■■Random genetic drift is the change in allele fre-
quencies caused by chance events of survival, re-
production, and meiosis. Unlike selection, it does
not systematically favor one allele or another.
■■drift is stronger in smaller populations, causing
faster evolution.
■■drift tends to erode genetic variation within
a population, and causes differences among
populations to accumulate.
■■Some effects of drift are best understood by
looking backward in time at the gene tree that
reflects the genealogy of the copies of a gene
that are carried by living individuals. going back-
ward in time along the gene tree, two genes
coalesce at their most recent common ancestor.
All the copies of a gene in any population, spe-
cies, or group of species ultimately trace back to
a single ancestral gene copy at some point in the
past.
■■The strength of drift is determined by the effec-
tive population size, Ne. A natural measure for the
strength of drift is 1/Ne, which determines the
size of the random fluctuations in allele frequen-
cies going forward in time. It also determines the
average time in the past of the most recent com-
mon ancestor of two gene copies now carried by
living individuals.
■■Ne is smaller (sometimes much smaller) than the
actual population size, as the result of factors that
include fluctuating population sizes and unequal
sex ratios. Population bottlenecks are short, se-
vere reductions in population size.
■■The heterozygosity at a dNA site that is evolv-ing neutrally (with no selection) is expected to
be proportional to Ne. This relation can be used
to estimate Ne from genetic data. dNA bases at
which deleterious mutations occur tend to be
less polymorphic. As a result, introns and regions
between genes are typically more variable than
the coding regions of a genome. for the same
reason, the third positions of codons are typically
more variable than the first and second positions.
■■An allele will evolve largely as if selection is not
acting when s << 1/Ne, while it will evolve largely
as if drift is not acting if s >> 1/Ne.
■■Because the relative importance of selection
and drift depends on the population size, many
of the genetic differences among species with
large Ne are expected to be adaptive, while in
species with small Ne many of the genetic dif-
ferences result from drift rather than adaptation.
drift has caused many deleterious mutations to
be fixed in humans and other species.
■■M any (but not all) genes evolve at a relatively
constant rate. A constant rate of molecular
evolution is called a molecular clock, and it can
be used to estimate the time since two species
shared a common ancestor. Constant rates are
expected when genes evolve neutrally, but rela-
tively constant rates are also seen in some genes
that are evolving by positive selection.
■■Several methods are used to detect selection
acting on dNA sequences. Recent positive selec-
tion can be detected using variation within and
differences between species. loci identified this
way are of interest because they tell us about the
molecular basis of adaptive evolution.TERMS ANd CoNCEPTS
background
selection
coalesce, coalescent
codon bias
dN/dS ratio
effective population
size (Ne)founder event
gene tree
genetic drift
genome scan
inbreeding load
local adaptationMK test
molecular clock
neutral theory
of molecular
evolution
population
bottleneckpositive selection
pseudogene
purifying selection
selective constraintSUggESTIoNS foR fURTHER REAdINg
The human brain is not well adapted to thinking
about probability, and so random genetic drift
is a difficult concept to grasp. A lucid introduc-
tion to the theory is found in J. H. gillespie’s
book Population Genetics: A Concise Guide(Johns Hopkins University Press, Baltimore,
Md, 2004). The texts suggested at the end of
Chapter 5 are also excellent on this topic.
The neutral theory of molecular evolution was
wildly controversial when it was first proposedSUMMARy
07_EVOL4E_CH07.indd 188 3/23/17 9:09 AM