Selection Experiments

1
Lecture 13 Selection Experiments Experimental
Evolution
2
Selection Experiments(a.k.a. Experimental
Evolution)
1. Experimental way to study evolution in
action 2. The earliest form of genetic
engineering 3. A way to produce useful
organisms 4. The most direct and convincing test
of whether a trait shows any additive genetic
variance in the population 5. A modern corollary
to the August Krogh Principle. If a suitable
model does not exist, then create one! Bennett,
A. F. 2003. Experimental evolution and the Krogh
Principle generating biological novelty for
functional and genetic analyses. Physiological
and Biochemical Zoology 761-11.
3
Selection Experiments
6. A powerful way to demonstrate
mechanism, i.e., how organisms work a. Select
on an organismal trait b. Observe correlated
response in lower- level trait that you
hypothesize causes the organismal
difference c. Test that hypothesis by
performing a second experiment, selecting on
the lower-level trait d. Does the organismal
trait change as predicted?
4
Hypothetical Example
a. Select for long life span in mice b. Observe
correlated increase in anti-oxidant enzyme
activities c. Select for high anti-oxidant
enzyme activities (e.g., biopsy individuals to
score their phenotype and then choose
breeders) d. Does life span increase as
predicted?
5
Types of "Selection Experiments" Artificial
Selection Captive populations in which
individuals in each generation are measured for a
phenotypic trait (or combination of traits). Some
top or bottom fraction of individuals is then
chosen as the breeders to produce the next
generation. This is called "truncation
selection" or "mass selection." One variation is
taking the highest-scoring (or lowest-scoring)
male and female from within each
family. Within-family selection increases the
effective population size (Ne), reduces rate of
inbreeding, and helps to eliminate confounding
influences of some maternal effects. But, it also
reduces the possible intensity of selection as
compared with "mass selection," which involves
choosing breeders without regard to their family
membership.
6
Laboratory Natural Selection Individual
phenotypes are not measured each generation, nor
are breeders specifically chosen by the
investigator. Rather, a freely breeding
population is exposed to altered environmental
conditions, such as different temperatures or
salinities, or to altered husbandry conditions,
which could favor changes in demographic
schedules. Assuming that additive genetic
variance exists for relevant traits, the
population will adapt to the new conditions.
Most common with non-vertebrates, including
Drosophila, bacteria, and viruses, but have also
been employed with vertebrates Barnett and
Dickson housed mouse colonies at room
temperature or around 0o Celsius.
7
Intentional Field Introductions
Manipulations David Reznick's guppies in
Trinidad Anolis lizards introduced to Caribbean
islands (Tom Schoener, Jonathan
Losos) "Accidental" Introductions
Manipulations Drosophila studied by Ray Huey and
colleagues House Sparrows Adaptations of fishes
to ponds heated by nuclear power plants Adaptatio
ns of plants to living on mine tailings
8
Examples of Selection Experiments
9
Note physical lower limit of zero oil
Ridley, 1996, p. 238
10
r h2 s only works for initial generations
Ridley, 1996, pp. 236-239
11
Mice
Bunger, L., A. Laidlaw, G. Bulfield, E. J. Eisen,
J. F. Medrano, G. E. Bradford, F. Pirchner, U.
Renne, W. Schlote, and W. G. Hill. 2001. Inbred
lines of mice derived from long-term growth
selected lines unique resources for mapping
growth genes. Mammalian Genome 12678686.
12
Rats
Tryon, R. C. 1929. The genetics of learning
ability in rats. Univ. Calif. Publ. Psychol.
471-89.
13
Rats
Ridley, 1996, p. 45
Hunt, H. R., C. A. Hoppert, and S. Rosen. 1955.
Genetic factors in experimental rat caries. Pages
66-81 in R. F. Sognnaes, ed. Advances in
experimental caries research. American
Association for the Advancement of Science,
Washington, D.C.
