Simulation - Indiana University Bloomington

POPULATION GENETICS SIMULATIONS By F. Frey and C. Lively, Indiana University How to get the program Go to http://evolution.gs.washington.edu/popgen/popg.html (This is correct
for 2008. Use a browser other than Safari). Download the appropriate version of the popgen program for your computer.
Software Notes
See "Running the program" in the url above
Purpose of this exercise Drawing from past courses and interactions with other faculty (especially
Dr. Fortier and Dr. Dudle), Frank Frey and C. Lively developed this series
of exercises. This program allows you to set parameter values (fitness of
genotypes, mutation & migration rates, population size, etc.) and watch
allele frequencies change through time in a series of simulated
populations. These exercises should solidify your understanding of how
selection, mutation, migration, and drift affect the evolutionary process.
Additionally, you should come away with an understanding of how beneficial
or deleterious recessive alleles may or may not persist in a population
depending on the evolutionary forces at work and understand the effects of
interacting evolutionary forces on the evolutionary process. How to set values in the program and start a simulation In the menu, select run, then select "new run".
With the mouse, click on the box neighboring the value you would like to
set (or use the TAB button)
Type in the new value.
When all values are set, click the 'run simulation' button to start the
simulation
Parameters and explanations (alleles are italicized in this exercise) Initial frequency of allele A (not necessarily dominant - depends on
relative fitness values)
wAA, wAa, waa: Relative fitness of each genotype in the population (may
vary between 0 & 1)
Migration rate: Number of migrants from source population per generation to
your population
Mutation rate of allele A to allele a. This may vary between 0 & 1
Mutation rate of allele a to allele A. This may vary between 0 & 1
Number of different populations to simulate at the same time (may vary
between 1 & 10)
Number of generations the simulation will run through (may vary between 1 &
10000)
Population Size: Size of each simulated population (may vary between 1 &
10000) NATURAL SELECTION
Consider a simple case of overdominance alone (no other evolutionary
forces) using the parameters: p = 0.01, wAA = 0.9, wAa = 1.0, and waa =
0.9, where p is the frequency of the A allele.
1) Why is this a case of overdominance?
2) Will allele A get more or less common through time?
3) Will allele A be fixed (p = 1), lost (p = 0), or maintained at some
intermediate frequency?
4) On the figure below, graph your prediction of the frequency of allele
A through time. Consider the shape of the curve as you do so (e.g.,
Is the spread or loss of the allele constant? Does the rate of spread
or loss of the allele slow down or speed up over time?, etc.). Is the
output from Popgen, the dashed line is the analytical prediction,
assuming an infinite population size. Simulation In the popgen simulation, set the number of populations to 1. Set
Population Size to 10000 (so the effect of drift is small). Set the
initial frequency of the A allele to 0.01. Set the mutation and migration
values to zero.
Now set wAA = 0.9, wAa = 1, waa = 0.9. Set the number of generations to
200. Click Okay to run simulation.
5) Do your predictions above match the results of the popgen simulation?
6) In the space below, summarize how overdominance alone affects the
evolutionary process.
Now consider a case where allele A is recessive to allele a. Set p = 0.5
and waa = 1.0. All other parameters remain the same. 7) How do the three fitness values (wAA, wAa, waa) show that allele A is
recessive? Is it a beneficial or deleterious allele?
8) Will allele A get more or less common through time?
9) Will allele A be fixed (p = 1), lost (p = 0), or maintained at some
intermediate frequency?
10) On the figure below, graph your prediction of the frequency of allele
A through time. Consider the shape of the curve as you do so (e.g.,
Is the spread or loss of the allele constant? Does the rate of spread
or loss of the allele slow down or speed up over time?, etc.)
Simulation
Make sure p = 0.5, waa = 1.0, and all other parameters are the same as the
previous simulation. Click on 'run simulation'. 11) On the figure above, sketch the results of the simulation. Do your
predictions match the results?
12) In the space below, summarize how recessiveness alone affects the
spread or loss of an allele in the population.
13) How would you increase the strength of selection against allele A?
Try simulating this situation. How does the strength of selection
affect the rate of spread or loss of a recessive allele?
