How natural selection works at the level of genes, alleles, genotypes, & phenotypes.

Key Points

  • Natural selection can cause microevolution (change in allele frequencies), with fitness-increasing alleles becoming more common in the population.
  • Fitness is a measure of reproductive success (how many offspring an organism leaves in the next generation, relative to others in the group).
  • Natural selection can act on traits determined by alternative alleles of a single gene, or on polygenic traits (traits determined by many genes).
  • Natural selection on traits determined by multiple genes may take the form of stabilizing selection, directional selection, or disruptive selection.

Introduction

We've already met a few different mechanisms of evolution. Genetic drift, migration, mutation...the list goes on. All of these mechanisms can make a population evolve, or change in its genetic makeup over generations.
But there's one mechanism of evolution that's a bit more famous than the others, and that's natural selection. What makes natural selection so special? Out of all the mechanisms of evolution, it's the only one that can consistently make populations adapted, or better-suited for their environment, over time.
You may have already seen natural selection as part of Darwin’s theory of evolution. In this article, we will dive deeper – in fact, deeper than Darwin himself could go. We will examine natural selection at the level of population genetics, in terms of allele, genotype, and phenotype frequencies.

Quick review of natural selection

Here is a quick reminder of how a population evolves by natural selection:
  • Organisms with heritable (genetically determined) features that help them survive and reproduce in a particular environment tend to leave more offspring than their peers.
  • If this continues over generations, the heritable features that aid survival and reproduction will become more and more common in the population.
  • The population will not only evolve (change in its genetic makeup and inherited traits), but will evolve in such a way that it becomes adapted, or better-suited, to its environment.

Natural selection can cause microevolution

Natural selection acts on an organism’s phenotype, or observable features. Phenotype is often largely a product of genotype (the alleles, or gene versions, the organism carries). When a phenotype produced by certain alleles helps organisms survive and reproduce better than their peers, natural selection can increase the frequency of the helpful alleles from one generation to the next – that is, it can cause microevolution.

Example: Rabbit coat color

As an example, let's imagine a population of brown and white rabbits, whose coat color is determined by dominant brown (B) and recessive white (b) alleles of a single gene. If a predator such as a hawk can see white rabbits (genotype bb) more easily than brown rabbits (BB and Bb) against the backdrop of a grassy field, brown rabbits are more likely than white rabbits to survive hawk predation. Because more brown than white rabbits will survive to reproduce, the next generation will probably contain a higher frequency of B alleles.
We can demonstrate this to ourselves by working through an example. Let's start with a set of allele and phenotype frequencies, shown in the diagram below, and see how they change in a generation if half of the white rabbits (but none of the brown rabbits) are eaten by hawks:
For that, we can use the Hardy-Weinberg equation.
You may think, that shouldn't work, because there's selection! You're right that the population is not in Hardy-Weinberg equilibrium, because its allele frequencies are changing from generation to generation. And if you check, you'll find that the allele frequencies and genotype frequencies of the survivors group do not match Hardy-Weinberg expectations.
However, once we know the allele frequencies in the survivors group, we can assume that the survivors will mate randomly with one another and use the Hardy-Weinberg equilibrium to predict the likely genotype and phenotype proportions of their offspring.
Allele frequencies:
Calculation of genotypes based on allele frequencies using Hardy-Weinberg:
Calculation of number of individuals of each genotype (population size =20= 20):
We would predict about 33 BB, 1010 Bb, and 77 bb rabbits. Since BB and Bb rabbits are brown, we would expect 3+10=133 + 10 = 13 brown rabbits. Only bb rabbits are white, so we would expect 77 white rabbits.
In this example, the frequency of the survival-promoting B allele rose from 0.30.3 to 0.40.4 in a single generation. The percent of the population with the survival-promoting brown phenotype also rose from 50%50\% to 65%65\%. (We can predict the next generation by assuming that the survivors mate randomly and leave equal numbers of offspring on average.) This is a made-up example, but it gives us a concrete sense of how natural selection can shift allele and phenotype frequencies to make a population better-suited to its environment.
Will the recessive b alleles disappear from the population due to selection? Maybe someday, but not right away. That's because they can “hide” from predators in the heterozygous (Bb) brown rabbits. This is a good reminder that natural selection acts on phenotypes, not genotypes. A hawk can tell a brown rabbit from a white rabbit, but it can't tell an BB rabbit from an Bb rabbit.

Fitness = reproductive success

The phenotypes and genotypes favored by natural selection aren't necessarily just the ones that survive best. Instead, they're the ones with the highest overall fitness. Fitness is a measure of how well organisms survive and reproduce, with emphasis on "reproduce." Officially, fitness is defined as the number of offspring that organisms with a particular genotype or phenotype leave behind, on average, as compared to others in the population.
Survival is one important component of fitness. In order to leave any offspring at all in the next generation, an organism has to reach reproductive age. For instance, in the example above, brown rabbits had higher fitness than white rabbits, because a larger fraction of brown rabbits than white rabbits survived to reproduce. Living for a longer period of time may also allow an organisms to reproduce more separate times (e.g., with more mates or in multiple years).
However, survival is not the only part of the fitness equation. Fitness also depends on the ability to attract a mate and the number of offspring produced per mating. An organism that survived for many years, but never successfully attracted a mate or had offspring, would have very (zero) low fitness.

