Human Molecular Genetics IV. Genetics of common diseases/ Multifactorial genetics.
Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 ·...
Transcript of Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 ·...
![Page 1: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/1.jpg)
Population Genetics, Lecture 1
Nancy Lim Saccone
Bio 5488, Spring 2019
Monday 3/18/19
(with thanks to Don Conrad and slides from past years)
![Page 2: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/2.jpg)
What is population genetics?
• Very broadly: The science of genetic variation in
populations of organisms
• Origin, amount, frequency, distribution of this
variation in space and time
• Focus on human population genetics
c.f. Biol 4181, Alan Templeton
![Page 3: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/3.jpg)
Why study (human) population genetics?
• Demographic inference
• Our DNA records history of humankind:
population size changes, migrations, etc.
• Functional inference
• E.g. alleles deleterious to "fitness" are unlikely to
be common
• Complex disease
• What analysis approaches are appropriate, or
can leverage population history
• Especially now in the era of sequencing
• You will have your own genome sequence
![Page 4: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/4.jpg)
![Page 5: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/5.jpg)
Outline for today
Theory:
- Hardy-Weinberg
- Forward models: Wright Fisher model
- Decay of heterozygosity
- Backward models: Coalescent
![Page 6: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/6.jpg)
Hardy-Weinberg Law
Goal: describe the relationship of allele and genotype
frequencies in a population that follows Mendel's law
of segregation.
Under certain assumptions, an equilibrium is
maintained
![Page 7: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/7.jpg)
Hardy-Weinberg Law
Some preliminaries:
Locus with 2 alleles: A1 , A2
p = pr(A1)
q = pr(A2)
p + q = 1
If we know the genotype frequencies in a population
sample - pr(A1A1), pr(A1A2), pr(A2A2) - we can
calculate the allele frequencies:
p = pr(A1) = 1 * pr(A1A1) + 0.5 * pr(A1A2) + 0 * pr(A2A2)
q = 1-p = 0 * pr(A1A1) + 0.5 * pr(A1A2) + 1 * pr(A2A2)
What about the reverse?
![Page 8: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/8.jpg)
Hardy-Weinberg Law
What about the reverse: If we know the allele
frequencies, can we calculate the genotype
frequencies?
In general, allele frequencies do not uniquely determine
genotype frequencies:
p = 0.5, q = 0.5 can correspond to:
pr(A1A1) = 0.25 pr(A1A1) = 0.5
pr(A1A2) = 0.5 OR pr(A1A2) = 0.0 OR…
pr(A2A2) = 0.25 pr(A2A2) = 0.5
![Page 9: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/9.jpg)
Hardy-Weinberg equilibrium (2 allele locus)
Under certain assumptions, Hardy-Weinberg equilibrium holds:
that is, the probabilities of the three genotypes (A1A1, A1A2,
A2A2) are p2, 2pq, and q2, respectively.
HW law (under the assumptions below):
1) HWE is established after 1 generation of random mating.
2) HWE, once established, is maintained over the generations
Assumptions:
a. (infinitely) large population
b. discrete generations
c. random mating
d. no selection, no migration, no mutation
e. equal initial genotype frequencies in the two sexes
Note also we take genotypes to be unordered.
![Page 10: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/10.jpg)
Hardy-Weinberg equilibrium
To see 2): Suppose HWE holds. Then
Maternal
A1 (p) A2 (q)
Paternal A1 (p) A1A1 (p2) A1A2 (pq)
A2 (q) A2A1 (qp) A2A2 (q2)
Why is p the prob of a parent transmitting A1 ?
