Download Lecture 26 Molecular Evolution

NEW TOPIC:
MOLECULAR EVOLUTION
THE NEUTRAL THEORY OF MOLECULAR EVOLUTION
 Genetic drift causes more substitutions than does
natural selection.
 Molecular evolution is a balance between drift and
mutation.
Motoo Kimura (1968)
OBSERVATIONS THAT PROMPTED THE NEUTRAL THEORY
 Observed rates of amino acid substitutions
in proteins were surprisingly high.
 The amount of heterozygosity in natural
populations seemed too high to be explained
by selection.
Amino Acid Differences
UNIFORM RATE OF AMINO ACID SUBSTITUTION
FUNCTIONALLY IMPORTANT PARTS OF THE
AMINO ACID SEQUENCE EVOLVE MORE SLOWLY
THAN OTHER PARTS
VARIATION IN THE RATE OF EVOLUTION IN
DIFFERENT TYPES OF DNA
THE FIXATION OF NEUTRAL MUTATIONS
 For any locus, 2Nµ mutations enter a population each generation.
 Each new neutral mutation has a probability of being fixed equal to its
initial frequency, 1/(2N).
 The average number of substitutions per locus per generation is equal to
the product of these,
2Nµ x 1/(2N) = µ
BASIC TENANT OF THE NEUTRAL THEORY:
The substitution rate = the mutation rate
=µ
EQUILIBRIUM HETEROZYGOSITY
 Basic Neutral Theory models consider only panmictic
populations, so migration and nonrandom mating are
ignored.
 By eliminating considerations of selection as an
important force in molecular evolution, neutral theory
models a balance between drift and mutation.
 As a result, at equilibrium,
F = 1/(1 + 4Nµ), and
H = 4Nµ/(1 + 4Nµ)
PREDICTION:
 Because the strength of random genetic drift
depends on Ne, neutral mutations should take
longer to drift to fixation in large populations (on
average, 4Ne generations).
 Therefore, large populations should contain
many alleles at intermediate frequency,
whereas small populations should exhibit little
polymorphism.
POSITIVE RELATIONSHIP BETWEEN
HETEROZYGOSITY AND POPULATION SIZE
PREDICTION:
 GENES, OR GENE REGIONS, WITH A
HIGHER SUBSTITUTION RATE SHOULD
SHOW INCREASED POLYMORPHISM.
Low µ
High µ
REGIONS OF PROTEIN CODING GENES WITH HIGH
SUBSTITUTION RATES HAVE HIGH LEVELS OF
POLYMORPHISM
FIXATION OF NEUTRAL VS. BENEFICIAL ALLELES
Drift Dominates if:
1/2N >> 2s
Selection dominates if:
1/2N << 2s
Probability of fixation of a new mutant allele:
Drift:
Selection:
1/2N
2s
 The rate of mutation (per year) may increase with
metabolic rate and decrease with generation length:
ENDOTHERMS
ECTOTHERMS
MUTATIONS MAY HAVE DIFFERENT
EFFECTS ON FITNESS
 Synonymous mutation: do not alter the
amino acid sequence of the protein
 Often selectively neutral
 Non-synonymous mutation: alter the amino
acid sequence of the protein
 More likely to be subject to selection
COROLLARY OF THE NEUTRAL THEORY:
 NEUTRAL (SILENT) MUTATIONS HAVE HIGHER
SUBSTITUTION RATES THAN MUTATIONS UNDER
SELECTION.
 Most of the human genome (>95%) is noncoding. Most
mutations in these regions are silent.
 Approx. 24% of base substitutions in protein coding
regions are silent (synonymous).
SYNONYMOUS VS. NONSYNONYMOUS
SUBSTITUTIONS IN THE  - GLOBIN PROTEIN CODING
REGION
NONSYNONYMOUS
SYNONYMOUS
SYNONYMOUS VS. NONSYNONYMOUS SUBSTITUTIONS
SYNONYMOUS RATE
NONSYNONYMOUS RATE
DETECTING SELECTION ON DNA SEQUENCES
 Synonymous substitutions: do not change protein
 Should evolve at a neutral rate
 Estimated by ks (number of synonymous substitutions / 1 kb)
 Nonsynonymous substitutions: change protein
 Faster evolution than synonymous sites indicates positive selection
 Slower evolution than synonymous sites indicates purifying
selection
 Estimated by ka (number of nonsynonymous substitutions / 1 kb)
 The ratio ka/ks can be used to infer the action of selection
on protein coding genes.
