Genetic and isotopic
analysis and the UK
Border Agency
Can DNA analysis really be used to screen asylum seekers by
identifying their country of origin? The immigration authority
apparently believed so, and almost put such a scheme into immediate
action – without, it seems, consulting academic scientists on the
matter. David Balding, Michael Weale, Michael Richards and
Mark Thomas examine a worrying story.
Gene and isotope
tests of nationality
have no validity,
yet were to be
used to screen
asylum seekers
58
The UK Border Agency is concerned about people
entering Britain and seeking asylum who claim to be
refugees from strife-torn countries such as Somalia,
when in fact they are opportunistic economic migrants
from nearby but relatively peaceful countries such as
Kenya. For a number of years the Agency has being
using language analysis to help confirm or refute an
individual’s claim that they come from one or other side
of an international border. Recently, it has turned to
genetic tests and isotope analysis.
A letter to Border Agency staff dated September
11th, 2009 suggests that implementation of these technologies was then well advanced. It refers to the “Human
Provenance” pilot project that was due to start three days
later, using samples that were voluntarily given by asylum
seekers. A 15-page “Nationality Swapping” instruction
document for Agency staff gave further details, and indicated that the trial would be limited to those claiming
to come from Somalia but for whom language analysis
suggested that this claim was doubtful.
Many of us are familiar with TV shows and
commercial services that claim to be able to trace an
june2010
individual’s genes back to a village in Wales, or Denmark,
or somewhere in Africa. The DNA-based ancestry
testing market worldwide is worth $26 million per year
(2008 estimate1), and growing, and is widely perceived
by the general public and media to be based on robust
science2. For example, a mitochondrial DNA (mtDNA)
match between a prehistoric skeleton and a local man
living near Cheddar, western England, has been accepted
by some historians as proving a deep genetic continuity
of the British people dating back to prehistory, thus
discounting a role for subsequent large-scale migration
events3. Stable isotope analysis was widely claimed to
have been instrumental in identifying the mutilated
torso of a boy of African origin found in the Thames
in 20014. So why not harness such apparently effective
technology in the service of the state to help distinguish
the genuine asylum seekers from the fraudulent?
Unfortunately, there is more hype than hard science to these reports. These examples can be found in
media coverage and some secondary academic literature,
but not in the primary scientific literature. As we explain
below, and as often discussed in the pages of Significance,
© 2010 The Royal Statistical Society
the underlying sources of uncertainty are
widely underestimated.
Isotopic analysis was originally developed
in archaeology to help identify immigrants
or “non-locals” in a burial population. More
recently it has become widely used in ecology
to track movements of migratory species such
as birds, and to source animal products such as
ivory. The method is based on using as a signature the ratio of two isotopes of an element,
such as strontium, lead, oxygen or hydrogen.
As animals and humans absorb these elements
from food and drink, these isotopic signatures
are more or less faithfully recorded in tissue
such as bone, teeth, hair and nails. The value of
the isotopic signature depends on the conditions in the locality from which the food and
drink were obtained. Strontium, which is one
of the main elements used for isotopic analysis
in humans, is derived from soil, and hence from
the underlying bedrock and from rainwater.
Geologies of different ages usually have different strontium isotope values, so if, for example,
you consumed bread made from wheat grown
in a field on an old geological substrate then
the strontium isotope signature obtained from
your bone or hair could be very different than
if your bread was made from wheat grown
on a young substrate. Oxygen and hydrogen
isotope signatures are mainly derived from
water and largely reflect climate, so can be used
to distinguish an animal that lived in a warm
environment when their tissues were forming
from one that lived in a cooler environment.
On the face of it, isotope signatures seem
to provide a straightforward way of inferring
the provenance of both animals and humans,
especially if multiple elements are used – lead
and oxygen as well as strontium, for example.
Unfortunately, things are not as simple as that.
The isotope signatures of strontium and other
elements can be similar over very large areas,
and even between remote parts of the world.
Thus it is only in unusual cases, such as very
old rocks from South Africa or Canada, that
we may be able to pinpoint the exact origin of
an individual based on their strontium values.
Indeed, the common practice in archaeology is to measure the local strontium isotope
signature (by sampling modern plants and
animals) and then categorise the individuals
under study simply into “local” or “non-local”.
It is rare to attempt to pinpoint the source
of the “non-locals”. Oxygen and hydrogen
isotope signatures cannot distinguish between
broadly similar climates. For example, Ireland,
western France and Portugal all have similar
rainfall oxygen isotope values. We also lack
the detailed large-scale isotopic maps of different areas of the world that would be needed
for proper inference of recent geographical
origin.
