Topic 10: Population Bottlenecks Flashcards
Does it take longer for genetic variation or population size to recover from a bottleneck?
Genetic variation
What are 5 causes of bottlenecks?
Habitat loss/fragmentation
Exploitation
Disease
A founding event
Other stochastic human induced or natural events
What are 4 things that bottlenecks can lead to?
Reduction in effective size
Loss of genetic variation
Limits to the population’s ability to adapt
Inbreeding
What is Ne?
Effective population size, the number of individuals that effectively participate in producing the next generation
All individuals reproduce equally, equal numbers of males and females. generations are discrete, no inbreeding, population size is constant and stable
Determines the strength of drift, can be estimated from detailed demographic data
Ne can also be measured using genetic data independently
What is Nc?
Consensus population size, easier to measure than Ne but rarely does Ne=Nc, and often times Ne is only 10-50% of Nc
What happens to genetic variation over time in an isolated population?
Genetic variation declines in small isolated populations due to inbreeding-like effects of drift (ignoring mutation and selection) each generation
Inbreeding accumulates over generations, and there are three terms in the equation to account for the probability of drawing the same allele twice by chance (identical by descent), probability of drawing two different alleles, and probability two different alleles are identical by descent in the previous generation
What type of genetic data are used to detect population bottlenecks?
Microsats offer a way of detecting whether past bottlenecks have occurred and allow us to measure the severity of a bottleneck.
Most methods look at the changes that occur in heterozygosity and allelic diversity and infer changes in effective population size from this.
What are the two genetic methods we can use to study changes in population size? (detect bottlenecks)
Two sample methods:
- compare levels of of polymorphism (obs/ex heterozygosity, polymorphic loci, # alleles per locus) which will be lower in bottlenecks
-compare allele frequency distribution (common vs rare)
compare variance in allele frequency (changes a lot in small populations)
One sample methods:
- allele size distribution
M-ratio
What happens in a two sample method to test bottlenecks?
You sample the population before the bottleneck and then currently (after the bottleneck)
This requires that the bottleneck was anticipated and observed, historic or museum specimens that are available pre-bottleneck, a current population exists that resembles what your bottleneck population would have been like in the past (pre-bottleneck)
What is involved in the two sample method of changes in polymorphism?
Genetic variation is expected to decline when the effective population size declines
Compare Hex, Hobs, proportion of polymorphic loci and/or # of alleles per locus (k) using a sign test or other non-parametric stats tests to see if they are correlated
What is involved in the two sample method of changes in allele frequency distribution?
During a bottleneck, rare alleles are more likely to be lost than common alleles due to drift. This changes the distribution of allele frequencies in a population by reducing the proportion of rare alleles.
Use samples pre and post bottleneck to compare the allele frequency distributions, and use a chi square to compare them and see if they are significant.
Rare alleles are LOST much faster
Use a chi square test for significance of the change in allele distribution from source to founded
Can use the change in allele frequencies to estimate the effective size during the bottleneck
What is involved in the one sample method of using range in allele size?
During a bottleneck, rare alleles are lost which creates a gap in the microsatellite allele frequency distributions
Because the alleles lost are not necessarily from the extreme ends of the size distribution, the range of allele size (r) will not be as affected as much as the number of alleles observed(k)
Therefore we expect the M=k/r ratio to decline following a bottleneck
How do you test for a meaningful reduction in M-ratios?
Using a critical value of 0.68 derived from putatively stable wild populations in Garza and Williamson
Comparing the mean observed M-ratio to an expected distribution determined by simulations
What are two and one sample tests used for?
Two sample tests: changes in number of alleles, changes in expected heterozygosity, changes in allele frequency distribution, estimation of Ne using allele frequency variance
One sample tests: loss of alleles vs reduction in allele size range
