demos (``spatially homogeneous'' models). Each deme has a fixed number of adults and the probability of propagation (distribution of the distance of propagation from any deme) does not depend on the spatial position of the deme. For spatially homogeneous models, the measure of convergence stability can be calculated in terms of fitness functions and the classical concept of genetic identity by origin in models with repeated mutation. However, this expression and its derivation presented earlier are contrary to common sense, since the challenge of the rate of repetitive mutation is not required to calculate the probability of fixation and any measure of fitness that follows from it. The article presents a different conclusion that does not account for repeated mutations, but uses simple results from the theory of fusion in population genetics. These results show that the indicator of ``fitness'' can be expressed in terms of the average time of fusion of pairs of genes. There is no need to take into account the repetitive mutation rate. However, algebraic methods of repetitive mutation models remain a convenient tool for analyzing models. The arguments in this article can be generalized to less symmetric models.