Parental Prob of having Prob of Joint
genotype this genotype transmitting ‘A1' probability
A1A1 p2 1 p2
A1A2 2pq 1/2 pq
A2A2 q2 0 0
Total: p2+pq = p(p+q)=p
Because HWE holds in current pop
![Page 11: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/11.jpg)
Hardy-Weinberg Equilibrium
To see 1) HWE is established after 1 generation of random
mating:
Notation:
Let p = Pr(A1) = allele freq of A1 in parental generation,
q = Pr(A2) = allele freq of A2 in parental generation
Pr(AiAj) = frequency of genotype AiAj in initial (parental)
generation
(we don't assume these equal p2, 2pq, q2 for A1A1, A1A2,A2A2)
Pr1 is notation for probability (frequency) in the next (offspring)
generation
![Page 12: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/12.jpg)
To see 1):
Given genotype frequencies in a (parental) population, Mendelian
principles of segregation dictate the probability distribution for
the genotypes in the offspring population
Father Mother Prob that
Offspr is
A1A1
Prob that
Offspr is
A1A2
Prob that
Offspr is
A2A2
A1A1 A1A1 1 0 0
A1A1 A1A2 0.5 0.5 0
A1A1 A2A2 0 1 0
A1A2 A1A1 0.5 0.5 0
A1A2 A1A2 0.25 0.5 0.25
A1A2 A2A2 0 0.5 0.5
A2A2 A1A1 0 1 0
A2A2 A1A2 0 0.5 0.5
A2A2 A2A2 0 0 1
![Page 13: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/13.jpg)
To see 1):
In offspring, genotype probabilities are:
Pr1(A1A1) = 1*Pr(A1A1)Pr(A1A1) + 0.5Pr(A1A1)Pr(A1A2) + 0.5
Pr(A1A2)Pr(A1A1) + 0.25 Pr(A1A2)Pr(A1A2)
= [Pr(A1A1) + 0.5Pr(A1A2)]2
Pr1(A2A2) = ... = [Pr(A2A2) + 0.5Pr(A1A2)]2
Pr1(A1A2) = ... = 2[Pr(A1A1) + 0.5Pr(A1A2)][Pr(A2A2) + 0.5Pr(A1A2)]
Father Mother Prob that
Offspr is A1A1
Prob that
Offspr is A1A2
Prob that
Offspr is A2A2
Freq of this mating
type
A1A1 A1A1 1 0 0 Pr(A1A1)Pr(A1A1)
A1A1 A1A2 0.5 0.5 0 Pr(A1A1)Pr(A1A2)
A1A1 A2A2 0 1 0 Pr(A1A1)Pr(A2A2)
A1A2 A1A1 0.5 0.5 0 Pr(A1A2)Pr(A1A1)
A1A2 A1A2 0.25 0.5 0.25 Pr(A1A2)Pr(A1A2)
A1A2 A2A2 0 0.5 0.5 Pr(A1A2)Pr(A2A2)
A2A2 A1A1 0 1 0 Pr(A2A2)Pr(A1A1)
A2A2 A1A2 0 0.5 0.5 Pr(A2A2)Pr(A1A2)
A2A2 A2A2 0 0 1 Pr(A2A2)Pr(A2A2)
![Page 14: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/14.jpg)
HWE. To see 1):
In offspring,
Pr1(A1A1) = [Pr(A1A1) + 0.5Pr(A1A2)]2
Pr1(A2A2) = ... = [Pr(A2A2) + 0.5Pr(A1A2)]2
Pr1(A1A2) = ... = 2[Pr(A1A1) + 0.5Pr(A1A2)][Pr(A2A2) + 0.5Pr(A1A2)]
Note also that in the parents' generation,
p=Pr(A1) = Pr(A1A1) + 0.5Pr(A1A2)
q=Pr(A2) = Pr(A2A2) + 0.5Pr(A1A2)
Thus,
Pr1(A1A1) = [Pr(A1)]2 = p2
Pr1(A2A2) = [Pr(A2)]2 = q2
Pr1(A1A2) = 2Pr(A1)Pr(A2) = 2pq
And, again, in the offspring generation
p1=Pr1(A1) = Pr1(A1A1) + 0.5Pr1(A1A2)= p2 + 0.5(2pq)=p(p+q)=p
q1=Pr1(A2) = Pr1(A2A2) + 0.5Pr1(A1A2)=q
So the allele frequencies are unchanged in the next generation;
thus Pr1(A1A1) = p12, Pr1(A2A2) = q1
2, Pr1(A1A2) = 2p1q1.
![Page 15: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/15.jpg)
gcbias.org
![Page 16: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/16.jpg)
Testing for HWE
ex/ consider locus with alleles a,A. Suppose in a sample of 550
individuals, we observe 200 aa, 300 aA, 50 AA. Is this consistent
with HWE?