Fitness depends on the environment

Which traits are favored by natural selection (that is, which features make an organism more fit) depends on the environment. For example, a brown rabbit might be more fit than a white rabbit in a brownish, grassy landscape with sharp-eyed predators. However, in a light-colored landscape (such as sand dunes), white rabbits might be better than brown rabbits at avoiding predators. And if there weren't any predators, the two coat colors might be equally fit!
In many cases, a trait also involves tradeoffs. That is, it may have some positive and some negative effects on fitness. For instance, a particular coat color might make a rabbit less visible to predators, but also less attractive to potential mates. Since fitness is a function of both survival and reproduction, whether the coat color is a net "win" will depend on the relative strengths of the predation and the mate preference.

Natural selection can act on traits controlled by many genes

In some cases, different phenotypes in a population are determined by just one gene. For example, this was the case with our hypothetical rabbits. It's also true in some real cases of natural selection for coat color (e.g., in mice)1,2^{1,2}.
However, in many cases, phenotypes are controlled by multiple genes that each make a small contribution overall result. Such phenotypes are often called polygenic traits, and they typically form a spectrum, taking many slightly different forms. Plotting the frequency of the different forms in a population often results in a graph with a bell curve shape. Height (see graph below) and many other traits in humans are polygenic.
Histogram showing height in inches of male high school seniors in a sample group. The histogram is roughly bell-shaped, with just a few individuals at the tails (60 inches and 77 inches) and many individuals in the middle, around 69 inches.
Image modified from "Continuous variation: Quantitative traits," by J. W. Kimball (CC BY 3.0).
We can see if natural selection is acting on a polygenic trait by watching how the distribution of phenotypes in the population changes over time. Certain characteristic shifts tell us selection is occurring, even if we don’t know exactly which genes control the trait.

How natural selection can shift phenotype distributions

There are three basic ways that natural selection can influence distribution of phenotypes for polygenic traits in a population. To illustrate these forms of selection, let's use an imaginary beetle population, in which beetle color is controlled by many genes and varies in a spectrum from light to dark green.
  1. Stabilizing selection. In stabilizing selection, intermediate phenotypes are more fit than extreme ones. For example, medium-green beetles might be the best camouflaged, and thus survive best, on a forest floor covered by medium-green plants. Stabilizing selection tends to narrow the curve.
  2. Directional selection. One extreme phenotype is more fit than all the other phenotypes. For example, if the beetle population moves into a new environment with dark soil and vegetation, the dark green beetles might be better hidden and survive better than medium or light beetles. Directional selection shifts the curve towards the favorable phenotype.
  3. Disruptive selection. Both extreme phenotypes are more fit than those in the middle. For example, if the beetles move into a new environment with patches of light-green moss and dark-green shrubs, both light and dark beetles might be better hidden (and survive better) than medium-green beetles. Diversifying selection makes multiple peaks in the curve.

Summary

Natural selection can cause microevolution, or a change in allele frequencies over time, with fitness-increasing alleles becoming more common in the population over generations. Fitness is a measure of relative reproductive success. It refers to how many offspring organisms of a particular genotype or phenotype leave in the next generation, relative to others in the group.
Natural selection can act on traits determined by different alleles of a single gene, or on polygenic traits (traits determined by many genes). Polygenic traits in a population often form a bell curve distribution. Natural selection on polygenic traits can take the form of:
  • Stabilizing selection: Intermediate phenotypes have the highest fitness, and the bell curve tends to narrow.
  • Directional selection: One of the extreme phenotypes has the highest fitness. The bell curve shifts towards the more fit phenotype.
  • Disruptive selection: Both extreme phenotypes have a higher fitness than intermediate phenotypes. The bell curve develops two peaks.

Attribution:

This article is a modified derivative of "Natural selection and populations - Advanced," by Douglas Wilkin and Barbara Akre, CK-12 Foundation, CC BY-NC 3.0.
The modified article is licensed under a CC BY-NC-SA 4.0 license.

Works cited:

  1. Bradt, S. (2009, August 27). Mice living in Sandhills quickly evolved lighter coloration. In Harvard gazette. Retrieved from http://news.harvard.edu/gazette/story/2009/08/mice-living-in-sand-hills-quickly-evolved-lighter-coloration/.
  2. Linnen, C. R., Kingsley, E. P., Jensen, J. D., and Hoekstra, H. E. (2009). On the origin and spread of an adaptive allele in deer mice. Science, 325(5944), 1095-1098. http://dx.doi.org/10.1126/science.1175826.

Additional references:

Evolution. (2014). In Scitable. Retrieved from http://www.nature.com/scitable/definition/evolution-78.
Fitness (biology). (2016, May 19). Retrieved May 21, 2016 from Wikipedia: https://en.wikipedia.org/wiki/Fitness_%28biology%29.
Purves, W. K., Sadava, D. E., Orians, G. H., and Heller, H.C. (2003). The results of natural selection. In Life: The science of biology (7th ed., pp. 470-475). Sunderland, MA: Sinauer Associates.
Reece, J. B., Urry, L. A., Cain, M. L., Wasserman, S. A., Minorsky, P. V., and Jackson, R. B. (2011). Natural selection is the only mechanism that consistently causes adaptive evolution. In Campbell biology (10th ed., pp. 491-498). San Francisco, CA: Pearson.
University of California Museum of Paleontology. (2016). Misconceptions about natural selection. In Understanding evolution. Retrieved from http://evolution.berkeley.edu/evolibrary/article/evo_32.
University of California Museum of Paleontology. (2016). Natural selection. In Understanding evolution. Retrieved from http://evolution.berkeley.edu/evolibrary/article/evo_25.
Wilkin, D. and Brainard, J. (2016, May 12). Natural selection. In CK-12: Eco pop. Retrieved from http://www.ck12.org/user:amzpzwxkqg11cnnklm9yzw../book/Eco-Pop/section/19.0/.
Raven
Campbell Concepts & Connections
Evolutionary analysis
Loading