Gene Frequencies

Outline
• Objective

• Introduction

• Allelic and Genotypic Variation

• Hardy-Weinberg Equilibrium

• Factors Affecting Allele Frequency

• Summary
Objectives
• Demonstrate the relevance of population genetics concepts to plant breeding
populations.

• Demonstrate the relevance of a purely theoretical Ideal Population to plant

breeding populations.

• Demonstrate understanding for purpose of populations in Hardy-Weinberg

Equilibrium

• Distinguish populations in Hardy-Weinberg Equilibrium from the Ideal

Population.

• Describe impact of mutation, selection and drift on breeding populations.

Introduction
• The challenge of Quantitative Genetics is to connect traits measured on
quantitative scales with genes that are inherited and evaluated as discrete units.
• Population genetics characterizes how discrete units, i.e., alleles, change in breeding
populations.
• Such characterization is the basis for understanding structure of genomes and breeding
populations.

• The forces of mutation, migration, selection and drift will alter the structure of
breeding populations.

• Herein we will learn how to characterize population structure at one or two loci in
diploid crop species.
Allelic and Genotypic Variation
Ideal Population
• Quantitative traits are, by necessity, studied
within the context of a population.
• Knowledge of how genes behave in
populations is therefore fundamental regardless
of whether the genes are known or whether
they represent unknown alleles at a generic
quantitative trait locus.
• In order to understand the genetic structure of
a population it is necessary to establish a
standard reference population so that the
breeding population can be characterized Fig. 1 Reference population. Adapted from Falconer
relative to the standard. and Mackay, 1996.

• For this purpose an ‘ideal’ conceptual base population can be defined as infinitely large with
potential to extract finite sub-populations through sampling such as depicted in the following
figure and described in Falconer and Mackay (1996): 
• Population is a group of interbreeding individuals that exist together in time and space
• Breeding populations are created by breeders to serve as a source of cultivars that meet specific
breeding objectives.
Ideal population assumptions
• In the ideal population depicted in Fig. 1, the
following assumptions are true:
1. The base population is infinite, or at least too
large to count.
2. There is no migration between sub-
populations.
3. There is no breeding between overlapping
generations.
4. The number of breeding individuals is the
same in each sub-population.
5. There is random mating within a sub-
population.
6. There is no Selection.
7. There is no Mutation.
• Of course in real populations these assumptions are violated.
Allelic and Genotypic Frequencies
•• A
  population can be characterized in terms of its
genotype frequencies and allele frequencies at a
locus.
• We first model a single locus with only two alleles
in an ideal breeding population of diploid
individuals. Define the following:
• N = number of breeding individuals in a sub population
(population size)
• t = time usually measured in terms of generations
• q = frequency of one of two alleles at a locus within a sub
population
• p = 1 – q = frequency of a second allele at a locus within a
sub population
• = frequency of a second allele across the sub populations
(the mean of p)
• = frequency of a second allele in the base population
 • Due to the assumptions associated with an ideal reference population, at any
stage or generation of the sampling process, so can be used interchangeably with .
Allelic and Genotypic Frequencies…
Table 1. Frequencies of AA, Aa and aa individuals
  Alleles Genotypes Total
  A a AA Aa aa
Number 240 240 120 600
Frequency  
Frequency after random mating

••   Suppose a diploid breeding population is segregating at a locus with two alleles, “A” and “a”
(Table).
• The genotype frequencies refer to the proportion of individuals that have a particular genotype.
• A population of 600 diploid individuals is equivalent to a population of 1200 alleles.
• In the example above, the genotype frequencies are PAA = 240/600=0.40 for the AA homozygote,
PAa=240/600=0.40 for the Aa heterozygote, and Paa=120/600=0.20 for the aa homozygote.
• The relationship between allele frequencies and genotype frequencies can be expressed as: 
• The frequency of the dominat allele:

• The frequency of the recessive allele:

• Where, p + q = 1 and PAA + PAa + Paa = 1.​
Allele frequencies in breeding populations
• 
• Allele frequencies are generally unknown in breeding populations created from
non-inbred parents or from three or more inbred parents.
• But breeding populations in both self-pollinated and cross-pollinated crops are
often created by crossing two inbreds.
• As such the allele frequencies at segregating loci are known even if the alleles
themeslves are unknown.
• In F1 anf F2 populations:
• The allele frequencies are at all loci that differ between the two parental inbreds

• In backcross generation to either parents: suppose one parent has the A1A1 genotype
where as the other parent has the A2A2 genotype
• BC1 to A1A1 >>>>>
•
Variance of Allele Frequency
• 
• A particular sub-population is a random sample of N individuals or 2N
gametes (for a diploid) from the base population.
• Therefore, the expected gene frequency of a particular allele in the sub-
populations is q0 and the variance of q is

• Since q0 is a constant, the variance of the change in allele frequency (q1 -
q0) is also:
Frequency Estimators
••  
In addition to the genetic sampling process depicted in Fig. 1, a statistical sampling
process can be used to estimate frequencies, variances and covariances of alleles
and genotypes in a sub-population.

• If we sample n individuals from a population of size N, and notationally let

then and

is the sample frequency of the A allele and

is the sample frequency of genotype AA.

Hardy-Weinberg Equilibrium
HWE: Introduction
•• G.H.
  Hardy was a British mathematician and W. Weinberg was a German
physician who, in 1908, independently deduced the relationship between allele
frequencies and genotype frequencies under random mating.

• Three key features of Hardy-Weinberg equilibrium are

1. The allele frequencies remain constant from generation to generation

2. The square of the array of the allele frequencies is equal to the array of genotype
frequencies, i.e., .

3. If allele frequencies change due to external factors, one generation of random mating will
lead to a new set of equilibrium genotype frequencies.

• Conditions need for Hardy-Weinberg equilibrium are random mating, a large

population and the absence of selection, mutation and migration.
HWE: Introduction…
•  Notation (single-locus)
• Probability of observing each state of an allele

• Probability of observing each state of a locus (genotypic frequencies)

 • Two-loci case:
• Allele:
• Single-locus expression: • Loci:
• Sum of probabilities always add to 1 • Takes many generations for neighbor loci HWE
• Allele:
• Locus:
• FOR BREEDING: Allelic and genotypic frequencies from multiple loci across the
HWE: Introduction…
• The
  figure shows several characterstics of random-
mated and random-mated backcross populations
1. The frequency of heterozygotes in a population in
HWE is maximum when . This result indicates that the
proportion of heterozygotes is maximum in an
population.
2. If p is greater than 2/3, then the proportion of
heterozygotes is intermediate between the proportions
of the two homozygotes, i.e., . This result indicates that
if a BC1 or any other backcross generation is random
mated, the resulting proportion of Aa gentotypes will
always be intermediate to the proportions of the AA
Figure 1. Genotype frequency at Hardy- and aa genotypes.
weinberg equilibrium 3. If an allele is rare, then that allele will be present in
mostly in heterozygotes rather than in homozygotes
https://en.wikipedia.org/wiki/File:Hardy-Weinberg.svg

FOR BREEDING: Observed within (but not across) F2 populations. Used to create
genetic maps and to estimate segregation distortion. HWE is often assumed to
simplify calculations.
Assumptions
• The
  proof of HWE requires the following assumptions (Falconer and Mackay,
1996):
• Allele frequency in the parents is equal to the allele frequency in the gametes
• Assumes normal gene segregation
• Assumes equal fertility of parents
• Allele frequency in gametes is equal to the allele frequency in gametes forming zygotes
• Assumes equal fertilizing capacity of gametes
• Assumes large population
• Allele frequency in gametes forming zygotes is equal to allele frequencies in zygotes
• Genotype frequency in zygotes is equal to genotype frequency in progeny
• Assumes random mating
• Assumes equal gene frequencies in male and female parents
• Genotype frequencies in progeny does not alter gene frequencies in progeny.
• Assumes equal viability
• For a two allele locus in a population in HWE:
,,
• HWE at a given genetic locus is achieved in one generation of random mating. Genotype
frequencies in the progeny depend only on the allele frequencies in the parents and not on
the genotype frequencies of the parents.
Disequilibrium
• 
• As discussed there are several processes that can force allelic
and genotypic frequencies to deviate from HWE.

