Topic 9 Flashcards
(32 cards)
1
Q
metapopulation
A
A ‘large’ population composed of two or more partially (or completely) isolated demes, such that mating within the metapopulation is not random
2
Q
subpopulation or deme
A
groups of individuals that mate randomly with each other in the absence of inbreeding and assortative mating
3
Q
population subdivision
A
- Species almost never exist as a single panmictic population, but rather as subpopulations or demes that are typically spatially isolated
4
Q
wahlund effect
A
- the perception of a heterozygote deficit caused by treating two different subpopulations as a single population.
5
Q
multiple populations of the same species
A
- We have just seen that population subdivision reduces heterozygosity and therefore increases F, the inbreeding coefficient when we treat more than one population as a single population.
- Note that in the previous example, F did not increase because of ‘inbreeding’ within the two subpopulations, they are both randomly mating.
- The nonzero (+ve) value of F resulted from non-random mating and HW deviations at the metapopulation level.
6
Q
population subdivision increases F in the metapopulation
A
- Although F increases in the metapopulation because of population subdivision, the value of F in the metapopulation will also reflect reductions in heterozygosity as a result of any nonrandom mating within the individual demes or subpopulations.
- We need to develop a method to separate these contributing factors.
7
Q
Fis
A
- FIS is an average measure of F among subpopulations, or the average fractional reduction (or increase) in heterozygosity
(HW deviations) within subpopulations dues to all factors (inbreeding, outbreeding, assortative mating) and ranges from -1 to 1
8
Q
Fst
A
- FST is a measure of the fractional reduction in heterozygosity within the metapopulation from what is expected based on the average allele frequencies, which occurs because of population subdivision alone
- ranges from 0 to 1
- FST gives us the fractional reduction in heterozygosity in the metapopulation
compared to that expected in the metapopulation due to differences in allele
frequencies among subpopulations. FST will be 0 when all subpopulations have identical allele frequencies, and 1 if they are all fixed for different alleles - FST measures the fractional reduction in heterozygosity within the metapopulation because of allele frequency differences within
demes, in comparison to what we would expect if all demes were fused into a single panmictic population with allele frequencies p and q. In
other words, the heterozygosity is reduced in comparison to 2pq. In the absence of migration, selection, or mutation, FST can be interpreted to arise because of genetic drift. However, any factor that changes allele frequencies among demes will change FST
9
Q
interpreting Fst
A
0- 0.05: little genetic differentiation
0.05- 0.15: moderate genetic differentiation
0.15- 0.25: great genetic differentiation
> 0.25: very great genetic differentiation
10
Q
FIT
A
- FIT is the TOTAL departure from HW expectations (differences in heterozygosity) in the metapopulation because of deviations
within demes caused by localized inbreeding etc. (FIS) , as well as allele frequency differences among the demes or subpopulations (FST) - FST can also be calculated and interpreted in terms of allele frequency variances. In this case, the interpretation is that FST is the proportion of the total genetic variation in the metapopulation that results from differences in allele frequencies among the subpopulations
11
Q
nei’s genetic difference
A
- Over the years, numerous methods have been developed to quantify allelic differences between two populations. These differ in their assumptions, depending on the type of allele frequency data being used. For example, some are only applicable to microsatellite data.
- Although they give different values of genetic distance, they are all correlated with each other. A historically widely used distance measures is Nei’s genetic distance. Masatoshi Nei has made many important contributions to evolutionary genetics
- Nei’s genetic distance can theoretically take on values between 0 and infinity. In the case of allelic data, by way of some arithmetic, it can be interpreted as the number of allelic substitutions at a locus. In our previous example, we did not have any fixed allelic differences, and the allele frequencies were not hugely different, hence a D value of 0.016 substitutions between populations at the locus
- The key is that the larger the value of Nei’s D, the greater the difference in allele frequencies among populations or subpopulations
12
Q
overview of FIT
A
- Quantifies the total decrease in heterozygosity (HW deviation) at the metapopulation level because of any deviations within each subpopulation due to all factors such as inbreeding, outbreeding and assortative mating, as well as due to subdivision within the metapopulation
13
Q
overview of Fis
A
- Quantifies the average reduction (or increase) in heterozygosity within subpopulations because of factors such as assortative mating, inbreeding, and outbreeding (note that outbreeding and negative assortative mating will increase heterozygosity).
