Mendelian Genetics in Populations II

1
Mendelian Genetics in Populations II

• Migration, Genetic Drift, and Nonrandom Mating

2
Chance events can alter allele and genotype
frequencies (Fig. 6.10a)
In the sampling of egg and sperm to make
zygotes, there were 14 A1 and 6 A2 alleles. So
the allele frequencies changed to 0.7 for A1 and
0.3 for A2 Note these alleles are not affected
by natural selection neutral alleles
3
Range of possible outcomes when 20 gametes are
sampled from a population with allele frequencies
of p 0.6 and q 0.4 (Fig. 6.11)
4
Genetic drift is evolution that happens by chance

• The figure on the previous slide demonstrates
  that
• Allele frequencies are more likely to change than
  stay the same across generations
• Small per generation changes in allele frequency
  are more likely than large changes
• If the initial frequency of an allele is gt 0.5,
  it is more likely to increase in frequency that
  decrease in the next generation

5
Computer simulations of random genetic drift

• Random sample of 20 gametes from a population
  with p(A1) 0.6 and q(A2) 0.4
• Using Populus 5.3
• http//www.cbs.umn.edu/populus/Download/download.h
  tml

6
Evolutionary effects of genetic drift 1

• Genetic drift results in the eventual fixation of
  one allele, and the loss of all other alleles
  (provided no other forces are acting neutral
  alleles)
• As one allele drifts toward fixation, the
  heterozygosity of the population declines
• The probability of eventual fixation of a neutral
  allele is equal to its current frequency in the
  population (Note in a diploid population of size
  N, a new mutation will have a frequency of 1/2N)
• Larger populations require more time for fixation
• Drift also operates on alleles that are affected
  by selection (more on that later)

7
Genetic drift and heterozygosity (Fig. 6.15)
8
Genetic drift in experimental lab populations107
populations of size N 16, initial allele
frequencies 0.5, incomplete dominance
9
Loss of heterozygosity in a laboratory population

• Dashed line is the theoretical prediction for
  population size N 16
• Solid gray line is theoretical prediction for N
  9
• The effective population size was smaller than
  the actual population size
• See previous slide for additional experimental
  details

Heterozygosity
Generation
10
Population size and genetic diversity in natural
populations (Fig. 6.19a,b)
Polymorphism proportion of loci for which the
frequency of the most common allele is lt
0.99 Allelic richness average number of
alleles per locus
11
Genetic variation in Ozark glade populations of
collared lizard (Fig.

• The colors in the pie chart at left represent 7
  different multilocus genotypes based on malate
  dehydrogenase (MDH) genotypes, mtDNA haplotypes,
  and ribosomal DNA (rDNA) genotypes.
• Populations in different glades are represented
  by colored circles on the maps
• A solid-color circle indicates that only one
  multilocus genotype is present in that population.

12
Founder effects and bottlenecks are special cases
of genetic drift

• Founder effects occur when a new population is
  started with a small number of migrants from
  another (usually larger) population
• Bottlenecks occur when a large population is
  reduced to a small size
• In both cases drift is strong because of small
  population size, which may persist for a number
  of generations

13
How founder effects and bottlenecks change the
genetic composition of populations

• Reduce genetic variation through loss of low
  frequency alleles
• Increase the frequency of some very rare
  (probably harmful) alleles that happen to be
  present in founders or survivors of bottlenecks
• Lead to inbreeding, which increases the chance
  that recessive mutations will be homozygous

14
The founder effect in an island-hopping birdthe
silvereye, Zosterops lateralis (Fig. 6.13)

• Progressive founder events reduce genetic
  diversity as measured by the average number of
  alleles at six microsatellite loci

15
A founder effect in a human population
achromatopsia in the Pingelapese people of the
Eastern Caroline Islands

• The current population of about 3,000 people on
  the Pingelap Atoll are descended from 20
  survivors of a typhoon in about 1775
• The frequency of achromatopsia (complete
  colorblindness, extreme sensitivity to light, and
  poor visual accuity), a homozygous recessive
  disorder, is about 1 in 20 (compared to less than
  1 in 20,000 in most populations)
• At least one of the survivors of the typhoon
  carried at least one copy of the defective
  allele, which means that among the survivors the
  frequency of the defective allele was at least
  2.5
• The frequency of the defective allele has since
  increased to gt 20 under the influence of drift

16
Migration and genetic drift

• Migration of individuals (movement from one
  population to another) retards the rate of drift
  within each population
• If migration among populations is high enough,
  the several populations become, in effect, one
  larger population in which drift is slower
• In general, 2 or more migrants into a population
  per generation is enough gene flow to make
  several populations behave like one larger
  population

17
Selection and genetic drift 1

• Genetic drift also affects the frequencies of
  alleles that are under selection
• Suppose the fitnesses (relative survivorships) of
  genotypes at a locus are as follows
• According to what we have learned, if A2 is a new
  mutation, directional selection should result in
  the fixation of allele A2 because A2A2
  homozygotes have the highest survivorship that
  is, the Pr(fix A2) 1.0
• However in a finite population of size N, its
  not so simple
• If the quantity 4Ns is greater than about 2, then
  Pr(fix A2) 2s, which will be much less than 1.0
  for realistic selection coefficients
• If the quantity 4Ns is less than about 1, then
  Pr(fix A2) 1/2N, which is the probability of
  fixation of a new neutral mutation by drift, and
  will also be less than 1.0
• Either way, most beneficial mutations are lost
  from populations because of random genetic drift!
  And its even worse if A2 is recessive!

