## Info

The mutation rates at microsatellite or simple sequence repeat (SSR) loci are also of interest since such loci are widely employed as selectively neutral genetic markers to study a wide range of population genetic processes. These repeated DNA regions have very high rates of mutation between 1 x 10-2 and 6 x 10-6 per sexual generation (Ellegren 2000; Steinberg et al. 2002; Beck et al. 2003).

How can such a low-probability event like mutation add more than a trivial amount of genetic variation to populations? Let's calculate an initial answer to that question using humans as our example. Averaging over coding and non-coding parts of the genome, an approximate nuclear genome mutation rate in humans is about 1 x 10-9 mutations per base pair per generation. The haploid genome (one sperm or egg) contains about 3.2 x 109 base pairs (bp). Each genome of each diploid individual will have:

(1 x 10-9 mutations bp-1 generation-1) (2 x 3.2 x 109 bp-1) = 6.4 mutations (5.1)

where the factor of 2 is due to a diploid genome. Each of us differs from one of our parents by half this amount, or about three mutations on average. If all mutations are random events evenly distributed throughout the genome, every pair of individuals differs by twice this number of mutations or about 13 mutational differences on average. The overall effect of mutation on available genetic variation depends on the size of a population. The human population is currently about 6.486 billion people (see www.census.gov/main/www/popclock.html). Based on this population size, there are a total of

(6.4 mutations individual-1 generation-1)(6.486 x 109 individuals) = 41.5 x 109 mutations (5.2)

or over 41 billion mutations expected each generation! This means that the absolute numbers of mutations per generation are potentially high and depend on the rate of mutation, the size of the population, and the size of the genome. We will revisit this topic later in the chapter to make a more formal prediction about the levels of heterozygosity expected when the input of genetic variation due to mutation and the loss of genetic variation caused by genetic drift are at equilibrium.

The impact that a mutant allele (as part of a heterozygous or homozygous genotype) has on the pheno-type of an individual can vary greatly. Since natural selection along with genetic drift are critical processes that determine the fate of new mutations, the phenotype is most often considered in the context of its survivorship and reproduction, or fitness. The range of the possible fitnesses for an individual mutant allele can be thought of as a mutation fitness spectrum like that shown in Fig. 5.1. The fitness of all mutation effects on the phenotype is relative to the average fitness of a population (see Chapter 6 for explanations of fitness and average fitness). Detrimental or deleterious mutations reduce survival and reproduction while mutations that improve survival and reproduction are advantageous. Severely deleterious mutations such as those that result in death (called lethals) or failure to reproduce viable offspring are acted strongly against by natural selection and will likely not last for a single generation. Mutations that

Relative fitness of a new mutation

Figure 5.1 A hypothetical distribution of the effects of mutations on phenotypes that ultimately impact the Darwinian fitness of genotypes. Mutations that have a mean fitness less than the mean fitness of the population (w) are decreased in frequency by natural selection. The shaded area around w indicates the zone where mutations have small effects on fitness relative to the effects of genetic drift (the width of the neutral zone depends on the effective population size). The shaded area near zero mean fitness indicates mutations that cause failure to reproduce or are lethal. Lethals are more common since it is a category that includes many degrees of severity resulting from diverse causes. The fitness effects of mutations are inherently difficult to measure because of the rarity of mutation events, the small effect of many mutations, and the dependence of fitness on environmental context.

are strongly deleterious and nearly lethal are sometimes called sublethals. Mutations that have small positive or negative effects on fitness (the shaded zone around the mean fitness in Fig. 5.1) are called neutral or nearly neutral since their fate will be dictated either totally or mostly by sampling error of genetic drift. The final type are beneficial mutations that increase survival and reproduction above the average fitness of the population. It is important to note that the fitness effects of mutations may depend greatly on environmental context (see Fry & Heinsohn 2002) and the genotype at other loci. These different types of mutation will be the subject of models later in this chapter that show how fitness relates to the chance that a new mutation is lost or reaches fixation in a population.

The mutation fitness spectrum plays a central role in a wide range of hypotheses to explain a multitude of phenomena in population genetics and evolution (see Charlesworth & Charlesworth 1998; Lynch et al 1999; Orr 2003; Estes et al. 2004). Explanations for phenomena as general and diverse as inbreeding depression, the evolution of mating systems, the evolution of sex and recombination, and the rate of adaptation depend in part on the nature of the mutation fitness spectrum. Strongly deleterious or strongly beneficial mutations will be steadily and predictably driven to loss or to fixation, respectively, by natural selection. However, fixation and loss of mutations that have a small impact on fitness (relative to the effective population size) is due in whole or in part to random genetic drift. A consequence is that mildly deleterious mutations may reach fixation by chance and accumulate in a population over time. Similarly, some mildly beneficial mutations may be lost from populations by chance. An accumulation of mildly deleterious mutations reduces individual fitness and may increase the risk of extinction, resulting in natural selection for processes that reduce the load of deleterious mutations in a population. The frequency of beneficial mutations may also place limits on the rate of evolution by positive natural selection. Thus, the specific shape of the frequency

Mutation fitness spectrum The frequency distribution of the average fitness of new mutations measured relative to the average fitness of a reference population.

distribution shown schematically in Fig. 5.1 provides crucial information about the fate of individual mutations as well as the long-term consequences of continual mutation in populations.

A commonly employed method to estimate the shape of the mutation fitness spectrum relies on founding a series of genetically identical populations and then allowing some to experience mutations for many generations while maintaining a control population that does not experience mutation. Viability and reproduction phenotypes of the mutated populations are then compared with the control population at intervals to estimate the average change in fitness caused by the mutations. Such comparisons are called mutation-accumulation experiments since mutations are repeatedly fixed by genetic drift over time in the mutation populations.

If there was absolutely no mutation, the replicate populations in a mutation-accumulation experiment would all maintain identical viability over time since each population started out being genetically identical. Mutation, however, will occur at random and cause independent genetic changes in the different populations, causing the populations to diverge in viability. Imagine that the mutation fitness spectrum is symmetric around the mean fitness so that the frequency of deleterious and beneficial mutations of the same magnitude is equal. That would produce no change in the average viability of lines in a mutation-accumulation experiment since there would be equal chances of beneficial or deleterious mutations of the same size that would cancel each other out in a sample of many mutations. However, there would be an increase in the variance in viability because the range of viabilities among the populations would increase with more and more mutations. Next imagine a mutation fitness spectrum like that shown in Fig. 5.1 where deleterious mutations are more common than beneficial mutations. As mutations accumulate, the average viability of lines should decrease since deleterious mutations are more common than beneficial mutations. The more skewed the distribution is toward deleterious mutations the faster the average viability should decrease in the replicate populations.

The results of several classic mutation accumulation studies that estimated the frequency distribution for mutations that affect viability in Drosophila melanogaster have had a major impact on perceptions of the mutation fitness spectrum (Mukai 1964; Mukai et al. 1972). Mutation-accumulation experiments in Drosophila rely on special breeding designs

0 0