Download Numbers of - TeacherWeb

LAB #12 - POPULATION GENETICS1
Objectives: This is an observational lab, wherein you will do simulations that help
to understand the basic principals of population genetics. When you finish this lab,
you should:
1. Understand the underlying genetic principals behind the HardyWeinberg equilibrium equation
2. Understand the role of assortative mating in the disruption of the
Hardy-Weinberg equilibrium
3. Understand the role of large population size in the maintenance of the
Hardy-Weinberg equilibrium
I. Background:
Darwin’s theory of evolution by means of natural selection, as developed in The
Origin of Species (1859), was incomplete. No one knew the source of heritable
variation in organisms, nor did any one know the mechanism for transmitting
variation from parents to offspring. The rediscovery of Mendel’s laws of
inheritance (1900), together with DeVries’ discovery of mutations (1903), and
Bovari and Sutton’s insight linking gene transmission with chromosome behavior in
meiosis (1902) provided the missing elements. Today, we refer to the combination
of evolution through natural selection and genetics as neo-Darwinism.
To fully understand the connection between evolution and genetics, we need to
first review two basic concepts, phenotype and genotype. The phenotype of an
individual is a compilation of it’s measurable attributes, both external (such as
flower color, stem length, number of seeds) and internal (such as photosynthetic
rate, presence or absence of CAM respiration). These phenotypic attributes may
be due to genetic make up, but they may also be strongly influenced by
environment: for instance the number of seeds produced by a plant is influenced
strongly by environmental factors such as rainfall and soil nutrients. Evolution is
defined as changes in a population’s mean phenotype over time. The emphasis in
evolutionary biology must be on the population because although the unit of
selection is the individual, which may or may not survive and reproduce, an
individual’s heritable phenotype cannot change through evolution; it is fixed at
birth! The measurement of evolutionary change is the change in the mean
1 Parts of this exercise are modified from: McCourt, R.M. 1988. Laboratory
Manual. Random House, New York
1
phenotype of a population.
It then follows that if there are changes in the heritable phenotypes in a
population, there should be changes in the relative frequencies of the alleles
causing these different phenotypes in that population. Populations are thus
characterized as having a mean phenotype, and having certain allele frequencies
that produce the mean and variation in phenotype. A population can be described
as the sum total of all the genes possessed by all of the individuals in that
population at any one time; that is, it is seen as a gene pool. When alleles change
frequency within such gene pools - accompanied by corresponding changes in mean
phenotype - evolution has occurred.
In 1908, G.H. Hardy, a British mathematician, and W. Weinberg, a German
physician, independently pointed out that the frequency of the possible
combinations of a pair of alleles at a particular genetic locus in a population is
described by the expansion of the binomial (p + q)2. In the derivation, we let
p = Frequency of Dominant allele = freq (A)
q = Frequency of Recessive allele = freq (a)
and
p+q=1 (there are no other alleles)
Then,
Frequency of genotype AA = p2
Frequency of genotype Aa = 2pq
Frequency of genotype aa = q2
All of the possible genotype frequencies must add up to 1, so
p2 + 2pq + q2 = 1
or
(p+q)2 = 1
Further, these authors pointed out that in a population, the allele frequencies will
remain constant from generation to generation if certain conditions are met. This
constancy of allele frequencies over time is fundamental to the modern study of
evolution because deviation from constancy is often a researcher’s first indication
that evolution is taking place. The conditions under which allele frequencies are
constant are now called Hardy-Weinberg conditions.
The Hardy-Weinberg conditions are:
2
1. Random mating. The choice of mating pairs is not influenced by the phenotype
of the individuals, but is governed by the rules of chance. Remember that mating
choice need only be random with respect to the trait being studied.
2. No mutation or migration. The allele frequencies must not be changed because
of mutations or because individuals with unusual allele frequencies are leaving or
entering the group.
3. Large population. In small populations, random events can significantly alter
allele frequencies. This random process is called genetic drift. To avoid drift, the
Hardy-Weinberg conditions include assumption of an infinitely large population
number.
4. No selection. The various phenotypes produced by each genotype should cope
equally well with the environment and be equally fertile at the time of
reproduction.
If any one of these conditions is not met, the Hardy-Weinberg equilibrium will be
upset, and a population will be evolving, not static. In this lab we will use “beanbag” genetics models to demonstrate the Hardy-Weinberg model and examine the
effects of violation of the assumptions of random mating and infinite population
size (sexual selection and genetic drift models of evolution).
PROCEDURE
PART I. DEMONSTRATION OF HARDY-WEINBERG EQUATION
A. Allele frequencies in human populations. It is possible to use data from your
class to determine the allele frequencies for common genetic traits. Several
human traits are determined simply by one genetic locus with two alleles that
interact in Mendelian fashion (i.e., one dominant and one recessive allele). You will
determine your own phenotype for each of these characteristics, then collect the
data from the entire class.