Another important issue is the length of
time that it takes for a change of geographical
location to be recorded in body tissues. The
UK Border Agency proposal is to take samples
of hair and nails from asylum seekers for
isotopic analysis. These tissues have a relatively
rapid turnover in the human body, and at best
will record the isotope signatures of food and
drink consumed during the few months before
arriving in the UK. Information from further
Is this man from Kenya or from Somalia? DNA tests are powerless to say. © iStockphoto.com/Lone Elisa
Plougmann
back in time would require bone biopsies, an
invasive and damaging procedure that would
be ethically untenable. Isotope signatures
would therefore seem of limited value as an
indicator of an asylum seeker’s real country of
origin.
What then of DNA analysis, the second
scientific strand apparently proposed by the
Border Agency? Unlike isotopes, the genotypic
variants we carry today are unchanged from
the day we were born. They are also, with few
alterations, the same variants that we inherited
from our parents, and they from theirs. This of
course is the basis for genetic ancestry testing;
but it also poses problems for the inference
of recent origins. Someone may have been
born in one country – say, Somalia – and be
a bona fide citizen of it, yet if their parents or
their recent ancestors were immigrants – say,
from Kenya – then they will carry the genetic
signatures of that more distant ancestry, not of
their current nationality.
The evidence points to the history of the
human race being a history of migration. The
ancestors of modern non-Africans began to
migrate from Africa to other parts of the world
perhaps between 60 000 and 100 000 years
ago5, and we humans have been on the move
ever since, both within and outside Africa.
Indeed, recent studies of ancient DNA6,7 and
radiocarbon dates8 suggest episodes of dramatic human migration activity over the last
10 000 years. This has led to patterns of spatial
variation that are remarkable for their uniformity, not their diversity, relative to other primate
species9. Like the pattern seen in isotopes, the
frequencies of genetic variants typically vary
only gradually with distance, making precise
localisation difficult. There can, however, be
pockets of more substantial variation, due to
small population size and genetic isolation, or
recent long-distance migration.
As with isotopes, one would need a very
extensive sampling programme to ascertain
the subtleties of the spatial pattern of genetic
variation. To do this properly would require
not only a sample from each country but also
appropriate within-country sampling to account for regional variation as well as religious,
cultural and linguistic groupings. Currently,
there is no sufficiently comprehensive sample
repository of this type. The requirement for
informed consent can hinder systematic sampling, and the available studies typically rely
on convenience samples. Furthermore, many
of the geographically specific DNA collections
that are available to academic researchers are
unlikely to be available for use by the Border
Agency or any branch of government, because
june2010
59
this would not have been covered by the
informed consent governing the use of those
collections.
Even without any study of genetic or
demographic history, it is widely recognised
that international borders, and particularly
