Multivariate Characters
Univariate Multivariate
Continuous distribution, one dimension Correlated responses
Drosophila bristle number, Stalk-eyed Stalk-eyed flies: male eyespan and
fly eyespan female preference
Genetic basis of phenotypes Arm and leg length and height (a tall
Additive gene effects (h^2 = VA/VP) person will also have long legs and long
Traits evolve independently arms)
Trait specific adaptation/optimisation Hair colour and eye colour etc.
Straightforward response to
directional selection
Where do multivariate characters come from?
The same genes affect more than one trait
- Different characters of a same individual
- Same character in males and females
- Same character in different environments
How does the shared genetic basis affect trait evolution?
Describe genetic variation
- For one trait additive genetic variance = σ2A
- For 2 traits, variation can be represented as a
matrix
Evolution of multivariate characters
For a univariate character Δz = σ2Aβ
(change in trait value, additive genetic
variance, strength of selection)
For multivariate characters
- Interference if traits covary (σ2A,12,
σ2A,21 ≠ 0) – selection on one trait
affects the other trait
- Correlated responses: one trait
selected, both change
- Accelerated responses: eg. positive correlation, selection aligned
- Adaptive conflicts: eg. positive correlation, selection opposed
Correlated Responses (Reeve and Fairbaim, 1996)
, Artificial selection on D. melanogaster
thorax length
Additional intersexual correlation rMF =
0.93
Additive intersexual covariance σ2A,12 > 0
Correlated responses between the
sexes – similar pattern regardless of
which one was selected or both sexes
Adaptive Conflicts
Sexual Dimorphism
Reproductive roles demand for different
types of resources
- Males : Acquire mating partners so
the more mates, the more offspring is
produced
- Females: Take resources to rear offspring
acquiring many mates is not necessary, if fact more
than 1 mate (number of offspring decreases )
(Bateman 1948)
There are opposing selection pressures on the 2
sexes
- Eg. Locomotion in D. melanogaster (Long and Rice,
2007)
- Females with higher locomotory activity have
lower fitness, but males are the opposite
(presumably more motion, energy to perform mating dance and to compete with
other males)
However, interspecific genetic correlations are high, as most genes are shared
between the sexes (autosomes)
- Y/W chromosomes carry few genes
- How is sexual dimorphism brought about?
By definition, males and females of any species
with separate sexes differ phenotypically (Hesketh et al., 2013)
, - Sexual dimorphism reflects divergent selection on male and female
morphological, physiological and behavioural characters
- Male phenotype often significantly shaped by sexual selection on traits that
increase fertilisation success, whereas females will reflect for increased
fecundity
Evolution of sexual dimorphism is often incomplete and sexual antagonism persists
in many populations
- Sexual antagonism arises when alleles that benefit the fitness of one sex causes
deleterious effects in the other sex
- Sexually antagonistic alleles are an evolutionary step towards the evolution of
sexual dimorphism
Theory predicts that divergent selection
on male and female phenotypes can favour
invasion of antagonistic mutations, as long
as their benefit in one sex outweighs the
cost in the other
- Evidence for sexual antagonism in D.
melanogaster look at 40 genotypes
between females and males
negative correlation/relationship between males and females
Once antangonistic variation segregates in the population, selection will favour the
invasion of modifiers that limit the expression of antagonistic alleles to the
favoured sex, thus resolving antagonism and allowing sexes to diverge phenotypically
(Rice, 1984)
- Mutations that break down genetic correlation, sex-specific gene expression
- Conflict resolution in waltzing flies for various traits, eg. Wing, thorax length
(Bonduriansky and Row, 2005)
- Acrosss species – meta-analysis also gives similar correlation (Poissant et al.,
2009)
- Adaptive conflicts are resolved (but data: correlations!)
- Sexual dimorphism is ubiquitous – some antagonistic variation remains –
unresolvable? Turnover creation-resolution?