Answer: use chi-squared test ,1 df:
N = total number of alleles observed = 2*550 = 1100
Pr(a) = (200*2 + 300)/1100 =7/11
Pr(A) = (300 + 50*2)/1100 = 4/11
Expected genotype counts Observed genotype
under HWE: counts
aa: (49/121)*550 = 222.727 200
aA: 2*(28/121)*550 = 254.545 300
AA: (16/121)*550 = 72.727 50
2 (observedexpected)
expected
2
2 [(49 * 550 /121)200]2
222.727
[(56 * 550 /121)300]2
254.545
[(16 * 550 /121)50]2
72.727
2.319 8.11689 7.102 17.54 p value 0.00002816 2.816 *105
![Page 17: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/17.jpg)
Wright-Fisher Model
In real life, populations are not infinite, and allele and genotype
frequencies do change over time
How do we extend from HWE to account for finite population, key
processes?
![Page 18: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/18.jpg)
Wright-Fisher ModelAssumptions:
• 2 allele system
• N diploid individuals in each generation
• 2N gametes
• Random mating, no selection, no mutation
• Discrete generationsA, red
a, green
Each generation, the new population is made by sampling with
replacement from the previous generation
![Page 19: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/19.jpg)
Let's play a few rounds of this game
![Page 20: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/20.jpg)
Faster by computer:
Generations
Alle
le c
ounts
Fixation!
![Page 21: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/21.jpg)
Again:Fixation!
Fixation!
![Page 22: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/22.jpg)
Let's investigate this phenomenon:
Change population size
Change allele frequencies
![Page 23: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/23.jpg)
Smaller population size….
![Page 24: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/24.jpg)
Larger population size….
![Page 25: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/25.jpg)
Larger population size….
![Page 26: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/26.jpg)
Initial red allele frequency > green allele frequency
![Page 27: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/27.jpg)
Initial red allele frequency > green allele frequency
![Page 28: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/28.jpg)
Initial red allele frequency > green allele frequency
![Page 29: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/29.jpg)
![Page 30: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/30.jpg)
Forw
ard
in tim
e
![Page 31: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/31.jpg)
Decay of Heterozygosity
Define Gt = homozygosity at generation t
= probability that a random draw of 2
chromosomes from the pop results in 2 of the
same allele
Ht = 1 – Gt = heterozygosity at generation t
= probability that a random draw of 2
chromosomes from the pop results in 2 different
alleles
Under Wright-Fisher assumptions, what happens to Ht (or Gt)
over time?
![Page 32: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/32.jpg)
Decay of HeterozygosityTwo ways to get 2 of the same allele:
Identical by
descent
Generation 0 Generation 1
Probability
= 1
2𝑁
Generation 0 Generation 1
Probability
= 1 −1
2𝑁* G0
Therefore 𝐺1 =1
2𝑁+ 1 −
1
2𝑁∗ 𝐺0
![Page 33: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/33.jpg)
Decay of HeterozygosityTwo ways to get 2 of the same allele:
Identical by
descent
Generation t Generation t+1
Probability
= 1
2𝑁
Generation t Generation t+1
Probability
= 1 −1
2𝑁* Gt
Therefore 𝐺𝑡+1 =1
2𝑁+ 1 −
1
2𝑁∗ 𝐺𝑡
![Page 34: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/34.jpg)
Decay of Heterozygosity: Proof
Therefore 𝐻𝑡 = 1 −1
2𝑁
𝑡
∗ 𝐻0
𝐺𝑡+1 =1
2𝑁+ 1 −
1
2𝑁∗ 𝐺𝑡
𝐻𝑡+1 = 1 − 𝐺𝑡+1 = 1 −1
2𝑁+ 1 −
1
2𝑁∗ 𝐺𝑡
= 1 −1
2𝑁− 𝐺𝑡 +
1
2𝑁∗ 𝐺𝑡
= 1 − 𝐺𝑡 −1
2𝑁1 − 𝐺𝑡
= 1 −1
2𝑁𝐻𝑡
= 1 −1
2𝑁
2
𝐻𝑡−1 = … = 1 −1
2𝑁
𝑡+1
𝐻0
![Page 35: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/35.jpg)
Insights from 𝐻𝑡 = 1 −1
2𝑁
𝑡
∗ 𝐻0
Half-life of H: at what t is Ht = ½ H0?
