Transcript Document

Molecular
clocks
1
Molecular clock?
• The molecular clock hypothesis
was put forward by
Zuckerkandl and Pauling in
1962.
• They noted that rates of amino
acid replacements in animal
hemoglobins were proportional
to time of divergence—as
judged from the fossil record.
2
3
4
5
Molecular clocks?
• Zuckerkandl and Pauling, therefore,
proposed that for any given protein,
the rate of molecular evolution is
approximately constant over time in
all lineages.
6
7
The molecular clock hypothesis
If proteins evolve at constant rates, then the
number of substitutions between two
sequences may be used to estimate
divergence times.
This is analogous to the dating of geological
times by radioactive decay.
8
Example:
The rate of nonsynonymous substitution for a-globin is
0.56  10–9 nonsynonymous substitutions per
nonsynonymous site per year.
Rat and human a-globins differ by 0.093
nonsynonymous substitutions per nonsynonymous site.
If the universal molecular-clock hypothesis is correct,
then human and rat diverged from a common ancestor
0.093/2  0.56  10 –9 = 83 million years ago.
9
Allan C. Wilson
Morris Goodman
10
The “sacrament” of the straight line
11
Q: How to draw a straight line?
A1: Have no more than two observation points.
12
Q: How to draw a straight line?
A2: With more than two observation points, use a very
thick line.
13
Q: How to draw a straight line?
A3: With more than two observation points, deny the
accuracy of the measurements on one or both axes.
14
15
Relative Rate Tests
16
Sarich & Wilson’s
Test
17
KAB = KOA + KOB
KAC = KOA + KOC
KBC = KOB + KOC
18
KOA = (KAC + KAB – KBC)/2
KOB = (KAB + KBC – KAC)/2
KOC = (KAC + KBC – KAB)/2
19
If the molecular clock hypothesis is correct, then
20
KAC – KBC = 0
Not significantly different from 0
21
22
No such difference is seen at
nonsynonymous sites, indicating that
mutational differences, rather than
selectional differences, are involved.
23
24
25
The results of the relative rate test
depend on knowledge of true tree.
26
Tests involving
duplicated genes
27
KA B
1 1
 KAA
 KOA  KOB  KBB
1
1
28
KA B
2 2
 KAA
 KOA  KOB  KBB
2
2
29
If A1 evolves at the same rate as A2, and B1 evolves at the same rate as
B2, then
KA B  KA B  0
1 1
2 2
30
31
32
33
34
A = adult; E = embryonic; F = fetal
Relative rate tests have
shown that there is no
universal molecular clock.
However, sufficiently accurate
local clocks may exist.
35
slow
fast
36
Mutation rate per site per year versus genome size
(Gago S, Elena SF, Flores R, Sanjuán R. Extremely high mutation rate of a
hammerhead viroid. 2009. Science 323:1308.)
37
The ranking of organisms started with the
Aristotelian Scala Naturae…
38
39
… and was used by
Linnaeus in his
Systema Naturae.
Primates (humans and monkeys)
Secundates (mammals)
Tertiates (all others)
40
In the literature one often encounters the
adjective “primitive” attached to the name of
an organism. For example, sponges are defined
as “primitive.”
41
Humans, on the other hand, are always
referred to as “advanced.”
42
43
Advanced
Primitive
44
Causes of variation in substitution
rates among evolutionary lineages
The factors most commonly invoked to explain the
differences in the rate of substitution among lineages are:
(1) replication-dependent factors, i.e.,
mutation.
(2) replication-independent factors, i.e.,
selection.
45
46
Generation Time
47
Rates of evolution tend to
correlate with generation
times.
48
Metabolic rate = amounts of O2
consumed per weight unit per time
unit.
49
metabolic-rate effect
mice
newts
whales
sharks
50
51
Rates of evolution tend to
correlate with metabolic
rates.
52
Generation times tend to
correlate with metabolic
rates.
The big ones are the slow
ones.
