Inferring recent or ongoing selection - - PowerPoint PPT Presentation

Inferring recent or ongoing selection http://popgen.dk/albrecht/BAG2017/web/

Anders Albrechtsen

Introduction Signatures of recent/ongoing selection

Recent selection

within species / using shared variation

Introduction Signatures of recent/ongoing selection

Sorry about the Human-centric talk

Good candidates for genes under recent selection

Introduction Signatures of recent/ongoing selection

Methods is applicable for most organisms

Examples of organisms with DNA

Introduction Signatures of recent/ongoing selection

Discrete time Write-Fisher Model

Figure: 5 generations

• We will sample with replacement to fill up the next jar.
• We assume the total number of marbles in the jar stays constant
• We are interested in the number of blue marbles at time t
• Let f n

blue = frequency of blue marbles at time n

• P[ no. blue marbles at time = n +1 |no. blue marbles at time = n]

Introduction Signatures of recent/ongoing selection

haploid view of selection based on the Wright fisher model

fitness and allele frequency changes (Expected) f n+1

A

= wANA wANA + waNa f n+1

A

= wAf n

A

wAf n

A + wafaAn

= f n

A

f n

A + wa/wAfaAn

selection coefficient

wa wA = 1 − s

f n+1

A

=

NA NA+(1−s)Na

Introduction Signatures of recent/ongoing selection

Selection over time (expectations)

Figure: generations given f 0

A = 0.01 s = 0.01

Introduction Signatures of recent/ongoing selection

Selection in diploid (expectations)

Diploid selection wAa wAA = 1 − SAa waa wAA = 1 − Saa Additive saa = 2sAa Dominant sAa = saa Recessive sAa = 0

Introduction Signatures of recent/ongoing selection

Allele frequency trajectory

Alleles frequency over time - what kind of selection?

Figure: time

Introduction Signatures of recent/ongoing selection

Allele frequency trajectory

neutral - drift - starting frequency 1/2N

Introduction Signatures of recent/ongoing selection

Probability of fixation

Introduction Signatures of recent/ongoing selection

strong positive selection

7 of 50 simulations reach fixation

Introduction Signatures of recent/ongoing selection

Weak positive

2 / 50 simulation reach fixation

Introduction Signatures of recent/ongoing selection

Weak negative

1 of 50 simulation reach fixation

Introduction Signatures of recent/ongoing selection

Neutral selection

Alleles can be removed,polymorphic or fixed figure from Matteo Fumagalli

Introduction Signatures of recent/ongoing selection

strong negative selection

alleles can be removed or be polymorphic

Introduction Signatures of recent/ongoing selection

Strong positive selection

Alleles can be removed, polymorphic or fixed

Introduction Signatures of recent/ongoing selection

Balancing selection

Alleles can be removed, polymorphic or fixed

Introduction Signatures of recent/ongoing selection

Summary

selections effect on alleles Neutral/weak removed, polymorphic or fixed Strong negative removed or polymorphic Strong positive removed, polymorphic or fixed Balacing removed, polymorphic or fixed Strong selection Depends on the population size Conclusion Allele frequency is (almost always) not enough to determine selection

Introduction Signatures of recent/ongoing selection

Need for additional information

Option 1 use information from the genomic region Option 2 Use information from mulitple species/populations Options 3 selection experiments External information

• Candidate genes/biological knowledge
• Functional categories
• Association to phenotypes

Introduction Signatures of recent/ongoing selection

Common methods used to detect selection

Introduction Signatures of recent/ongoing selection

Signature of selection

• Neutral locus
• Lots of variability

Introduction Signatures of recent/ongoing selection

Signature of selection

• Mutation enters the

population

Introduction Signatures of recent/ongoing selection

Signature of selection

• Negative selection

removed the allele

Introduction Signatures of recent/ongoing selection

Signature of selection

• Mutation enters the

population

Introduction Signatures of recent/ongoing selection

Signature of selection

• Mutation enters the

population

• Mutation increases in

frequency due to positive selection

Introduction Signatures of recent/ongoing selection

Signature of selection

• Increases LD
• Affects the variability

Introduction Signatures of recent/ongoing selection

Signature of selection

• Increases haplotype similarity

Introduction Signatures of recent/ongoing selection

Signature of selection

• Increases differences with
• ther populations in the

whole region

Introduction Signatures of recent/ongoing selection

What is the site frequency spectrum

Ind 11 T C G T C T C A A T 12 T C G T C T C C A G 21 A G G T C G C C A T 22 A C G T G G T C A T 31 A C T A G G C C T T 32 A C T A G G T C A T # Minor 2 1 2 2 3 2 2 1 1 1 Number of minor alleles (folded) η = (0.4, 0.5, 0.1)

