A simple expression for the strength of selection on recombination generated by interference among mutations

Abstract

Significance Recombination between parental chromosomes during meiosis represents an important source of genetic novelty and is thought to be the main evolutionary benefit of sexual reproduction. However, the evolutionary forces driving the rapid evolution of recombination rates demonstrated by comparisons between populations or closely related species remain obscure. This article provides mathematical quantification of the selective advantage of a mutation increasing the genetic map length (average number of cross-overs occurring at meiosis) of a whole genome due to the increased efficiency of selection against deleterious alleles. It shows that the advantage of recombination can be expressed as a simple expression of the mutation rate per unit map length, providing a simple way of evaluating its plausible order of magnitude. One of the most widely cited hypotheses to explain the evolutionary maintenance of genetic recombination states that the reshuffling of genotypes at meiosis increases the efficiency of natural selection by reducing interference among selected loci. However, and despite several decades of theoretical work, a quantitative estimation of the possible selective advantage of a mutant allele increasing chromosomal map length (the average number of cross-overs at meiosis) remains difficult. This article derives a simple expression for the strength of selection acting on a modifier gene affecting the genetic map length of a whole chromosome or genome undergoing recurrent mutation. In particular, it shows that indirect selection for recombination caused by interference among mutations is proportional to NeU2/NeR3, where Ne is the effective population size, U is the deleterious mutation rate per chromosome, and R is the chromosome map length. Indirect selection is relatively insensitive to the fitness effects of deleterious alleles, epistasis, or the genetic architecture of recombination rate variation and may compensate for substantial costs associated with recombination when linkage is tight. However, its effect generally stays weak in large, highly recombining populations.