18
Selection and genetic drift 2

• Lets look at some numbers
• Suppose N 200, and s 0.01 (i.e., a 1
  selective advantage to A2 when heterozygous)
• Then 4Ns 8, which is large, and the
  probability of eventual fixation of A2 if it is a
  new mutation 2s 0.02
• On the other hand
• Suppose N 5, and s 0.01
• Then 4Ns 0.2, which is small, and the
  probability of eventual fixation of A2 is 1/2N
  0.1
• In this particular example, then, our new
  favorable mutation, A2, is more likely to become
  fixed in a small population, where its initial
  frequency is higher, than in a large population,
  where its initial frequency is lower.

19
The Neutral Theory of Molecular Evolution 1

• It is important to distinguish between evolution
  of proteins (i.e., amino acid sequences) and
  evolution of nucleotide sequences
• It is reasonable to assume that much nucleotide
  sequence evolution is selectively neutral or
  nearly so pseudogenes, introns and other
  non-translated regions synonymous substitutions
• On the other hand, we might expect many amino
  acid substitutions to be under selection
• The long-standing argument between neutralists
  and selectionists is really about whether most
  protein evolution ( amino acid substitutions
  non-synonymous nucleotide substitutions) is
  neutral, or is adaptive and driven by natural
  selection

20
The Neutral Theory of Molecular Evolution 2

• Neutralists do not claim that most mutations are
  neutral
• However, they do argue that most of the protein
  variation that we see within populations has no
  fitness consequences (i.e. alleles are neutral),
  and that most of the evolutionary change in
  proteins that we see between related taxa is due
  to drift acting on selectively neutral alleles.
  According to neutralists, positive selection (
  adaptation) plays little or no role in evolution
  at the molecular level
• For neutralists, as well as selectionists, most
  mutations are harmful and are removed from
  populations by the action of natural selection

21
Neutral theory predicts a molecular clock

• Let the neutral mutation rate be µ( proportion
  of new mutant copies of a gene per generation,
  e.g., 1 in a million)
• In a diploid population of size N, there will be
  2Nµ new mutations at a gene per generation
• Since these mutations are neutral, the
  probability of eventual fixation of any one
  mutation is 1/2N, and its probability of loss is
  1 - (1/2N) which will be very close to 1.0 for
  reasonably large populations
• Most new neutral mutations will be lost by drift
  within a few generations, but occasionally a new
  mutation will increase infrequency under drift
  and replace previously existing alleles this is
  known as an allelic substitution

22
Mutation and allelic substitution over time (Fig.
6.20)
23
The average rate of allelic substitution, which
is the rate of neutral evolution, is equal to the
neutral mutation rate

• The average number of allelic fixations per
  generation is equal to the number of new
  mutations per generation x the probability that
  any one mutation eventually becomes fixed 2Nµ x
  1/2N µ
• So, the predicted average amount of time between
  allelic substitutions is 1/µ generations this is
  the molecular clock

24
Molecular divergence among three related taxa
A
B
C
t
Time since most recent common ancestor
2t

• Neutral theory predicts
• The amount of molecular difference between A and
  C will be the same as the amount of difference
  between B and C, since the amount of evolutionary
  time is the same in both comparisons
• The amount of molecular difference between A and
  C (or B and C) will be twice the amount of
  difference between A and B, since the common
  ancestor of A and B lived half as long ago as the
  common ancestor of A and C (or B and C)

25
Synonymous and non-synonymous base substitutions
(Fig. 6.21)
26
Molecular evolution in influenza viruses is
consistent with neural theory (Fig. 6.21c)
27
A paradox and an inconsistency

• The paradox is that under neutral theory, in
  which allelic substitutions are due only to drift
  (rather than directional selection) the rate of
  evolution does not depend upon population size
  but only the mutation rate to neutral alleles
• The inconsistency is that the theory predicts a
  clock that ticks in generation time, yet much
  of the evidence for a molecular clock is based on
  clock time

28
Nearly neutral theory 1

• Neutralists have addressed the problem of time
  scale with what is known as nearly neutral theory
• The idea is that many slightly deleterious
  mutations will be effectively neutral, depending
  on population size
• In general, as we saw above, drift will govern
  the fate of alleles if the quantity 4Ns is
  small, where N is effective population size and
  s is the coefficient that describes the action of
  natural selection on the heterozygote