Traits:
i) Ear lobes may be free (detached) or attached to the head. The dominant allele
is “detached”.
ii) “Hitchhikers” thumb is a trait wherein the middle joint of the thumb is hyperextensible, so that the thumb bends back at nearly right-angles to the plane of the
hand. The alternative phenotype is a straight thumb that does not hyper extend.
3
The dominant allele is “hitchhikers”.
iii) Widow’s peak is when the hair line across the forehead forms a distinct “V” in
the center (note that this is distinct from pattern baldness in men). The
alternative phenotype is a straight hairline, without the “V”. The dominant allele is
“widow’s peak”.
iv) Dimpled chin is a distinct depression in the middle of the chin. The alternative
phenotype is a smooth chin. The dominant allele is “dimpled”.
Your phenotypes:
Earlobes_______________ Thumb_________________
Hairline________________Chin________________
Calculating class allele frequencies:
Table 1. Phenotype and allele frequencies of four traits for the entire class
# dominant
phenotype
# recessive
phenotype
frequency
(dominant
allele)
frequency
(recessive
allele)
Ear lobes
Thumb
Hairline
Chin dimples
Sample calculation: We cannot know the frequency of the dominant allele (why?)
and must therefore calculate it using the Hardy Weinberg equation. If the number
of students with attached earlobes (the recessive allele) is 9 out of 20, then we
know that q2 = 9/20 = 0.45 and q = 0.67. We can then calculate p (the frequency
of the dominant allele) as 1-q = 0.33.
Note that in some cases, the dominant alleles and their corresponding phenotype
4
are not the most common phenotype in the population. Allelic “dominance” is not
correlated with population frequency, although this is a common misconception.
5
PART II. HARDY-WEINBERG EQUILIBRIUM
Using split peas, we will simulate a Hardy-Weinberg equilibrium for one locus with 2
alleles. Working with a partner, obtain 100 yellow and 100 green split peas. Each
pea represents an allele (y or g) at the “split pea” gene locus. The imaginary species
is diploid, so each individual in the population is represented by a pair of peas. The
three genotypes will be gg, gy, and yy, and you will decide the initial frequencies of
each. Let p be the frequency of green alleles in the population and q frequency of
yellow alleles. An interesting case can be simulated using intermediate allele
frequencies (p & q between .3 and .7). Geneticists typically call the starting
population Generation 0 (zero). Once you have established your initial conditions,
you will use random “matings” to generate subsequent generations, and then record
the genotypic frequencies and allele frequencies for each of these descendent
generations. If our “split pea” species is in Hardy-Weinberg equilibrium, what
should happen with allele frequencies over time? Genotype frequencies?
a) Generation 0 (zero). We will start with an initial population of 80 individuals
using 160 peas. Decide on the initial genotype frequencies, calculate allele
frequencies and then calculate how many peas of each color you will need for
your starting population (i.e. multiply p x 160 and q x 160). Count the right
number of yellow and green peas and put them in a container. Record the
initial frequencies of genotypes and alleles in the “Generation 0” row of
Table 2.
b) The 160 peas in your beaker represent the gene pool, or the frequencies of
the two gamete types produced by Generation 0. During reproduction, 80
pairs of gametes will be drawn from this gene pool to create the next
(diploid) generation. We now need to simulate the process of reproduction
among the 80 members of our population. In real life, each individual would
produce perhaps hundreds or thousands of gametes and mate multiple times
(human females produce relatively few gametes. Imagine these are corn or
fruit flies). From your beaker, randomly select 2 peas. Record the genotype
on a piece of scratch paper then return the 2 peas to the container (so that
the “parents” can reproduce again). Repeat this process until you have
tallied a total of 80 offspring, simulating random mating among your 80
individuals to produce an offspring generation with the same population size.
In Table 2, generation 1, record the numbers of each genotype, the
numbers of each allele and the frequency of each allele.
c) repeat step (b) to produce genotype and allele frequencies for generation 2.
6
Table 2. Genotype and allele frequencies by generation for simulation 1: HardyWeinberg equilibrium
Generation
gg
Numbers of:
Genotypes
Alleles
gy
yy
g
y
Frequencies
p
q
0
1
2
Results:
Describe what happened to genotype frequencies in going from the starting
population through successive generations. Did allele frequencies change?
To determine if any deviations you observed are significant, we will use a chi-square
test, as in the Mendelian genetics lab (#8)
7
Table 3: Chi-square test for Hardy-Weinberg equilibrium with no selection
genotype
gen. 0
expected
frequency count, gen 2
observed
count, gen 2
O-E
(O-E)2
E
gg
gy
yy
totals
Chi-square calculations:

2
= ∑ (observed count - expected count)2
expected count
Look up the probability for the calculated value of X2 in table provided at the end
of this lab.
p<
<p
PART III. NON-RANDOM MATING
In most plants and animals non-random mating is the rule rather than the
exception. Males and females may choose their mates using a variety of criteria,
such as color, body size, territory size, or display behavior. Even when mate choice
is not occurring, differential survival of particular phenotypes in particular
environments may increase the probability that organisms with similar genotypes
mate. In our bean-bag model, we will simulate non random-mating by subdividing
the gene pool according to parental genotype/phenotype, and allowing only
individuals within each gene pool to mate with each other.
