Transcript population

Human Genetics, part II
Liisa Kauppi (Keeney lab)
Human populations: origins
Implications of population history for disease
mapping
The first demonstration of
world-wide differences in
human A, B and O allele
frequencies (1919)
Human population history
Genetic evidence is always
considered alongside linguistic,
anthropological and archeological
evidence
demic diffusion or cultural diffusion?
Two phenomena influencing gene/allele frequencies:
Founder effect
Small number of individuals settles new
area, then population grows
Bottleneck effect
Population size collapses due to e.g.
famine or epidemic
 genetic variability decreases
Additional forces influencing allele frequencies:
・Genetic drift
random effects, stronger when population size is small
・Gene flow
between neighboring groups
・Selection
For example infectious disease
Mixing or replacement?
Classical marker studies
Differences in allele frequency  genetic distances
Genetic diversity outside of Africa is a subset of diversity in Africa
Based on 120 protein-coding genes in 1,915 populations
Cavalli-Sforza & Feldman (2003) Nature Genet. 33, 266-275
Human genetic diversity is
evenly distributed
Most variation
between
populations
Most variation
within
populations
Templeton (1999) Am. J.
Anthropol. 100, 632-650
A large fraction of global human diversity
is contained within populations
Within
population
Median, all
84.6
autosomal
polymorphisms
Between
populations,
within
continent
Between
continents
5.1
9.9
AMOVA (analysis of molecular variance)
Father
Mother
Son
Courtesy of Mark Jobling
Non-recombining systems
Y chromosome “haplogroups”
“mitotypes”
1
2
3
4
Molecular clocks
Most recent common ancestor
mtDNA
Maternal - language
Y chromosome
Paternal - surname
Sociocultural factors
Patrilocality in most human populations
Polygamy
Colonizations: mostly males
Y chromosome lineages - fathers to sons
“Y chromosomal
Adam” and
“mitochondrial Eve”
were not alone!
Courtesy of Mark Jobling
Phylogenetic trees commonly
indicate a recent origin in Africa
90 (50 - 130) KYA, Hammer and Zegura
59 (40 - 140) KYA, Thomson et al.
90
69 (56 - 81) KYA, Hammer and Zegura
40 (35 - 89) KYA, Thomson et al.
80
KYA
70
60
50
40
30
20
10
0
A
B C D E F* G H
I
J K* L M N O P* Q R
Y chromosome
Y haplogroup distribution
A
B C D E F* G H
I
J K* L M N O P* Q R
Jobling & Tyler-Smith (2003) Nature Rev. Genet. 4, 598-612
An African origin
A
B C D E F* G H
I
J K* L M N O P* Q R
In Europe, there is a southeast to northwest cline in Y haplogroups
QuickTime™ and a
TIFF (LZW) decompressor
are needed to see this picture.
Gradients of allele
frequencies indicate
migration of people
Anatomically modern humans arrived in Europe via Asia 35,000 - 40,000 years ago.
Europeans are descendants of:
Paleolithic hunters and gatherers
Neolithic farmers
QuickTime™ and a
TIFF (LZW) decompressor
are needed to see this picture.
Upper Paleolithic
Late Paleolithic
Neolithic
Modern human mtDNA is distinct from
Neanderthal mtDNA
Neanderthal people lived in Europe 300,000 - 30,000 years ago
Krings et al. (1997) Cell 90, 19-30
Different genetic marker systems tell
different stories
Sample from La Plata, Argentina
45.6% native American maternal lineages
10.6% native American paternal lineages
Autosomal markers:
68% European, 26% native American, 7% West African
Martinez et al., 2004, Hum Biol 76, 543-57
More recent reshaping of diversity
a cluster of closely related lineages
• ‘Star cluster’ Y haplotype originated in/near Mongolia ~1,000 (700-1,300) years ago
• Now carried by ~8% of men in Central/East Asia, ~0.5% of men worldwide
• Suggested association with Genghis Khan (social prestige as a selective force)
Zerjal et al. (2003) Am. J. Hum. Genet. 72, 717-721
Are you a descendant of Genghis Khan?
http://www.oxfordancestors.com/genghis_khan.html
Paternal Clan certificate:
QuickT i me™ and a
T IFF (Uncompressed) decompressor
are needed to see this pi cture.
Quic kTime™ and a
TIFF (Unc ompres sed) dec ompres sor
are needed t o s ee t his pict ure.
Matriline service
Seven Daughters of Eve certificate
Lactase persistence
• All infants have high lactase enzyme activity
to digest the sugar lactose in milk
• In most humans, activity declines after
weaning, but in some it persists:
LCT*P
Population history and mapping
of genetic diseases
“Unrelated” individuals…
All humans are related if you look back far enough
…but some are more related than others
Founder effect
People are on average more related to each
other than in an “outbred” population
What’s special about isolated populations?
In a more “inbred” population, patients suffering from a
disease are more likely to share a common ancestor
More likely to have just one type of causative mutation
(no allelic heterogeneity)
In a younger population, LD blocks are longer (less
generations - less time for meiotic recombination)
Rare recessive diseases maybe much more prevalent
The first replicated ABO
association study (1954)
Mechanism: ABO blood
group binding adhesin
BabA in H. pylori
Admixture Mapping
• “Admixed” population is homogeneous but
each individual’s genome is a mosaic of
segments from different populations
• May be used to map disease loci
– multiple sclerosis susceptibility (Reich et al. 2005)
Admixture Mapping - requirements
• Disease has to show a difference in incidence
between the two “ancestral” populations, for example:
multiple sclerosis  in Africans vs. Europeans,
hypertension  in Africans vs. Europeans
• Must have polymorphic markers that differ in
frequency in the ancestral populations (HapMap
SNPs)
• Must have at least 10% admixture
Smith and O’Brien (2005) Nat Rev Genet 6, 623-632
Admixture mapping
Disease
allele must
have different
frequencies
in populations
1 and 2
Darvasi and Shifman, Nature Genetics 37, 118 - 119 (2005)
Assigning ancestry of chromosomal segments
Smith and O’Brien (2005) Nat Rev Genet 6, 623-632
Admixture Mapping
• Patient cohort of black Americans with multiple sclerosis (MS)
• MS  in Africans vs. Europeans
• Admixture: 20% European, 80% African
• Assign chromosomal segments (haplotypes) as “African” or
“European”
• Patients with MS should show an excess of “European”
chromosome segments across disease locus
Pharmacogenetics
“individualized medicine/therapy”
Optimize drug efficacy and minimize toxicity
Clinical trial: GSK3-beta gene and bipolar disorder
SNP (-50 T/C) in promoter region
Recurring episodes reduced with lithium in C/C homozygotes and C/T heterozygotes
Benedetti et al., (2005) Neurosc Lett 376, 51-55
“It is no surprise that skin pigment is a lousy surrogate for drugmetabolism status or most any aspect of human physiology.”
McLeod (2001), News and Views commentary on “Population genetic structure of
variable drug response”, Nature Genetics 29, 265 - 269
group fluidity and overlap!
14% of black American vs. 49% of White Americans benefiting
from angiotensin-converting enzyme inhibitor for heart failure
Yet - BiDil is now the first FDA approved drug for use in a specific ethnic group