FITNESS

1
FITNESS ADAPTATION
Fitness is a relative concept an allele with a
10 greater than average chance of representation
in the next generation has a 10 greater than
average fitness

• Adaptation is an absolute concept
• used to describe both
• (verb) the process by which a mean individual
  becomes fitter than its ancestors in a given
  environment, and
• (noun) a trait that is the outcome of past
  natural selection
• traits are adaptive if they improve mean fitness
  of individuals that have the traits in question

2
HOW TO STUDY ADAPTATION
Three conceptual stages

• Identify variants of trait
• Develop hypothesis of function of trait (develop
  alternative hypotheses and predictions)
• Test predictions using one of three methods
  (observation, experiment, comparison)

3
HOW TO STUDY ADAPTATION
Observational study

• Use when
• trait cannot be manipulated, and
• trait varies among individuals of the same
  species
• Type of data collected
• measurement(s) of the trait that is potentially
  adaptive (should vary among individuals within
  the same species)
• measurement(s) of fitness among individuals that
  vary in trait in questions (any measure of
  fitness will do life span, offspring, resource
  acquisition, etc.)

4
HOW TO STUDY ADAPTATION
Experimental study

• Use when
• trait can manipulated
• Type of data collected
• measurement(s) of the trait that is potentially
  adaptive (difference between control and
  experimental groups)
• measurement(s) of fitness among individuals that
  vary in trait in questions (any measure of
  fitness will do life span, offspring, resource
  acquisition, etc.)
• Control group should be manipulated in a
  similar way than the experimental group to insure
  that the variation in fitness is not caused by
  other factor than the trait that has been
  manipulated

5
HOW TO STUDY ADAPTATION
Comparative study

• Use when
• trait cannot be manipulated, and
• trait does not vary among individuals of the
  same species
• Type of data collected
• measurement(s) of the trait that is potentially
  adaptive (should vary among species)
• measurement(s) of fitness among species that
  vary in trait in questions (any measure of
  fitness will do life span, offspring, resource
  acquisition, etc.)
• use independent contrasts (see following slides)
  to establish if there is a correlation between
  trait in question and fitness

6
COMPARATIVE METHOD

• Conceptual idea
• all species share some evolutionary history with
  other species, and because of this shared
  history, species are not independent from each
  other
• This is a problem when testing hypotheses of
  cause and effect OR associations between traits
• if we find a significant relationship or
  correlation, we would conclude that the cause had
  the presumed effect or there is a correlation
  between traits because it occues across species.
• but this is WRONG this relationship or
  correlation might have occurred only once and is
  observed in multiple species because of shared
  evolutionary history, so that effectively this is
  a sample size of one (useless to test hypothesis)

7
COMPARATIVE METHOD
Solution compare species only across the
stretches on the tree for which they do not share
a common history
A
B
C
D

• Independent comparisons
• A to B
• C to D
• AB to CD

C D
A B
8
COMPARATIVE METHOD
For example, lets say you have values for two
traits (X and Y) for each species in your tree
X1 X2 X3 X4 Y1 Y2 Y3 Y4
A
B
C
D
9
COMPARATIVE METHOD
For example, lets say you have values for two
traits (X and Y) for each species in your tree
X1 X2 X3 X4 Y1 Y2 Y3 Y4
A
B
C
D

• You can compare
• A to B, and
• C to D, and

10
COMPARATIVE METHOD
For example, lets say you have values for two
traits (X and Y) for each species in your tree
X1 X2 X3 X4 Y1 Y2 Y3 Y4
A
B
C
D
X6 (X3X4)/2 Y6 (Y3Y4)/2
You can calculate averages to estimate trait
values for ancestral nodes on the tree, so that
comparison between AB to CD becomes possible
X5 (X1X2)/2 Y5 (Y1Y2)/2
A B
C D
11
COMPARATIVE METHOD
For example, lets say you have values for two
traits (X and Y) for each species in your tree
X1 X2 X3 X4 Y1 Y2 Y3 Y4
Independent contrasts
?Y
?X
A
B
C
D
X1-X2
Y1-Y2
X3-X4
Y3-Y4
X5-X6
Y5-Y6
X6 (X3X4)/2 Y6 (Y3Y4)/2
X5 (X1X2)/2 Y5 (Y1Y2)/2
The difference or contrast (?) should always be
calculated in the same direction on the tree
(subtract from left to right or right to left,
but be consistent for all the comparisons).
12
COMPARATIVE METHOD
You can then plot the contrasts in Y against
contrasts in X on a graph note because
contrasts are differences, the regression line
will always pass by the origin (0,0).
Each data point is now independent and the
relationship or correlation can be properly
tested without violating the assumption of
statistical independence of data points due to
shared evolutionary history.
?Y (INDEPENDENT CONTRAST FOR X)
0
0
?X (INDEPENDENT CONTRAST FOR Y)
13
STUDY OF ADAPTATION

• How to design a good experiment
• Consider the alternative hypotheses carefully
• Find clear predictions that distinguish among
  hypotheses
• Use good controls when appropriate, and
  replication if possible
• Interpretation of the data you should be able to
  tell how you would interpret the results in light
  of the predictions before you do the experiment.

14
ADAPTIVE LANDSCAPE

• An adaptive landscape is
• a tool to study fitness of populations
• is dynamic and varies as environmental effects
  on fitness vary

high
fitness
low
genotypes
15
ADAPTIVE LANDSCAPE
We can imagine a scenario where a few individuals
(a small population) colonize a new habitat. At
first, the pop will likely not be at the peak of
fitness. However, with time, selection will
favour genotypes that express phenotypes with
higher fitness.
high
selection
fitness
low
genotypes
16
ADAPTIVE LANDSCAPE
Our pop will slowly climb the fitness slope
towards the nearest fitness peak (the steeper the
slope, the faster the climb), until it reaches
it. Important selection can only drive pops up,
never down the fitness slope!
high
fitness
low
genotypes
17
ADAPTIVE LANDSCAPE
Mutations occurring at every generation will tend
to smear the pop slightly down the peak (remember
that mutations are usually slightly deleterious
or neutral).
high
fitness
low
genotypes
18
ADAPTIVE LANDSCAPE
But selection will act to bring the pop back to
the fitness peak (stabilizing selection).
high
selection
fitness
low
genotypes
19
ADAPTIVE LANDSCAPE
Adaptive landscapes can be fairly complex and
have more than one fitness peak.
high
fitness
low
genotypes
20
ADAPTIVE LANDSCAPE
Lets say that our newly arrive pop starts on the
slope of the lower peak even if that pop doesnt
know that there is a higher peak on the
landscape, it would do better if it was there.
How could it reach it?
high
selection
fitness
low
genotypes
21
ADAPTIVE LANDSCAPE
How could it reach it? -gt genetic drift can allow
a pop to go down the fitness slope through the
random fixation of alleles.
high
drift
fitness
low
genotypes
22
ADAPTIVE LANDSCAPE
Or, if the fitness landscape changes due to
environmental change, a pop could move to a new
region of the landscape through selection.
high
selection
fitness
old landscape
low
new landscape
genotypes