DEFINITION OF EVOLUTION
To Darwin, evolution was: "descent with modification".
Evolution can be
a change in morphology, ecology, behaviour or physiology. Underlying
all these changes must be genetic change,
if these changes are evolutionary. A modern, genetic definition:
WHAT CAUSES
EVOLUTION?
a) Natural selection
b) Mutation
c) Genetic drift, or neutral, random evolution
e) Migration or gene flow
This lecture: simple examples of evolution by natural selection.
WHAT IS NATURAL
SELECTION?
Selection may
be for VIABILITY (survival) vs. FERTILITY (number of offspring).
Natural selection often causes evolution, but natural selection may prevent evolutionary change (e.g. stable polymorphism; next lecture), and evolutionary change does not require natural selection (e.g. neutral evolution, or genetic drift). Potentially important: over 90% of your genome may be evolving neutrally -- our DNA may be literally mostly "junk"!
Some of the most interesting parts of evolution, of course, involve natural selection.
Evolution is a fact (or is it?); for a couple of the many other philosophical questions about evolution, click here.
SELECTION AT A SINGLE LOCUS
Many "quantitative traits" like size and behaviour patterns are inherited at multiple loci. However, simpler single-locus traits provide ideal examples of evolution by natural selection. Many single-locus traits seem to have evolved because of stress, especially recent stress imposed by humans. Selection has led to the evolution of resistance to these challenges. Some examples:
How does evolution by natural selection work? Evolution by natural selection is an inevitable, mathematical process. The frequency of a particular allele will change; its rate of change will depend mathematically on the advantage (or relative fitness) of that allele.
MATHS! EEK! This doesn't sound very pleasant, and indeed the maths can get very complicated (though not in 2007). However, much of the excitement in modern evolutionary theory is about theory. Even though we don't expect you to get to grips with all of it, we will try to give you an outline.
We give tutorials, as well as lectures, and practice doing the problems to help you understand. After doing these examples you will be able to do exam problems easily.
Mathematical evolutionary theory is useful. For example, given information about natural selection, how rapidly will evolution occur? The answers help us to take precautions about antibiotic resistance, or pest resistance, for instance. Evolution is a predictive science!
Evolution is controlled by the strength of selection, or, to put it another way, the fitnesses of different genotypes. The actual fitness is the number of offspring that an individual has during its lifetime. However, in evolution we are interested whether there is a tendency for one form to replace another, not in the actual numbers of individuals or actual fitnesses. We are more interested in the relative fitness.
We can represent selection against a genotype by some fractional amount, call this s, which is the selection coefficient. The relative fitness can be represented as 1 - s, compared with 1 for a different genotype.
Genotypes |
|
|
|
|
Relative fitness, W |
|
|
|
|
From the Hardy-Weinberg law: |
||||
Frequencies before selection |
|
|
|
|
Frequencies after selection |
|
|
|
|
= 1 - sq2 | ||||
So ... new genotype frequencies |
|
|
|
|
What is the new gene frequency, p’ ?
p’ = new frequency of AA + 0.5
x new frequency of Aa
What is the RATE of evolution per generation? We need to know:
the CHANGE OF GENE FREQUENCY, (obtained by subtracting old gene frequency from the new gene frequency).
This doesn't look like much fun, but equations like these ARE THE BASIS OF ALL EVOLUTION BY NATURAL SELECTION!!
We now have a basic equation for the effect of natural selection on the change in gene frequency, or evolution. Arranging this as a flow diagram:-
So now you could easily write a computer programme for evolution by natural selection!
Complications – There are many! For instance:
1) Many different kinds of selection:
4) Dominance not complete; use intermediate
value of h for dominance;
Genotypes | AA | Aa | aa |
fitnesses | 1 | 1-hs | 1-s |
if h=0, then allele A is dominant
(genotypes AA and Aa have similar fitnesses);
if h=1, then A is recessive (Aa
and aa have similar fitnesses);
if h=0.5, then A is exactly "co-dominant",
or with intermediate fitness.
5) Multiple genes, and so on...
But in all these different cases, the basic principle remains the same! Simple models like the one we just studied are the basis of all models in population genetics.
So, we can now answer
the question:
Answer:
If p is small, ~0.01 or less, ,
i.e. RAPID
If p is large, so that q
~ 0.01 or less,,
i.e. very SLOW,
because q2 is the square of a very small number, and
so is itself even smaller!
SO:
Selection for/against a DOMINANT allele at low frequency is RAPID (~p)
Selection for/against a RECESSIVE allele at low frequency is SLOW (~q2)
THREE POSSIBLE METHODS:
1) Estimate from change of gene frequencies per generation ; the result of selection. Example: peppered moth melanic frequency (CC and Cc, see below) increased from 0.01 in 1848 to 95% in 1898; JBS Haldane estimated s = 0.20 agains the nonmelanic recessive, c.
2) Distortion of Hardy-Weinberg ratios - problems? see next lecture.
3) Comparison of birth or death rates between individuals (W = relative measure of fitness of the different genotypes). This is the MOST DIRECT METHOD
USING METHOD (3) TO ESTIMATE SELECTION
IN PEPPERED MOTH
e.g. survival in a field experiment on
Kettlewell's experiments on the peppered moth
A) Central Birmingham
number
released |
number
recaptured |
proportion
recaptured |
relative
fitness |
W relative
to typica |
|
typica | 144 | 18 | 0.125 | 0.43 | 1.00 |
carbonaria | 486 | 140 | 0.288 | 1.00 | 2.30 |
B) Dorset wood
number
released |
number
recaptured |
proportion
recaptured |
relative
fitness |
W relative
to typica |
|
typica | 163 | 67 | 0.411 | 1.82 | 1.00 |
carbonaria | 142 | 32 | 0.225 | 1.00 | 0.55 |
SUMMARY OF FITNESSES (Note, the fitness of cc genotypes, Wcc = 1 - sc):
typica | carbonaria | carbonaria | selection
coefficient against typica |
|
Wcc | WCc | WCC | sc | |
City | 0.43 | 1 | 1 | +0.57 |
Wood | 1.82 | 1 | 1 | -0.82 |
HOW FAST WILL CARBONARIA INCREASE IN FREQUENCY in a city (where sc = 0.57)?
= spq2/(1-sq2); suppose p = 0.5 to start with:
= 0.57 x 0.5 x 0.52 / (1 - 0.57x0.52) = 0.08; carbonaria will increase by 8% per generation.
FUTUYMA, DJ 1998.
Evolutionary Biology. Chapters 12 and 13 (pp. 371-381).
References
on natural selection at single genes for resistance
Science Lbrary:
View 2007 Teaching Collection by going to eUCLid;
use Keyword, Basic Search, All Fields: 2007