1
2= 1 −
1
2𝑁
𝑡
ln1
2= 𝑡 ∗ 𝑙𝑛 1 −
1
2𝑁
𝑡 =−ln(2)
ln 1 −12𝑁
𝑡 ≈ 2Nln 2 = 1.39 ∗ 𝑁
Use approximation
ln(1 + x) ≈ 𝑥
Larger N corresponds to longer half-life
For N = 10^4, t = (20000)*ln(2) = 1.39 * 10^4
![Page 36: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/36.jpg)
Insights from 𝐻𝑡 = 1 −1
2𝑁
𝑡
∗ 𝐻0
Half-life of H: t = 1.39 N
This indicates that even in a large population, eventually
every allele will have descended from a single allele in the
founding population! At a locus, all but 1 allele will have
"died off" (been lost)
(Remember: no selection, no mutation)
![Page 37: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/37.jpg)
Let's add selection to our model
Need to account for differing fitness conferred by differing
genotypes
Darwin versus Drift
![Page 38: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/38.jpg)
Define relative fitness for each possible individual
e.g.
Fitness RR = 1
Fitness RG = 1.1
Fitness GG = 2
Now the rules account for fitness:
Pick an individual with probability proportional to the fitness
of their genotype.
Here, GG is twice as likely to be chosen as RR.
Now choose 1 chromosome to put into the next generation
Wright Fisher v0.2
![Page 39: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/39.jpg)
What relative fitness should we select?
(Calibrate the scale)
Conserved elements < 0.01% increase in fitness
Wright Fisher v0.2
![Page 40: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/40.jpg)
Darwin versus Drift
I = # generations
A = # gametes
R = count of R allele
G = count of G allele
![Page 41: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/41.jpg)
Darwin versus Drift
![Page 42: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/42.jpg)
Darwin versus Drift
![Page 43: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/43.jpg)
Darwin versus Drift
![Page 44: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/44.jpg)
Survival of the fittest luckiest
Sometimes drift can overcome selection
Depends on allele frequency, population size
Most new advantageous mutations are NOT fixed!
Some startling results!
![Page 45: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/45.jpg)
Infinite alleles model
Mutation
![Page 46: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/46.jpg)
![Page 47: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/47.jpg)
Chance can play a large role in determining which
polymorphisms are fixed in a population
(Not obvious)
Amount of variation at a locus, and fate of individual alleles,
depends on mutation-selection-drift balance.
Summary so far
![Page 48: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/48.jpg)
Examined 131,060 Icelanders born after 1972
Compared with expectations from Wright Fisher model
• Considerable effect of genetic drift, even with rapid
population expansion rather than constant population size
![Page 49: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/49.jpg)
Recursion equations
Genotype Total
aa aA AA
Freq in generation t pt2 2ptqt qt
2 pt2 + 2ptqt + qt
2 =1
Fitness w11 w12 w22
Freq after selection pt2w11 2ptqtw12 qt
2w22 w' = pt2w11 + 2ptqtw12
+ qt2w22
𝑝𝑡+1 =𝑝𝑡2𝑤11 + 𝑝𝑞𝑤12
𝑤′
𝑞𝑡+1 =𝑞𝑡2𝑤22 + 𝑝𝑞𝑤12
𝑤′
Recursion equations for
analysis of selection
(No drift or mutation, discrete
generations, random mating)
![Page 50: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/50.jpg)
![Page 51: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/51.jpg)
![Page 52: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/52.jpg)
![Page 53: Population Genetics, Lecture 1genetics.wustl.edu/bio5488/files/2019/03/genomics...2019/03/18 · Population Genetics, Lecture 1 Nancy Lim Saccone nlims@wustl.edu Bio 5488, Spring](https://reader034.fdocuments.us/reader034/viewer/2022042303/5ecf05fd736fc27d4d7f0efe/html5/thumbnails/53.jpg)
The coalescent process
• "backward in time" process
• Lineage of alleles in a
sample traced backward in
time to their common
ancestor allele
• Genealogies are
unobserved, but can be
estimated
• Conceptual framework for
population genetic
inference: mutation,
recombination,
demographic history
• Kingman, Tajima, Hudson