- These reductions (or increases) within subpopulations contribute to the total deviation from the expected heterozygosity (2pq) at the metapopulation level (FIT).
14
Q
overview of Fst
A
- Quantifies the proportion of the total reduction in expected heterozygosity (2pq) at the metapopulation level which occurs because of population subdivision (The Wahlund Effect).
- It arises because of allele frequency differences within the subpopulations and provides a measure of genetic differentiation among the subpopulations
- FST arises because of allele frequency differences among subpopulations. Since allele frequency differences among subpopulations are directly related to the variance of the mean allele frequencies in the metapopulation, the variance of the mean allele frequencies is directly related to FST. The greater the allele frequency differences among the subpopulations, the greater the variance of the mean allele frequencies, and the greater the value of FST.
- Quantifies the proportion of the total genetic variation in the metapopulation that arises because of allele frequency differences among subpopulations
15
Q
Nst
A
- a FST variant for DNA sequence data
- NST (and other variants such as KST) are useful for examining population subdivision using DNA sequence data. Often in the literature, NST is erroneously called FST
- tells us the proportion of the total nucleotide diversity in the metapopulation that arises as a
consequence of population subdivision, or differences among subpopulations
16
Q
Gst
A
- a FST variant for haplotype or allelic data
- is useful for examining population subdivision using DNA sequence data. Often in the literature, GST is also erroneously called FST
- is the proportion of the total haplotype diversity in the metapopulation that arises because of subdivision, or haplotype frequency differences among subpopulations
17
Q
Gst and Nst
A
- GST: considers only haplotype frequencies (differences among populations) to quantify population differences
- NST: considers both haplotype frequencies, and the sequence divergence between those haplotypes to quantify population differences
18
Q
testing for significance of population subdivision
A
- We can use a contingency goodness of fit test to examine the statistical significance of population subdivision.
- Contingency tests are a special variant of the X2 test. With this variant, we test for differences in the observed number of alleles among populations, which can easily be calculated from genotypes.
- The contingency test is very useful because it can be applied to any type of data: allozymes, microsatellites, SNPs, as well as nuclear mitochondrial and chloroplast genes.
- Computer programs use sophisticated permutation tests to look at subdivision. The increased power of computers over the past three decades has allowed for ever more complex analyses.
19
Q
gene flow
A
- Gene flow refers to the movement of individuals from one population to another, and subsequent mating.
- The general impact of gene flow is that it reduces the level of population differentiation, as measured by FST or any other measure of population differentiation.
20
Q
the continent island model
A
- The Continent- Island Model of gene flow assumes one way migration from a large continental population to a small island population, but it can be adapted to other situations in which gene flow is unidirectional
21
Q
the island model
A
- The island model of migration describes the case where a large number islands randomly exchange individuals.
- The islands can represent actual islands, lakes, fragments of habitats, mountains separated by valleys etc. We assume that the population size on the islands is infinitely large
- Any population exchanges migrants with any other population at equal probabilities
22
Q
the stepping stone model
A
- The island model of gene flow is often unrealistic because populations are much more likely to exchange migrants with adjacent populations than more distant populations.
- The stepping stone model accounts for this
- Adjacent lakes are much more likely to exchange migrants than lakes that are not adjacent. In this case, Lakes A and E are the least likely to exchange migrants. This results in an
isolation by distance pattern. In an Isolation by distance pattern, geographic distance is correlated to genetic distance.
23
Q
model assumptions
A
- The Continent-Island, Island and Steppingstone models all have the general effect of homogenizing populations, though in the latter case, an equilibrium value for the frequency of an allele is difficult to arrive at mathematically.