29
Nearly neutral theory 2

• Imagine a species in which effective population
  size, N, is 500. If the selection coefficient,
  s, against a mutant heterozygote is 0.0005, then
  4Ns 1.0, which qualifies as small, and the
  mutation is effectively neutral
• On the other hand, the same selection coefficient
  in a population of 5,000 would mean that 4Ns
  10.0, which would be large and would mean that
  natual selection would act against the mutation
  (that is, the mutation is harmful not neutral)
• If species with long generation times tend to
  have small population sizes, then the mutation
  rate to effectively neutral mutations will be
  greater in species with longer generations, and
  this higher mutation rate per generation will
  compensate for the longer generation time
• The net result is that the molecular clock will
  tend to tick at the same speed in clock time for
  all species

30
Population size, generation time, and nearly
neutral mutations (Fig. 22)
31
Synonymous and non-synonymous substitutions and
the molecular clock 1

• Evolution of synonymous nucleotide substitutions
  generally supports a molecular clock (this is not
  controversial)
• In some cases, non-synonymous substitutions also
  appear to be clock-like (e.g., the influenza
  virus data in Fig. 6.21)

32
Synonymous and non-synonymous substitutions and
the molecular clock 2

• Data from many genes indicates that the rate of
  evolution of synonymous substitutions is higher
  (often much higher) than the rate of evolution of
  non-synonymous substitutions (see Table 6.1 in
  your text)
• This observation is consistent with the
  expectation that most amino acid changes will be
  deleterious and, therefore, that the neutral
  mutation rate will be lower for non-synonymous
  substitutions (as would be predicted by both
  neutralists and selectionists)

33
Molecular evolution in influenza viruses is
consistent with neural theory (Fig. 6.21c)
34
Evidence for molecular evolution by positive
selection

• Neutral theory states that if we compare protein
  coding sequences (exons) between species, there
  will be more nucleotide substitutions at
  synonymous than non-synonymous sites (because the
  neutral mutation rate is lower at non-synonymous
  sites) dN/dS lt 1.0
• On the other hand, if much evolution involves
  positive selection on alleles that increase
  fitness, then evolution at non-synonymous sites
  may be faster than at synonymous sites (because
  positive selection will substitute alleles faster
  than drift) dN/dS gt 1.0

35
Positive selection on the BRCA1 gene in humans
and chimpanzees (Fig. 6.23)Ratio of replacement
to silent mutations that is greater than 1.0 on
branches leading to humans and chimps is evidence
for positive selection
36
Molecular evolution summary

• Neutral theory appears describe much evolution at
  the nucleotide level in untranslated parts of the
  genome and at synonymous sites
• However, for many, but not all proteins, there is
  evidence that evolution has been driven by
  positive selection on non-synonymous
  substitutions
• recently duplicated genes that have attained new
  functions, disease-resistance loci (e.g., genes
  that code for antibody proteins), genes involved
  in interactions between egg and sperm at
  fertilization, genes that code for certain
  enzymes (e.g., alcohol dehydrogenase)

37
Nonrandom mating

• Most commonly, nonrandom mating takes the form of
  inbreeding, or mating with relatives
• Inbreeding can be systematic, such as by
  self-fertilization in hermaphroditic plants and
  animals (this is the most extreme form of
  inbreeding)
• Inbreeding can be accidental, as when mating
  occurs between related individuals in finite,
  especially small, populations
• In either case, the genetic effect of inbreeding
  is to reduce heterozygosity below Hardy-Weinberg
  expectation (2pq), and to increase homozygosity
• One way of defining the inbreeding coefficient
• F (HH-W - Hobserved) / HH-W

38
Inbreeding increases homozygosity the example
of selfing (Fig. 6.25a)
39
Inbreeding increases homozygosity the example
of selfing (Fig. 6.25b)
40
The effects of inbreeding

• Because inbreeding increases homozygosity, it
  makes it more likely that deleterious recessive
  alleles in a population will be expressed
• This increases the effectiveness of selection
  against harmful recessives
• However, in species that do not normally inbreed,
  it also leads to inbreeding depression

41
Inbreeding depression in humans (Fig. 6.28)

• Children of first cousins have higher mortality
  rates than children of unrelated parents (by
  about 4 percentage points)

42
Inbreeding depression in great tits (Fig. 6.30)

• F 1/4 is the inbreeding coefficient of children
  of brother-sister matings
• F 1/16 is the inbreeding coefficient of children
  of 1st cousins

43
Synergism between small population size, drift
and inbreeding 1

• Drift is especially strong in small populations
• Drift reduces genetic variation ( increased
  homozygosity)
• It is possible for deleterious alleles to
  increase in frequency under drift (think about
  bottlenecks and achromatopsia in Pingelapese
  islanders)
• In small populations, inbreeding (mating between
  relatives) is unavoidable (remember the
  Pingelapese who are all descended from 20
  survivors of a typhoon) This is true even if the
  population is mating at random

44
Synergism between small population size, drift
and inbreeding 2

• So, small population size results in loss of
  genetic variation, chance increases in the
  frequency of harmful alleles, and inbreeding
• Inbreeding results in inbreeding depression,
  which further reduces population size, which
  further enhances the effect of drift, and also
  results in more inbreeding, which results in more
  inbreeding depression, which results in further
  reductions in population size, etc., etc., etc.
• This has been referred to as the extinction
  vortex