Signature of positive selection
Positive selection on FOXP2
ESTIMATION OF THE MUTATION RATE PER
NUCLEOTIDE SITE
ANCESTRAL SPECIES
t
years
SPECIES A
SPECIES B
µ = mutation rate per nucleotide site per year
Expected number of changes per site between species A and
B at synonymous (silent) sites = 2 t µ
Observed number of changes per site between species A and
B=D
Estimated mutation rate = µ = D / (2 t)
1 sub./site/100 MY
1 sub./site/100 BY
SYNONYMOUS SUBSTITUTION RATE
(substitutions/site/billion years)
ESTIMATED RATES OF MUTATION PER NUCLEOTIDE SITE
n = NUCLEAR GENOME
m = MITOCHONDRIAL GENOME
c = CHLOROPLAST GENOME
FROM: Lynch & Blanchard. 1998.Genetica 102-103:29-39.
THE MOLECULAR CLOCK
 Because the substitution rate = µ, the rate of
molecular evolution should be a constant over time
as long as the mutation rate does not change over
time.
DATING THE TIME OF EVOLUTIONARY
DIVERGENCE BETWEEN TWO LINEAGES
ANCESTRAL SPECIES
t
SPECIES 1
AATGGTGGCT
SPECIES 2
AATCCGGGCT
d = estimated rate of nucleotide substitutions / site/ year
(from a calibrated molecular clock)
D = observed fraction of nucleotide sites differing between
two species (0.3 in the example above)
2 x t x d = expected fraction of nucleotide differences between
species
Estimated time of divergence = t = D / (2d)
The molecular clock can be
calibrated with dates from
the fossil record
46
32
30
21
16
DATING THE TIME OF EVOLUTIONARY
DIVERGENCE BETWEEN TWO LINEAGES
EXAMPLE:
 The substitution rate per site is d = 10-8, and the
nucleotide sequences from two species differ at a
fraction D = 0.2 of the nucleotide sites.
 We estimate the time since isolation between the two
lineages as:
t = 0.2 / (2 x 10-8) = 107 years
CAVEATS TO THE MOLECULAR CLOCK
 Two sequences may have similar sequences BUT
not because of common descent.
Species A:
T
C
G
Species B:
T
A
G
 Convergence – lineages may diverge and then
acquire the same base by further mutation.
MUTATIONAL SATURATION
 Since nucleotides have only four possible states
(A,G,T,C), DNA sequences will not be expected to
diverge 100% even after an infinitely long period of time.
Maximum Divergence: 75%
 Mathematical models can be used to correct for the
saturation effect by making the relationship between
divergence and time linear.
DIFFERENT GENES EVOLVE AT DIFFERENT RATES
 Variation in the
selective
constraints on
different genes
leads to differing
rates of molecular
evolution.
 Thus, a calibration
of the molecular
clock must be done
separately for each
gene.
NEUTRAL THEORY: SUMMARY
 The Neutral Theory provides an intentionally
simplified model of nature (drift/mutation balance),
which makes quantitative and testable predictions
about the patterns of molecular evolution in nature.
 The Neutral Theory provides a “NULL MODEL” for
statistical tests of selection at the molecular level.
 The predictions of the Neutral Theory are generally
supported by the empirical data.
NEUTRAL THEORY OF PHENOTYPIC EVOLUTION
t
 For quantitative traits the rate of increase of
between-population variance =
2 x U x average squared mutation effect
= 2Vmt
 The rate of divergence of a neutral character is equal
to twice the rate of polygenic mutation.
THE RATE OF DIVERGENCE OF SKELETAL
MORPHOLOGY IN MAMMALS TENDS TO BE WELL
BELOW THE NEUTRAL EXPECTATION
Upper and lower bounds
of the neutral rate (Vm / Ve)
Rate of increase of between-population variance
(in units of within-population variance)
FROM: M. Lynch. 1990. Am. Nat. 136:727-741.