- All of these models assume no mutation, selection and that populations are infinitely large (no genetic drift)
24
Q
migration rates and dispersal abilities
A
- The rate of homogenization and ∆p will depend on the migration rate, or fraction of alleles that are migrants (m).
- This will in turn be dependent on the dispersal capabilities of the organism in question. In general, organisms with stronger dispersal abilities have higher migration rates.
- EG Starlings have conquered North America
25
migration and the HW principle
- The movement of migrants into a population can alter allele frequencies, and initially cause HW disequilibrium, but one round of random mating will restore HW equilibrium with respect to the new allele frequencies.
26
migration and gametic disequilibrium
- Consider a population where all members possess the multilocus genotype AAbb. A group of individuals with the multilocus genotype aaBB then migrate into this population.
- This will immediately result in non-random genotypic associations among the two loci, and in this case the disequilibrium coefficient D will be equal to Dmin (notice that AAbb and aaBB individuals can only produce repulsion phase gametes).
- In the absence of additional migration, D will break down over subsequent generations at a rate defined by the recombination fraction r.
27
migration and genetic drift
- Genetic drift causes populations to diverge, and gene flow acts to homogenize populations. Hence, drift and migration oppose each other.
- How much gene flow is required to oppose genetic drift?
28
the infinite island model
- Reconsider the Island Model, but instead of an infinitely large effective population size, we have finite population sizes, but the number of islands is infinitely large.
- If gene flow is low, drift will be the dominant
force on each island and they will diverge from each other
- If gene flow is high, migration will be the dominant force. Given that we are assuming the same effective population size on each island, the effective number of immigrants will determine the fate of the system.
- The effective number of immigrants per generation on all islands is (Ne)(m), which is the product of the effective population size on each island, and the migration rate.
- This figure (topic 9 II slide 30) shows the distribution of allele frequencies in populations (or subpopulations) within a metapopulation
for different values of (Ne)(m) when the average allele frequency of p or p is equal to 0.5
- When Nem = 0.05 the distribution is U shaped
and drift fixes most subpops for either p or q
- When Nem = 0.5 the distribution is flat
- With larger values of Nem (5 or more) the distribution becomes clustered around 0.5, which is equal to p
- This shows that the variance of the mean allele frequencies in the metapopulation depends on the magnitude of Nem
- The relationships on the previous slide show that population divergence due to drift depends on Nem, or the number of migrants, this is irrespective of the population sizes.
29
why the infinite islands model
- Assume that Nem is 5, and that the effective population size is 20. In this case, drift will be very strong because population size is small. However, the effect of migration will also be very strong because a Nem of 5 represents a big fraction of the population.
- Now, Assume that Nem is 5 and that the effective population size is 1,000,000. Drift will be very weak because the population size is so large, but the effect of migration will also be weak because a Nem of five represents a small fraction of the population
30
Infinite islands model Fst
If the variance of the mean allele frequencies depends on Nem,
then FST must also depend on Nem. With a given amount of
gene flow, FST will reach an equilibrium value:
31
measuring gene flow
- The previous formula suggests that it is possible to quantify the number of migrants using FST, and many researchers have attempted to do so.
- However, we must keep the assumptions of this approach in mind: populations have the same effective population size, populations are at equilibrium, population structure is approximated by an island model with infinite islands, there is no selection or mutation, gene flow occurs randomly regardless of proximity, gene flow occurs randomly with respect to genotype
- These assumptions are highly unlikely to be met, but nevertheless FST is correlated with levels of gene flow.
32
Summary
- Migration (gene flow) is one of the four principal evolutionary forces (along with mutation, selection and drift)
Its overall effect is two homogenize populations with respect to allele frequencies, and reduce the value of FST in a metapopulation
Migration will act in opposition to genetic drift, and can either work with, or oppose selection
On its own, it can create both HW and gametic disequilibrium
