Transcript Null hypotheses in evolutionary biology

The neutral theory of
molecular evolution
Motoo Kimura (1968)
High levels of polymorphism (variation) in
protein and DNA sequences among
individuals and species are difficult to
reconcile with mutation-selection equilibrium
(Ch 5.4)
Most mutations affecting fitness are deleterious,
hence quickly eliminated by selection
Ergo: Essentially all new mutations eventually
fixed are neutral, and evolve only by genetic
drift
• Do evolutionary biologists ever tire of
debating whether selection or drift dominates
the evolutionary process?
Why use DNA and protein
molecules to study evolution?
In principle, character homology
and independence can be
assured
Very large number of characters
can be studied
Only 4 precisely-defined character
states in DNA; 20 in protein
DNA sequencing is easy, fast, and
cheap; genome projects
Many (perhaps the majority) of
living species lack distinctive
morphological features
Neutral mutations accumulate in a
clocklike manner
Drawbacks to
molecular methods
In practice, orthology and
paralogy can be difficult to
distinguish in gene families
Not generally applicable to
fossil taxa
Gene phylogenies vs. species
phylogenies
Not all mutations are neutral
What makes the
molecular clock tick?
Probability of fixation of a neutral allele = p (the
current allele frequency)
• What is the fixation probability for a new neutral
mutation in a diploid population?
• 1/(2N)
New mutations arise at a rate of m (# mutations per
DNA base pair per generation; typically ~10-8)
• What is the frequency of new mutations in a
diploid population?
• (2N)m * L (length of genome in base pairs;
typically 108-1010 in eukaryotes)
Rate of fixation of new neutral mutations =
1/(2N) * (2N)m * L = mL
Since genome length (L) is a constant within a
species, neutral mutations go to fixation at a rate
equal to the mutation rate – m is the “ticking
speed” of the clock.
Which type of mutation should “tick” faster?
Why do some clocks
tick faster than others?
Amino acid substitutions per site, vertebrate a -globin
Empirical evidence for the molecular clock; from Hartl and Clark Principles of Population
Genetics, Fig. 12 p. 361
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0
100
200
300
400
500
Divergence time (millions of years)
C Pro
od m
in ot
C g ( er
od sl
in ow
g
(fa )
st
)
I
3' ntr
fla on
N Ps nk
on eu in
-s do g
yn g
on en
Sy y e
no mo
ny us
m
ou
s
Nucleotide substitutions per site per billion years
Evolutionary rates for different regions of genes; from Hartl and Clark Principles of
Population Genetics, Fig. 17 p. 373
5
4.5
4
3.5
3
2.5
2
1.5
1
0.5
0
Variation in substitution
rates
Nonsynonymous (‘replacement’)
substitution rates are variable and
relatively low. Why?
Promoter (5’ upstream) flanking
regions of genes have
intermediate rates of substitution
even though they are noncoding.
Why?
Synonymous (‘silent’) substitution
rates are high.
Intron substitution rates are high.
Pseudogene substitution rates are
highest. Why?
dN/dS
When will dN/dS < 1 ?
When will dN/dS = 1 ?
When will dN/dS > 1 ?
Genes with dN/dS >> 1
Major histocompatibility
complex (MHC)
Immunoglobulins
Self-incompatibility loci in plants
Sperm-egg recognition proteins
in abalone