Finite population size

Sewall Wright (1889 – 1988)

Two important effects of finite population size

Genetic Drift

Inbreeding

Progeny from mating between two heterozygotes

Aa x Aa

Possible genotypes of two offspring

Probability

Allele frequencies

 

 

 

A

a

AA and AA

(1/4)(1/4) =

1/16

1.00

0.00

aa and aa

(1/4)(1/4) =

1/16

0.00

1.00

AA and aa

2(1/4)(1/4) =

2/16

0.50

0.50

Aa and Aa

(2/4)(2/4) =

4/16

0.50

0.50

AA and Aa

2(1/4)(2/4) =

4/16

0.75

0.25

aa and Aa

2(1/4)(2/4) =

4/16

0.25

0.75

 

 

When is drift important?

Founder event

Bottlenecking

Captive population

Founder event example

Rubus alceifolius (Asian bramble)

Native to mainland East Asia (e.g. Viet Nam)

Introduced to Oceanic Islands (e.g. Réunion)

Study by Amsellem L., et al. 2000. Mol. Ecol. 9: 443-455

Réunion             Viet Nam

Bottleneck example

Northern Elephant Seal (Mirounga angustirostis)

Hunted almost to extinction

By 1890, fewer than 20 animals

Population now numbers more than 150,000


Captive population (hypothetical example)

Concepts

Identity by descent

Coalescence time


Problem

The western tussuck moth (Orgyia vetusta) is an annual moth that feeds on the yellow bush lupine, Lupinus arboreus.  The population overwinters as eggs, which hatch in the early spring.  Larvae feed for about a month and then metamorphose into adults, which quickly mate.  The females oviposit (lay eggs) on the lupine bushes and the adults subsequently die, so that there is only one generation per year.  Larvae feed voraciously, and when populations are dense, they defoliate and kill entire patches of lupine.

Suppose you are studying a population of moths at Bodega Marine Reserve.  During your study, the Fast and Slow allozymes at the pgi locus have been at Hardy-Weinberg Equilibrium with p = freq(Fast) = q = freq(Slow).

After the El Niño rains of 1998, the moth population exploded and defoliated all the lupines on the Reserve.  When you returned in 1999, despite extensive searching with multiple black light stations over several weeks, you find only one moth.  You have reason to believe that there was no selection on this locus during this extreme population bottleneck (i.e. mortality was random with respect to genotype at this locus). 

 

1.      What is the allele frequency in the 1999 generation?

2.      What is the probability that the population lost a pgi allele in this bottleneck?

3.      If the population had declined to only three individuals (N = 3), what would the allele frequency be?

4.      In that case, what would be the probability that the population had lost a pgi allele in the bottleneck?

5.      Can you come up with a general equation that predicts the probability that a large population will lose one of the two alleles at a locus (become monomorphic) when it drops to any small size N?


Homework

Go to http://www.utm.edu/~rirwin/Drift.htm

Use this genetic drift model to observe outcome of drift

1.      Choose a pair of allele frequencies (p = freq(A), q = freq(a)) to examine (remember, p + q = 1)

2.      Key in this frequency

3.      Set population size to 100 individuals

4.      Set run time to 200 generations

5.      Run model

6.      Record final allele frequencies

7.      Run model with same allele frequencies for 10 runs

8.      Imagine that you are observing evolution by drift in 10 replicate small populations.  Use your results to test whether the mean allele frequencies averaged over the 10 populations equal p and q.

9.      Among the runs in which one allele fixes, calculate probability with which each allele fixes (e.g. ).