Basics of evolutionary theory
The first tutorial will be hands-on, to give you a chance to check your understanding of the basics of the comparative method and evolution by natural selection.
Using the Evolution Lab game (click “play game”), complete at least the three training trees (Red, green and gecko; Familiar faces; Tree of life: Vegetarian edition). The aim here is to understand why closely related species might be expected to share many traits, and how patterns of shared and differing traits between organisms can be informative about evolutionary history of traits and species (e.g. the patterns of relatedness between species, and when certain traits are likely to have evolved). NB. this app works in Chrome or Edge, your success in other browsers may vary, if necessary try a different browser! You can access the game with a guest account if you don’t want to log in with google.
Use the AlleleA1 web app to answer the questions below. The aim here is to get a basic understanding of the effects of selection and also genetic drift (changes in gene frequency driven by chance).
The AlleleA1 app allows you to simulate the evolution of a single gene in an imaginary population of organisms. There are two possible genetic variants, called A1 and A2; each organism has two parents and inherits a variant from each, so an individual might be characterised as A1A1 (inherits the A1 variant from both parents), A1A2 (inherits the A1 variant from one parent and the A2 variant from the other), or A2A2 (inherits the A2 variant from both parents). The app allows you to manipulate selection in favour of each possible genotype (A1A1, A1A2, A2A2), by manipulating the relative fitness of each combination of genetic variants. By default the app starts with all genotypes having equal fitness (they all have relative fitness of 1, so they all produce the same number of offspring), and it plots the frequency of A1 variants in an infinitely large population for 50 generations of evolution; in this scenario, no gene frequencies every change, so the plot is a very boring straight line.
You can change the course of evolution in this simulated population by changing the relative fitness of the three genotypes, by changing the 3 numbers in the box “Selection — relative fitnesses”. Relative fitness is an easy way of modelling differences in the number of offspring each genotype produces on average - a genotype with relative fitness of 1 leaves twice as many offspring as a genotype with relative fitness 0.5. Q1: Can you find relative fitness settings where genetic variant A1 takes over the population (i.e. the line moves to Frequency of Allele A1 close to 1) or dies out (i.e. the line moves to Frequency of Allele A1 close to 0)? Q2: What affects the speed with which this happens? You might want to change to plotting more than 50 generations to see what is happening.
Things get more interesting if you move from looking at infinitely large populations (where the maths is very neat and the lines very smooth) to populations of a finite size (i.e. where there are a set number of organisms at each generation). You can change this by changing the parameter “Number of finite populations to simulate” to some number other than 0 - e.g. if you set it to 5, it will simulate 5 populations for you, each of 100 individuals (and you control the population size in the next box in the app). The graph will plot the frequency of variant A1 in each population, one line per population; you can click “run again” to rerun the simulations, it will be different every time because these involve a random component for who lives, dies and reproduces – on average the relative fitnesses are as you specify them in the boxes, but on any given generation a given organism might get lucky and have more offspring than you expect, or get unlucky and have fewer. Q3: Reset the app to the default parameters (button in the top right) and then simulate 5 populations where all variants have equal fitness. What happens, and why do you think that happens? Q4: What happens if you change the population size to very small (e.g. 10 individuals) or very large (e.g. 1000)? Q5: What if you change the selection parameters as you did before - does the fittest variant always win?
All aspects of this work are licensed under a Creative Commons Attribution 4.0 International License.