(a) Start with a population of 80 with the equilibrium count and genotypes
calculated for part I, generation 0. Assume that the g allele is dominant
over the y allele, so that gg and gy individuals have the phenotype of “green”.
Record the starting genotype frequencies and allele frequencies in Table 4,
generation 0. We will assume that organisms prefer to mate with other
8
organisms of their own phenotype, positive assortative mating. In nature
such a preference would not be 100 percent effective, but we can simulate
perfect compliance by taking all the beans for gg and gy individuals and
putting them in one container, and putting all the beans for yy in another.
These two containers represent separate gene pools in the population.
(b) Simulate the mating of organisms having the green phenotype by drawing 60
pairs of beans from the gg/gy container and tallying the genotypes on a
piece of scratch paper. Because we assume that individuals can mate more
than once and produce multiple offspring, be sure to replace and mix the
peas between draws. We do not need to simulate mating in the yy container
because all the offspring in this case must be yy. There will be 20 offspring
from yy matings. The combined results of the gg/gy and yy matings
represent the generation 1 offspring for the entire population (be sure to
add the results of the two gene pools together). Record these genotype
frequencies and allele frequencies in the table 3, row “1”.
(c) Adjust the allele frequencies in each container to reflect the gene pool for
generation 1: this will involve taking yellow peas from the “green” gene pool
and putting them in the “yellow” gene pool, reflecting the number of “yellow”
offspring produced (why would the “green” gene pool produce yellow
offspring?). Be sure that the total number of individuals in the population is
80. Repeat the step (b) with these new gene pools to produce generation 2.
Tally the genotypes of the second round of matings on scratch paper and
record the combined second-generation results in Table 4.
Table 4. Genotype and allele frequencies by generation for simulation 3: Effects
of non-random mating on genotype frequency and allele frequency.
Generation
Numbers of:
Genotypes
Alleles
gg
gy
yy
g
y
Frequencies
p
q
0
1
2
9
Results
How does non-random mating affect the frequency of each genotype?
Are the results significant? To answer this question, compare allele frequencies in
generation 2 to those in generation 0 using a Chi-squared test:
Table 5: Chi-square test for Hardy-Weinberg equilibrium with assortative mating
genotype
gen. 0
expected
frequency count, gen 2
observed
count, gen 2
O-E
(O-E)2
E
gg
gy
yy
totals
Chi-square calculations:

2
= ∑ (observed count - expected count)2
expected count
p<
<p
PART IV. GENETIC DRIFT
In the previous exercises we have kept the size of the breeding population
constant and fairly high. In nature, reproductive populations may go through
periods of greatly reduced numbers. This can occur through natural events, such
as dispersal of a few individuals to a new habitat like an island, or disease. These
“bottlenecks” can also happen through man-made destruction of natural habitat. In
any case, the population is reduced often without any difference among phenotypes
in survival (i.e., we are assuming that the mortality is not biased towards certain
members of the population). When only a small number of larger population can
mate, there is potential for significant genetic drift - sharp changes in gene
frequency caused by sampling error or the random survival of just a few, randomly
10
selected individuals that likely do not have the same allele frequencies as the
initial, large population. We will simulate genetic drift by taking a small subsample
of individuals from the part I, generation 0.
Methods:
Return the beans to the starting population used in Part I by mixing 80
pairs of beans with the required genotypes in one container. This
represents the gene pool of the initial population. Pick out four pairs of
beans at random and record their genotypes in Table 6. These pairs
represent a pioneer population of four individuals. Replace the beans in
the container, mix well, and repeat the process for four more pioneer
populations (a total of five pioneer populations). Record the genotypes
of the four pioneers in each population in the chart. Then calculate the
genotype frequencies and allele frequencies in each of these populations.
Record the genotype and allele frequencies in Table 6.
Table 6. Genotype and allele frequencies by population for part IV: Effects of
genetic drift.
Population
gg
Numbers of:
Genotypes
Alleles
gy
yy
g
y
Frequencies
p
q
gen 0
1
2
3
4
5
Results:
Compare the frequencies of genotypes and alleles in the pioneer populations to
those in the starting populations. Genetic drift can cause the fixation or
extinction of alleles in the pioneer population. Fixation occurs when a single allele
11
achieves a frequency of one; extinction occurs when the frequency of an allele
drops to zero. Did you observe this in any of your pioneer populations?
How would you test whether significant genetic drift has occurred in each of your
new populations?
12