1. Genes may interact additively, multiplicatively, or epistatically
Epistatic selection favors individuals with specific combinations
of alleles at different loci
Epistasis is suggested by violation of two-locus hwe
Linkage disequilibrium is the non-random association of alleles at
different loci; D measures the degree of non-random
association, scaled by allele frequencies in the population
Transient LDE can be produced by drift or admixture; Permanent
LDE is caused by non-random mating or selection
Lde may be relatively uncommon; But direct estimation from pairs
Of loci likely to interact is difficult

2. Sexual reproduction
1) Physical recombination of chromosomes
- DNA repair
- allelic shuffling
2) Outcrossing
- meiosis
- syngamy (genes in zygotes are from two
unrelated individuals
Consensus: recombination initially a DNA repair
mechanism; allelic recombination has been
favored subsequently

3. Evolution of outcrossing
1) group selection explanations
- short term advantage to asexual taxa (cost of meiosis),
but adapted to a specific environment

4. Cost of Meiosis -- Maynard Smith (1978)
in a few generations, an asexual species will numerically
supplant a sexual one!

5. Evolution of outcrossing
1) group selection explanations
- short term advantage to asexual taxa (cost of meiosis),
but adapted to a specific environment
-long term certainty of environmental change
-predicts lifespan of asexual << sexual species

6. Whiptail Lizards
Cneimodophorus
e.g. “uniparens”

7. Bdelloid rotifers
asexual 40 million years
diversify into 360 species
gene diversity in 4 species vs
sexual rotifers (Welch and
(Meselson 2003)

8. Evolution of outcrossing
1) group selection explanations
- short term advantage to asexual taxa (cost of meiosis),
but adapted to a specific environment
-long term certainty of environmental change
-predicts lifespan of asexual << sexual species
BUT -relies on selection above the individual level
(species competition)
-selection acts in the present, not the future
(short term advantage ??)
-why not an asexual world with high species turnover??

9. selection is operating on the individual
but many species are
capable of both sexual
and asexual reproduction
sexual reproduction occurs
when the environment
becomes unstable
asexual reproduction under
stable environmental
conditions
selection is operating on
the individual

10. Evolution of outcrossing
1) group selection explanations
- short term advantage to asexual taxa (cost of meiosis),
but adapted to a specific environment
-long term certainty of environmental change
-predicts lifespan of asexual << sexual species
BUT -relies on selection above the individual level
(species competition)
-selection acts in the present, not the future
(short term advantage ??)
-why not an asexual world with high species turnover??
short-term, individual advantage to sexual reproduction

11. Linkage disequilibrium arises from:
genetic drift selection
avoid
deleterious
combine
favorable
Muller’s ratchet mutational load*
mutation rate environmental
heterogeneity

12. Individual selection models
1) Muller’s ratchet
asexual species
- clones vary in the total number of deleterious
mutations they carry
- each generation, the clone with the highest fitness
(lowest number of mutations) may be lost by
drift or additional mutation
- the average number of deleterious mutations in the
population will continually increase
- the average relative fitness will continually decline;
eventually the population becomes extinct
sexual species
- recombination generates individuals with
a decreased number of deleterious alleles
- recombination reconstitutes the highest fitness
categories (zero mutations)

13. How the ratchet works: in any generation the class with
the fewest mutations may
be lost by chance (drift) or
by acquisition of another
mutation
forward mutation is more
common than reversion;
the distribution will
continue to shift to the right
Number of Individuals
Number of deleterious mutations

14. 444 lines of Salmonella typhimurium
Andersson and Hughes 1996
444 lines of Salmonella typhimurium
~1700 generations (60 growth cycles)
5 lines (1%) had significantly
slower growth than
ancestor Wt
mut
mut
mut
mut
mut
28 generations

15. Linkage disequilibrium arises from:
genetic drift selection
avoid
deleterious
combine
favorable
Muller’s ratchet mutational load*
mutation rate environmental
heterogeneity