• Deviations from equilibrium are referred to as Disequilibrium, and are often

denoted with a disequilibrium coefficient, D.

• In the two allele case the genotypic frequencies can be represented as

;;

Thus,
Non-Random Mating
• Two methods of non-random mating that are important in plant
breeding are:
• Assortative mating
• Occurs when similar phenotypes mate more frequently than they would by chance
• One example would be the tendency to mate early x early maturing plants and late x late
maturing plants.
• The effect of assortative mating is to increase the frequency of homozygotes and decrease
the frequency of heterozygotes in a population relative to what would be expected in a
randomly mating population.
• Assortative mating effectively divides the population into two or more groups where
matings are more frequent within groups than between groups.
• Disassortative mating
• Occurs when unlike or dissimilar phenotypes mate more frequently than would be
expected under random mating.
• Disassortative mating leads to an excess of heterozygotes and a deficiency of
homozygotes relative to random mating.
• Disassortative mating can also lead to the maintenance of rare alleles in a population.
Factors Affecting Allele Frequency
Introduction
• The factors affecting changes in allele frequency can be
divided into two categories: 
• Systematic processes, which are predictable in both magnitude
and direction  
• Migration,

• Mutation, and

• Selection.

• Dispersive processes, which are predictable in magnitude but not

direction.
Systematic processes affecting changes
in allele frequency
Migration
• In population genetics, gene flow (also known
as gene migration or allele flow) is the transfer
of genetic material from one population to
another.
• Migrants change the distribution of genetic
diversity among populations, by
modifying allele frequencies (the proportion
of members carrying a particular variant of a
gene).
• High rates of gene flow can reduce the genetic
differentiation between the two groups,
increasing homogeneity
• In some cases dispersal resulting in gene flow
may also result in the addition of novel genetic
variants under positive selection to the gene
pool of a species or population (adaptive
introgression.)
 Migration…
•  Assume a population has a frequency of m new immigrants each
generation, with 1-m being the frequency of natives.
• Let  be the frequency of a gene in the immigrant population and q0 the
frequency of the same gene in the native population. Then the
frequency in the mixed population will be: 

Factors Affecting Allele Frequency…Migration

• The change in gene frequency brought about by migration is the
difference between the allele frequency before and after migration.

• Thus the change in gene frequency from migration is dependent on

the rate of migration and the difference in allele frequency between the
native and immigrant population.
Migration and gene flow
• RELEVANCE: Exogenous sources of alleles
from other populations
• FOR NATURE: Outbreeding
• FOR BREEDING: Occurs intentionally or
unintentionally
1. exchange of genetic material among
breeders;
2. incorporation of new material;
3. seed or pollen contamination.
• WATCH OUT FOR:
• Contaminations in small population with
high genetic value: crossing blocks and seed
production.
• Gene flow when managing genetically
modified material.
Mutations
• Mutations are the source of all genetic variation. Loci with only one allelic variant in a
breeding population have no effect on phenotypic variability.

• While all allelic variants originated from a mutational event, we tend to group mutational
events in two classes:
• rare mutations and

• recurrent mutations (mutation occurs repeatedly).

1. Rare mutations:
• only occurs very infrequently in a population.

• Therefore, the mutant allele is carried only in a heterozygous condition and since mutations are usually
recessive, will not have an observable phenotype.

• usually be lost, although theory indicates rare mutations can increase in frequency if they have a selective
advantage.
Fate of single mutation
• Consider
  a population of only AA individuals. Suppose that one A allele in the
population mutates to a.
• Then there would only be one Aa individual in a population of AA individuals.
• So the Aa individual must mate with a AA individual.