14
Selection for Ethanol Sleep Time in Laboratory
Mice
Plomin, R., J. C. DeFries, and G. E. McClearn.
1990. Behavioral genetics A primer. 2nd ed. W.
H. Freeman, New York. 455 pp.
15
Selection for Ethanol Sleep Time in Laboratory
Mice
Note complete separation of short- and
long-selected lines
Plomin, R., J. C. DeFries, and G. E. McClearn.
1990. Behavioral genetics A primer. 2nd ed. W.
H. Freeman, New York. 455 pp.
16
Selection on Open-field Activity in Mice
Method developed by C. S. Hall in 1930s to
measure levels of fear and emotional reactivity
in rodents
Video camera
17
Mice
Important features of experimental
design replication up, down, and control lines
The direct response to selection. Note
consistency of response between replicates.
18
A correlated response to selection. Note somewhat
lower consistency of correlated response between
replicates.
19
Coat color also changes! An example of
pleiotropy, one of the main causes of genetic
correlations. Again, note consistency of
replicates.
20
Selection for Thermoregulatory Nesting
Lynch, C. B. 1980. Response to divergent
selection for nesting behavior in Mus musculus.
Genetics 96757-765.
The base population was a genetically
heterogeneous stock of lab mice (Mus musculus)
originally derived from an 8-way cross among
inbred strains.
May be considered the first rodent selection
experiment in "evolutionary physiology."
21
Selection for Thermoregulatory Nesting
The overall realized heritability pooled across
lines and replicates was 0.18 0.02 (0.15 0.03
for high nesting scores and 0.23 0.04 for low
nesting scores), or 0.28 0.05 when adjusted for
within-family selection.
r h2 s
h2 r/s
22
Selection for Thermoregulatory Nesting
Lynch, C. B. 1994. Evolutionary inferences from
genetic analyses of cold adaptation in laboratory
and wild populations of the house mouse.
Pages 278-301 in C. R.
B. Boake, ed. Quantitative genetic studies of
behavioral evolution. Univ. Chicago Press.
60 50 40 30 20 10 0
High
Cotton Used in 4 Days (g)
Control
Low
0 5 10 15 20 25
30 35 40 45
Generation
23
Selection for Thermoregulatory Nesting
Lynch, C. B. 1994. Evolutionary inferences from
genetic analyses of cold adaptation in
laboratory and wild populations of the house
mouse. Pages 278-301 in C. R. B.
Boake, ed. Quantitative genetic studies of
behavioral evolution. Univ. Chicago Press.
24
(No Transcript)
25
Artificial Selection on Longevity in Mice Nagai,
J., C. Y. Yin, and M. P. Sabour. 1995. Lines of
mice selected for reproductive longevity. Growth,
Development Aging 5979-91. Mice were
pair-mated, and their offspring from parities5 -
9 were used as breeders for the next
generation. Age at Last
Young Life- At gener. 16 Parturition
Born span (days) Alive
(days) Selection line 1 297 77
378 Selection line 2 299 83
437 Control line 191 46 347
26
Remember that narrow-sense heritability indicates
whether a trait tends to "run in families" and
can be estimated as the slope of the regression
of offspring mean on midparent mean
27
Least-squares linear regression to estimate
heritability
N 50
Mean of Offspring - Trait A
Mean of Parents - Trait A
28
Similarly, a genetic correlation indicates
whether two traits tend to "run in families" and
can be estimated as the slope of the regression
of offspring mean for one trait on midparent mean
for the other trait (or vice versa)
29
This plot indicates that that Traits A and B tend
to go together in families, and so are positively
genetically correlated. Therefore, selection to
increase one would also cause an increase in the
other. Hypothetical example Trait A might be
body mass and Trait B might be tail length.
Mean of Offspring - Trait B
Mean of Parents - Trait A
30
SCOR1PCT 100 (Exam1 Exam2 Critique1
Critique2)/160