The latent load and the population-optimal mutation rate

Abstract

The rarity of positively selected sites may lead to the expectation that they have only a minor effect on total genetic load. A framework for the genetic load associated with positively selected sites in sexual populations is presented and analyzed. This framework defines the latent load as the genetic load associated with positively selected sites that are destined to fix, but for which beneficial alleles have not yet become established. A formula for the latent load is derived that incorporates various real-world complicating factors. In humans, the latent load is estimated to be larger than the mutational load, and much larger than the substitutional load. The germline spontaneous mutation rate that maximizes population mean fitness is predicted within this framework to occur when the mean latent load is approximately equal to the mutational load. This population-optimal mutation rate differs from the minimal mutation rate predicted by purely microevolutionary considerations that focus on individual-level selection. It is hypothesized that, over macroevolutionary timescales, species-level processes such as differential extinction may bias the mutation rates of surviving species toward the population-optimal value. Overall, this framework highlights the importance of positively selected sites to the total genetic load, and suggests a potential role for the latent load in shaping the long-term evolution of the mutation rate.