• From Li (1976; pp 388), this mating has the following outcomes:

• No offspring are produced in which case the mutation is lost.
• One offspring is produced: the probability of that offspring being AA is 1/2 so the probability
of losing the mutation is 1/2.
• Two offspring are produced: Aa can mate with more than one of the AA individuals in the
population, thus if Aa mates with two AA individuals, the probability of both offspring being
AA is 1/4, so the probability of losing the mutation is 1/4.
• If k is the number of offspring from the above mating then the probability of
losing the mutation among the first generation of progeny is (1/2)k.
Mutation…
•2.  Recurent Mutations
• Let the mutation frequencies be:

• Then the change in gene frequency in one generation is:

at equilibrium
Mutation…
••  
RELEVANCE: Mutation is the only natural source of new variation in all species.
Important for phylogenetics to trace speciation.

• RANDOMNESS: Mutation are more likely to occur in some parts of the genome,
such as euchromatin (less condensed DNA).

• CHANCES: Mutagenic agents increases the probability of mutation from to

• FOR BREEDING: Mutational breeding is deprecated in crops, but it is still big deal
in some clonally propagated species. The updated version is called ‘gene editing’
based on CRISPR-Cas9 (site specific changes as opposed to random modification).

• WATCH OUT FOR: Most mutation have deleterious effects.

Selection
• Selection is one of the primary forces that will alter allele frequencies in
populations. 
• Selection is essentially the differential reproduction of genotypes.
• In population genetics this concept is referred to as fitness and is measured by
the reproductive contribution of an individual (or genotype) to the next
generation.
• Individuals that have more progeny are more fit than those who have less
progeny because they contribute more of their genes to the population.
• The change in allele frequency following selection is more complicated than
for mutation and migration, because selection is based on phenotype.
• Thus, calculating the change in allele frequency from selection requires
knowledge of genotypes and the degree of dominance with respect to fitness.
• Selection affects only the gene loci that affect the phenotype under selection—
rather than all loci in the entire genome—but it also would affect any genes
that are linked to the genes under selection.
General types of selection
• Types
• Directional selection: One direction increases the fitness: either higher
and lower vales, but not both.
• Stabilizing selection: An equilibrium exist and the population is better
off under an intermediate phenotypic value (or because extremes are
lethal). an average phenotype is favored
• Diversifying (or Disruptive): Either extreme is a better fit than an
intermediate form.
• FOR BREEDING:
• Yield undergoes positive directional selection (more is better), while
composition and maturity often undergo stabilizing selection (keep
germplasm within target maturity).
• WATCH OUT FOR:
• Most traits of interest are genetically correlated, the best phenotype is
provided by combination of traits: breeders must manage genetic
tradeoffs (often through selection indexes).
Effects of Selection
• Change in allele frequency
The strength of selection is expressed as a coefficient of selection, s, which is the
proportionate reduction in gametic output of a genotype compared to a standard
genotype, usually the most favored. Fitness (relative fitness) is the proportionate
contribution of offspring. 
• Partial selection against a completely recessive allele
To see how the change in allele frequency following selection is calculated
consider the case of selection against a recessive allele:

Genotypes
  AA Aa aa Total
Initial Frequencies
Coefficient of Selection
Fitness
Gametic
Gametic Contribution
Contribution
Frequency Equations
•  The frequency of allele a after selection is:

The change in allele frequency is then:

Selection and allele frequency
• RISE AND FATE OF ALLELES:
• Maximum variability point occurs at p=0.5

• Overtime, high fitness alleles trend to go to fixation whereas low fitness allele trend to go to
extinction

• CONSEQUENCE: Phenotype of the population shifts towards the most fit balance.

• FOR BREEDING:
• Allele come from introducing new lines (migration rather than mutation).

• Breeders’ job is to sort out which allele are valuable for the program.