53
Organelles: Mutation Rates
Animals
nucleus
mitochondria
LOW
HIGH
Plants
mitochondria
LOW
chloroplast
nucleus
HIGH
54
Evolution of RNA viruses:
RNA VIRUSES evolve at rates that are about
106 times faster than those of DNA organisms.
Therefore, significant numbers of nucleotide
substitutions accumulate over short time
periods, and differences in nucleotide sequences
between strains isolated at relatively short time
intervals are detectable. This property allows
for a novel approach to estimating evolutionary
rates.
55
Model tree for RNA viruses:
l1 and l2 = numbers of substitutions on the
branches leading to isolates 1 and 2, respectively.
Sequence 1, which was isolated at t1, was collected
t years earlier than sequence 2, which was isolated
at t2. r = rate of substitution per site per year 56
l2 – l1 = rt2 – rt1 = rt
l2 – l1 = d23 – d13
d
23
13
r
t
d
57
58
Example:
Two strains of the HIV1 virus, denoted as 1
and 2 were isolated from a two-year-old child
on 3 October 1984 and 15 January 1985,
respectively. The child was presumed to have
been infected once perinatally by her mother
by a single strain of HIV1.
59
03 Dec. 1984
15 Jan. 1985
Reference
t = 3.4 months (0.28 year)
d13 = 0.0655
d23 = 0.0675
a = 7.1  10–3 substitutions/site/year
60
61
Tempo of Evolution:
62
63
Punctuated equilibrium (Punk eek)
Niles Eldredge & Steven J. Gould (1972).
Punctuated equilibrium: An alternative to
phyletic gradualism. pp. 82-115. In: T. J.
M. Schopf (ed.) Models in Paleobiology,
Freeman, Cooper & Co., San Francisco.
64
“... it is probable that the periods, during
which each [species] underwent
modification, though many and long as
measured by years, have been short in
comparison with the periods during
which each remained in an unchanged
condition.”
Charles Darwin, from the final 6th edition
(1872) of On the Origin of Species
65
time
Phyletic gradualism
change
66
time
Punctuated equilibria associated with speciation events
speciation
stasis
change
revolution
67
time
Punctuated equilibria disassociated from speciation events
change
68
69
During most mammalian
evolution, growth-hormones
evolved quite slowly (~0.3 
10–9 replacements per site per
A There
phylogenetic
tree two
year).
are, however,
forincreases:
the growth-hormone
rate
a 40-fold
increase
prior
to primate
gene
in mammals
divergence, and a 20-fold
increase prior to ruminant
divergence.
70
Possible explanations for
the increased rates in
ruminants and primates:
(1) an increase in mutation rate
(2) positive selection
(3) relaxation of purifying
selection
71
Rates of amino-acid replacement ( 109 standard error )
and the ratio of nonsynonymous (KA) to synonymous (KS)
substitution in gro wth-hormo ne genes during ma mmalian
evolution
_________________________________________________________
Rate of am in o-acid
Phase
repl ace men t
KA/KS
_________________________________________________________
Slow ph ase
Ru minant rapid phase
Pri mate rapid ph ase
0.3 0.1
5.6 1.4
10.8 1.3
0.03
0.30
0.49
_________________________________________________________
72
Mindell, Sykes & Graur Test
73
74
75
IS THERE A RELATIONSHIP
BETWEEN MOLECULAR
RATES OF EVOLUTION &
MORPHOLOGICAL RATES
OF EVOLUTION?
76
A living fossil:
Limulus polyphemus (Atlantic horseshoe crab)
fossil (500 mya)
extant
77
Living fossils
Blue shark (Prionace glauca)
Alligator (Alligator mississippiensis)
Molecularly fast-evolving lineages
78
Living fossils
Yellow mud turtle (Kinosternon flavescens)
Molecularly slow-evolving lineages
79
IS THERE A RELATIONSHIP
BETWEEN MOLECULAR
RATES OF EVOLUTION &
MORPHOLOGICAL RATES
OF EVOLUTION?
NO!
80
Some scientists have even
suggested that the lack of
relationship between the two
levels of description is so
total as to deserve to be
called:
“The Big Divorce”
81