Population Biology Simulations
Collected here are a few simple simulations (written in Java) I use
(or plan to use) when teaching principles of population genetics and
population ecology in various courses. If you have suggestions for
improving them or ideas for other simulations that might be useful,
please contact me at kent@darwin.eeb.uconn.edu.
I can't promise that I'll have the time to adopt any suggestions you
make, but I promise that I'll consider them.
Population Genetics
- Mother-offspring transmmission -- Illustration of how mother-offspring
combinations can be used to make inferences about mating patterns and
genetic transmission in populations.
- EM algorithm for ABO frequencies -- This
applet illustrates the EM algorithm for estimating allele frequencies
in the ABO blood system. Users may select from a variety of sample
configurations (including random allocation of phenotypes with three
different sample sizes) and several different starting guesses
(including random frequencies). Results from each iteration are
displayed, but only six iterations can be displayed
simultaenously.
- Genetic drift -- This simulation
illustrates how allele frequencies change over time as a result
of genetic drift in small populations. Users may select
from three different starting allele frequenciese (0.1, 0.5,
0.9), five different population sizes (10, 25, 50, 100, 250),
and three different numbers of generations for the simulation
(50, 100, 250). Results from up to eight simulations are
displayed simultaneously in different colors.
- Natural selection -- This
simulation illustrates how allele frequencies change over time
in response to natural selection on diploid genotypes. Users may
select from five different fitnesses (0.8, 0.9, 1.0, 1.1, 1.2)
for each of the three genotypes. The number of generations is
fixed at 100. Results from up to eight different iterations are
displayed simultaneously in different colors.
- Mean fitness -- This applet illustrates
the relationship between allele frequency and population mean
fitness for a simple one-locus, two-allele model of viability
selection. Users may select from a variety of different
fitnesses.
- Natural selection and genetic
drift -- This simulation illustrates the interaction between
natural selection and genetic drift. Users may select from three
different starting allele frequencies (0.01, 0.05, 0.1), five
different population sizes (10, 25, 50, 100, 250), and three
different numbers of generations for the simulation (50, 100,
250). Only a single set of fitnesses representing selection for
an initially rare allele are employed, specifically
w11 = 1.0, w12 = 0.9, w22 =
0.8.
- Genetic drift and mutation --
This simulation illustrates the interaction between mutation and
genetic drift. Users may select from three different population
sizes (25, 100, 250) and several different mutation rates (none,
0.0001, 0.001, 0.01). 32 populations are simulated
simultaneously and the results are displayed as a frequency
histogram.
- Genetic drift and migration --
This simulation illustrates the interaction between migration and
genetic drift. Users may select from three different population
sizes (25, 100, 250) and several different mutation rates (none,
0.001, 0.01, 0.1). 32 populations are simulated
simultaneously and the results are displayed as a frequency
histogram.
- t-allele polymorphism -- This
simulation illustrates the interaction among drift, selection,
and segregation distortion. Users may select from several
different population sizes and degrees of distortion. 32
populations are simulated simultaneously and the results are
displayed as a frequency histogram.
- Response to selection in a
quantitative trait -- This simulation illustrates the
response to selection in a quantitative trait. Users may select
from several different population means and variances, selective
optima and strengths of selection (through the variance of the
selection function), and heritabilities. The regression between
mid-parent and offspring is shown, individuals surviving a bout
of selection are highlighted in blue, and the selective
differential and response to selection are highlighted in
red.
- Divergence of DNA sequences -- This
simulation illustrates divergence of DNA sequences according to
two simple models of sequence evolution: Jukes-Cantor and Kimura
2 parameter. Users may select from several different
transition/transversion ratios, sequence lengths, and numbers of
samples. In addition, the display can illustrate either the
percent sequence difference as a function of expected number of
substitutions or the calculated distance between two sequences
as a function of expected number of substituions.
Population Ecology
- Stochastic demography -- This
simulation illustrates how to evaluate the persistence
probability of populations with age/stage structure. The default
data is taken from Menges' paper on Furbish's lousewort, but any
Leslie or Lefkovitch model with fewer than 6 stages can be
accomodated.
Important note: The simulations may not display
properly in all browsers. Some of the simulations have alternate
versions that should work in any browser that supports Java. If you
can't display the simulation when loaded, scroll to the bottom of the
page and see if there's a link to an alternate version.
Several users have asked for versions of the simulations that can
be used offline or that they can host on their own servers. Everything
you need to do that should be in simulations.zip. Just download it, unzip it
into a convenient directory and you're all set. The one thing you
might want to do is to change the lines that read
<link rel=stylesheet type="text/css"
href="http://darwin.eeb.uconn.edu/styles.css">
to
<link rel=stylesheet type="text/css"
href="styles.css">
That way your browser won't waste time trying to connect to my website
for the stylesheet. Enjoy!
Kent Holsinger