14) How could you change the genotype fitness values to make allele A a
beneficial recessive allele? Try simulating this situation.
Summarize the differences (at least 2) in evolutionary trajectory
between a situation where allele A is a beneficial recessive and where
allele A is a deleterious recessive.
MUTATION
Consider a simple case of mutation alone. Things will happen extremely
slowly if we use realistic mutation rates, so pretend our study population
is at Love Canal, Chernobyl, or Three Mile Island. Set u (Ato a) = 0.1, u
(a to A) = 0.01, p = 0, wAA = wAa = waa = 1. All other parameters remain
the same. 15) Is mutation from allele A to allele a deleterious, beneficial or
neutral?
16) Is mutation from allele a to allele A deleterious, beneficial or
neutral?
17) Will allele A get more or less common through time?
18) Will allele A be fixed (p = 1), lost (p = 0), or maintained at some
intermediate frequency?
19) On the figure below, graph your prediction of the frequency of allele
A through time. Consider the shape of the curve as you do so (e.g.,
Is the spread or loss of the allele constant? Does the rate of spread
or loss of the allele slow down or speed up over time?, etc.) Simulation Make sure all parameter values are set as described above. Click on 'run
simulation'. 20) On the figure above, sketch the results of the simulation. Do your
predictions match the results? Press control C to continue the
simulation for another 200 generations.
21) In the space below, summarize how mutation alone affects the spread or
loss of an allele in the population. 22) Try some more simulations with lower mutation rates. For example, try
u (A to a) = 0.01 & u (a to A) = 0.001. How do decreased mutation
rates affect the time it takes for an allele to reach an intermediate
frequency? Compare the results of the mutation simulations to the selection
simulations. Note that mutation alone is a very weak evolutionary force
relative to natural selection.
NATURAL SELECTION AND MUTATION
Even though mutation alone is a weak evolutionary force, mutation is an
extremely important evolutionary force. These next two examples will
illustrate the effects of mutation when combined with selection on the
evolutionary process.
CASE I Consider a case of mutation - selection balance and the loss of deleterious
recessive alleles. Set p = 0.5, wAA = 1, wAa = 1, waa = 0.5, u (A to a) =
0.001, u (a to A) = 0. All other parameters remain the same as before.
23) Will allele A be fixed (p = 1), lost (p = 0), or maintained at some
intermediate frequency?
24) On the figure below, graph your prediction of the frequency of allele
A through time. Consider the shape of the curve as you do so (e.g.,
Is the spread or loss of the allele constant? Does the rate of spread
or loss of the allele slow down or speed up over time?, etc.) Simulation Make sure all parameter values are set as described above. Click on 'run
simulation'. 25) On the figure above, sketch the results of the simulation. Do your
predictions match the results?
There are two possible reasons why allele A was not fixed in this
population (note: these are not mutually exclusive). First, mutation from
A to a may be recreating allele a faster than selection takes it out.
Second, allele a is 'completely recessive'. This means that heterozygotes
do not suffer a fitness cost because they have one copy of allele a. These
next two simulations will test two hypotheses for the maintenance of allele
a in this population. Hypothesis I: Mutation prevents the rapid fixation of allele A 26) Set the mutation rates to zero, leaving all other parameters the same.
Re-run the simulation and compare your results to the previous
simulation (with mutation rate A to a = 0.001). Does mutation alone
explain the maintenance of allele a in the population?
Hypothesis II: Allele A is not fixed in the population because allele a is
completely recessive 27) Leaving the mutation rates at zero, change wAa to 0.9. This makes
allele a 'partially dominant', meaning that one copy of allele a has a
small deleterious effect on the heterozygote. Re-run the simulation
and compare your results to the above two simulations. Does the
'complete recessiveness' of allele a alone explain the maintenance of
allele a in the population?
28) In terms of evolutionary dynamics, what is the difference between a
'completely recessive' allele and a 'partially dominant' allele? To understand the importance of mutation, reset the mutation rates to u (A
to a) = 0.001 and u (a to A) = 0. Leave all other parameters the same
(allele a is still 'partially dominant'). Re-run the simulation. 29) Compare these results to