• WATCH OUT FOR: Pushing selection too hard to avoid compromising long term
genetic gains
39
Dispersive processes affecting changes
in allele frequency
Small Population size
• Unlike the three systematic forces that are predictable in both amount
and direction, changes due to small population size are predictable
only in amount and are random in direction.
• The effects of small population size can be understood from two
different perspectives.
• It can be considered a sampling process and
• it can be considered from the point of view of inbreeding.
• Consequences of small population size
• Random genetic drift: random changes in allele frequency within a
subpopulation
• Differentiation between subpopulations
• Uniformity within subpopulations
• Increased homozygosity
Inbreeding and Small Populations
• Inbreeding is the mating together of Gen. Ancestors
individuals that are related by ancestry. 0 1
• The degree of relationship among 1 2
individuals in a population is 2 4
determined by the size of the 3 8
population. This can be seen by
4 16
examining the number of ancestors that
a single individual has: 5 32
• Just 50 generations ago note that a single 6 64
individual would have more ancestors than 10 1,024
the number of people that have existed or
could exist on earth. 50 1,125,899,906,842,620

• Therefore, in small populations 100 1,267,650,600,228,230,000,000,000,

000,000
individuals are necessarily related to
one another. t 2t
Identical by Descent and Identical by State
• In finite populations there are
two sorts of homozygotes:
• Those that arose as a consequence
of the replication of a single
ancestral gene — these genes are
said to be identical by
descent (Bernardo, 1996).
• If the two genes have the same
function, but did not arise from
replication of a single ancestral
gene, they are said to be alike in
state.

• It is the production of homozygotes that are identical by descent that

gives rise to inbreeding in a small population.
Genetic Drift – Sampling effect
• Genetic drift is a mechanism of evolution in which
allele frequencies of a population change over
generations due to chance (sampling
error). Genetic drift occurs in all populations of
non-infinite size, but its effects are strongest in small
populations.
• FOUNDER EFFECT: The magnitude of genetic drift
is aggravated under sampling bottlenecks – when a
new population is derived from a non-
representative subset of the original population.
• CONSEQUENCE: Diversity losses and random
changes in the genomic composition. Effects of drift
in the genome can be easily confounded with
selection.
• FOR BREEDING: Most breeders are not concerned about drift. Much of the selection
performed in early generations causes drift. Having the same family being selected in
separate experiments mitigates drift.
• WATCH OUT FOR: Unlike practical breeding per se, geneticist that attempt to infer
signatures of selection should pay close attention to drift.
Brief recap to Population Genetics
Random Genetic Drift
 Reduction of population size

Infinitely large population Sample

Founder effect
In population genetics, the founder effect is the loss of genetic variation that
occurs when a new population is established by a very small number of
individuals from a larger population.
Loss of diversity: Bottlenecks

A population bottleneck or genetic

bottleneck is a sharp reduction in the size of
a population due to environmental events
such as famines, earthquakes, floods, fires,
disease, and droughts or human activities.
Mutation and Selection
New variation
When new variation (mutation) Extinction
affects fitness

Fixation
 If blue is a desirable trait for
BREEDING…
Directional
Mutation Selection Fixation

Time
Migration
Gene flow
Sub-populations
and
Inbreeding
Inbreeding
Genetic Diversity
Wild relatives
Germplasm collection
Summary
• Write three new things you learned thus far.
• Write one more thing you wish to learn before the last date
of the course
• Linkage disequilibrium
• Gene mining
• F-statistics
• Examples – case studies
• GXE, Heterosis, Combining ability, Data analysis
• Selection index
• Marker traits association (GWAS)
• Linkage mapping
References
• Beavis, W. and K. Lamkey. 2016. Gene Frequencies. In Quantitative
Genetics, interactive e-learning courseware. Plant Breeding E-
Learning in Africa.
• Falconer, D.S., and T.F.C. Mackay. 1996. Introduction to quantitative
genetics. 4th ed. Pearson, Burnt Mill, England.
• Bernardo, R. 2010. Breeding for Quantitative Traits in Plants. 2nd ed.
Stemma Press, Minnesota, USA.