16. Individual selection models
2) environmental heterogeneity
assumes: trade-off in performance in different environments
asexual species – offspring competitively superior in
parent’s environment, but only in that environment
sexual species
– genetically diverse offspring can compete in a
large number of environments
requires rapid change in the environment ---> acquisition of new
combinations of alleles faster by recombination than
mutation

17. A1B1 - +
0 frequency of A + -
0 frequency of B
A2B2

18. fitness of host genotype with two types of parasites
H1 H2
parasite P
genotype P
fitness of parasite genotype in two types of host
parasite genotype
P1 P2
host H
genotype H

19. H Frequency of H2 H2
Frequency of P2 P2 P1
Host
Parasite

20. Red Queen Hypothesis Parasite Host “Now here, you see, it takes all
the running you can do, to keep
in the same place. If you want to
get somewhere else, you must
run at least twice as fast as that.”

21. Lively 1992
freshwater snail, Potamopyrgus antipodarum
two types of females:
obligate sexuals ---> %%, &&
obligate parthenogens ---> &&
trematode parasite, Microphallus
infected snails are sterile
lakes (snail popns) vary in:
relative frequency of sexuals
vs. asexuals
relative frequeny of trematode
infections

22. frequency of sexual reproduction increased with incidence of parasite

23. Dybdhal and Lively 1998
Snail and Trematode host
Red Queen model predicts negative frequency dependent selection
- oscillations in host genotype frequency are a consequence
of selection by parasites
Lake Poerua asexuals only
- measure abundance of 112 snail clones over 5 years
- only four are common (> 15% of total; 50-60% of sample)
- frequency of the 4 common clones in random and parasite
infested samples varied significantly among years

24. *
*** *

25. Dybdhal and Lively 1998
snail and trematode host
Red Queen model predicts negative frequency dependent selection
- oscillations in host genotype frequency are a consequence
of selection by parasites
Lake Poerua asexuals only
- measure abundance of 112 snail clones over 5 years
- only four are common (> 15% of total; 50-60% of sample)
- frequency of the 4 common clones in random and parasite
infested samples varied significantly among years
- significant overinfection was time-lagged

27. Dybdhal and Lively 1998
snail and trematode host
Red Queen model predicts negative frequency dependent selection
- oscillations in host genotype frequency are a consequence
of selection by parasites
Lake Poerua asexuals only
- measure abundance of 112 snail clones over 5 years
- only four are common (> 15% of total; 50-60% of sample)
- frequency of the 4 common clones in random and parasite
infested samples varied significantly among years
- significant overinfection was time-lagged
- the prevalence of infection in rare (n=40) clones was
significantly lower than in common clones

29. cost of meiosis suggests that asexually reproducing taxa
should be more common than sexually reproducing ones
group selection models explain the advantages of sex by long-term
pay-offs, but cannot account for a short-term advantage
most asexual taxa are relatively young (vs. related sexual species)
Muller’s ratchet works, but is too slow for other than microbes
hypotheses that depend on envionmental heterogeneity currently
have the best empirical support
parasites and pathogens may be an important selective force
favoring sexual reproduction; host-parasite systems are
inherently frequency-dependent

30. Evolution of Quantitative Traits
Phenotype
Frequency
Phenotypes are often (usually) continuous (rather than discrete)
and often have a normal distribution
Height, weight, bill depth, resistance to a parasite, etc.
Why??
Several to Many genes may affect phenotype
Effect of environment as well as the effect of genes

31. P = G + E The Phenotype is produced by the Genotype, and the
Environment
P = G E
Genotype
Phenotype

32. P = G + E The Phenotype is produced by the Genotype, and the
Environment
P = G E
Genotype
Phenotype

33. P = G + E The Phenotype is produced by the Genotype, and the
Environment
P = G E
Genotype
Phenotype

34. P = G + E The Phenotype is produced by the Genotype, and the
Environment
P = G E
We will always look at the Phenotype, the Genetic component and
the Environmental component with respect to the population mean
Genotype
Phenotype
Pi = pi - p _
Gi = gi - g
Ei = ei - e