1 2 3 Number of minor alleles Density 0.0 0.2 0.4

Introduction Signatures of recent/ongoing selection

What is the site frequency spectrum

Ind 11 T C G T C T C A A T 12 T C G T C T C C A G 21 A G G T C G C C A T 22 A C G T G G T C A T 31 A C T A G G C C T T 32 A C T A G G T C A T Outgroup A C T T C T C C A G # Derived 2 1 4 2 3 4 2 1 1 5 polarized SFS (unfolded) η = (0.3, 0.3, 0.1, 0.2, 0.1)

1 2 3 4 5 Number of minor alleles Density 0.00 0.10 0.20 0.30

Introduction Signatures of recent/ongoing selection

Frequency spectrum gives information about selection and demography

Introduction Signatures of recent/ongoing selection

Thetas are based on the frequency spectrum

Watterson θW = a−1 n−1

i=1 ηi, where a = n−1 i=1 1/i

Tajima θT = n

2

−1 n−1

i=1 i(n − i)ηi

Tajima’s D D =

θT −θW

√

Var(θT −θW ) under a neutral model* θT = θW

Introduction Signatures of recent/ongoing selection

Theta are based on the frequency spectrum

Watterson θW = a−1 n−1

i=1 ηi, where a = n−1 i=1 1/i

Tajima θT = n

2

−1 n−1

i=1 i(n − i)ηi

4 diploid individuals

0.0 0.2 0.4 0.6 η Ση i(n−i) = 0.39 0.19 0.13 0.1 0.08 0.06 0.06 0.25 0.43 0.54 0.57 0.54 0.43 0.25 1 1 1 1 1 1 1 watterson 0.39 tajimas 0.39 watterson 0.39 tajimas 0.39

Introduction Signatures of recent/ongoing selection

Theta are based on the frequency spectrum

Watterson θW = a−1 n−1

i=1 ηi, where a = n−1 i=1 1/i

Tajima π = θT = n

2

−1 n−1

i=1 i(n − i)ηi

4 diploid individuals

0.0 0.2 0.4 0.6 η η Ση i(n−i) = 0.66 0.17 0.07 0.04 0.03 0.02 0.01 0.39 0.19 0.13 0.1 0.08 0.06 0.06 0.25 0.43 0.54 0.57 0.54 0.43 0.25 1 1 1 1 1 1 1 watterson 0.39 tajimas 0.39 watterson 0.39 tajimas 0.39 watterson 0.39 tajimas 0.32 watterson 0.39 tajimas 0.32

Introduction Signatures of recent/ongoing selection

Thetas are based on the frequency spectrum

Watterson θW = a−1 n−1

i=1 ηi, where a = n−1 i=1 1/i

Tajima π = θT = n

2

−1 n−1

i=1 i(n − i)ηi

Fu & Li θFL = η1 Fay & Wu θH = n

2

−1 n−1

i=1 i2ηi

Zeng, Fu,Shi and Wu θL =

1 n−1

n−1

i=1 iηi

general ˆ θ = n

i=0 αiηi

Test statistics D =

θ1−θ2

√

Var(θ1−θ2) under a neutral model* θ1 = θ2

Difference weighting schemes for the SFS

Introduction Signatures of recent/ongoing selection

Why does selection affect the SFS

Introduction Signatures of recent/ongoing selection

Frequency spectrum gives information about selection and demography

Introduction Signatures of recent/ongoing selection

How to assess significance

slides stolen from Matteo Fumagalli

Introduction Signatures of recent/ongoing selection

How to assess significance

Introduction Signatures of recent/ongoing selection

How to assess significance

Introduction Signatures of recent/ongoing selection

How to assess significance

Introduction Signatures of recent/ongoing selection

How to assess significance

Introduction Signatures of recent/ongoing selection

How to assess significance

Introduction Signatures of recent/ongoing selection

How to assess significance

Introduction Signatures of recent/ongoing selection

How to assess significance

Introduction Signatures of recent/ongoing selection

How to assess significance

Introduction Signatures of recent/ongoing selection

Exercises

Let see how variability π and Tajimas D performs on famous examples of human adaptation. go to http://popgen.dk/albrecht/BAG2017/web/ Graphics When you will run analysis on the server you will need graphic (see above link)