those in Africa, often do not respect genetic
boundaries, nor have they acted as effective
barriers to migration. Many semi-nomadic or
pastoral groups straddle modern borders and
move with seasons and rainfall from one side
to the other and back. Indeed, the separation
of culturally and sometimes genetically similar
groups by an international border has often
contributed to tension and conflict. Even distinct cultures either side of a border provide no
guarantee of genetic distinctness: both theory
and observation tell us that a relatively small
level of migration, perhaps accompanying
trade or wars, can suffice to generate relative
●
(a)
SE
a)
NO
DE
SE
NO
FI
SE
LV
●
SE
NO
IE
SE
IE
GB IE IE
NL GB
DE SE
DE
IE IE IE
NL NL
IE
PL
IE
GB GB IEGB
DESE
GB IE
IE
GBIE
DE
IE
IE
GB
DE
DE
GB
SE
IE IE
GB
GB
DE
GB
DE NL
GB
SctGB
GB
GB
GB
IE
IE
PL
GB
GB
IEIE
GBGBGBGB
GBIEGB
IE GBIE
IE IE
NL
IE
IEIE
GB
GB
IE
GB
PL
GB
IEGB
RU RU
GB
GBGB
GB
DE
GB
IE IEGB IE IEIE GB Sct
PL PL
GB SE DE
DE
DE
GB
GB
GB
IE
IE NL
GBGBGBGB
DE
PL
IEIE IE
GB
GB IE
GB
GB
GB
GBGB
GBGBDE
GB
GB
Sct
GBIE
GB
RUPL
GBNL
PL
GB
IE IE IE GB
DE DE
PL
GB
NL
GB
DE
NL
GB GB
GB
GB
GB
GB
GB
PL
GB
GB
GB
GB
GB
GB
IE
DK
GB
IE
GB
NL
GB
PL
GB
GB
RU
GB GB GB
DE
IE
GB
DE
DE
DE
IE
GB
GB
GB
GB
GBIE GB
GB
IE
GBGBGB
GB
GB
NL
DE
GBGB
NL
GB
PL
GB
IE GB
GB
GB GB
GBGB GB
GB
DE DE
GB
NL
PL
GB
GB
PL
GBIE
IE GB
GB
GB
GBGB
GB
GB
DE
Sct
GB
GB
GB
GBGB
GB
GB
PL
GB
GB
GB
BE
GB
PL
GB
IE
BE
GB GB GBFR GB GB
GB
GB
GB
BE GB
GB
GB
IE DE
GB
GB GB
GB GBGB
RU
BE
Sct
DE DE DE
BE DE
GB
GB
BE
DE
GB
GB
GB
DE
DE
GB
GB
FR
CZ
GB
GB BE GB NL BEBE
GB NL
GB
DE
GBGB
GB
GB
GB GBGBGB GB
DE
CZ
DE DE
DE
IE GBGB GB GB FR
PL
GB
DE
GB
BE
GB
PL
CZ
NL BE
PL
CZ
BE
AT
GBBEBE
GB
DE DE
DE DE
CZ AT PL
CH−GGB
BE DE
CZ
CH−G
GBDENL
BE
AT PL
DE
CH−G
BE
DE
CZ
DE
BE
BE
BE FR
FR
CH−G
HU CZ
DE
BE
ATHU
BE
BE
HR
CH−G
CH−G
CH−G
CH−F
HU
BE
BE
CH−GCH−G
BECH−G
FRGB BE
HU
FR
ATDE
BE
DE AT
BE
CH−G
DE
FR BE
HU
BE
FR
CH−G
CH−G
GB
BE
BE BE
CH−G
BE
BEFR
CH−G DE
HU
AT HU
BE
DE
CH−G
BE FR
CH−F
CH−F
CH−G
CH−G
CH−G
CH−F
CH−G
BE
CH−G
GB
DE
CH−G
CH−F
CH−F
CH−G
FRFR
HU
DE
DE
CH−F
FR
AT
UA
HU CZ
CH−F
AT
AT
FR
CH−F
CH−G
CH−F
CH−G
PL
CH−G
FR
HU
CH−G
DE
CH−G
DE
CH−G
CH−F
DE
CH−F
CH−G
FR
CH−G
CH−F
FR CH−G
CH−G
HU
FR FR FR FRFR
CH−F
CH−G
CH−F
CH−F
CH−G
CH−G
GB CZ
HU HU SI
CH−F
CH−G
CH−F
CH−F
CH−F
CH−G
CH−F
CH−F
CH−F
FR
FR
CH−G
DE
CH−I
CH−F
CH−F
CH−F
CH−G
DE
CH−F
CH−G
DE CH−G
HU
CH−G
DE
FR BE
AT
CH−G
CH−G
CH−G
HR
CH−F
CH−F
BA
CH−F
CH−F
CH−G
CH−G
CH−F
FR
CH−G
CH−G
BE
DE
CH−F
IT
CH−F
CH−G
CH−G
FR
CH−G
CH−G
CH−G
CH−F
CH−G
CH−F
FR
CH−F
HU
FRCH−F
CH−F
CH−F
CH−G
CH−F
CH−F
CH−F
FR
CH−F
HR
FR
HR
CH−F
CH−F
CH−G
CH−G
CH−F
CH−F
CH−G
CH−F
DE
CH−F
FRCH−F
CH−F
CH−F
CH−G
CH−F
CH−F
CH−F
CH−F
BE
CH−F
BE
CH−F
CH−F
CH−G
CH−F
RO
CH−F
CH−G
FR FR
CH−F
CH−G
FR
CH−F
DE
CH−G
CH−F
CH−G
CH−F
FR
CH−F
AT
CH−G
CH−F
CH−F
CH−G GB
CH−F
CH−G
CH−G
FR
CH−F
CH−F
CH−F
CH−F
CH−F
CH−G
CH−F
FR
CH−F
CH−G
CH−G
CH−F
CH−F
CH−F
CH−F
CH−G
CH−F
HR
CH−G
CH−F
CH−F
CH−G
FRFR CH−F
FR
FR
CH−F
CH−GAT DE
FRFR
CH−F
SIHR
CH−F
FR
CH−F
CH−G
FRCH−F
CH−F
RO
DE
AT CH−F
CH−G
CH−F
FR
CH−F
FR
CH−F
FRCH−F
CH−F
CH−F
CH−F
CH−F
FRFR
RS
FR
FR
CH−F
CH−F
CH−F
FR CH−F
FR CH−F
CH−F
FR
CH−F
HR
CH−F
BABA
CH−F
CH−G
YGYG YG
FRIT CH−F
YG
CH−F
CH−F
CH−F
FRCH−F
BA
CH−F
BA
CH−G
CH−F
YG
FR FR
CH−F
IT
IT
IT
CH−F
YG BA
HU YG
CH−GRU
CH−I
YG
CH−F
CZ
FR FR
CH−F
BA
FR CH−F
FR
ES
RSYGBA YG
FR
FR FR FR
BA RS
IT IT
YG
RO
FR
FR
RO YG
FR
HU
YG YG
FR
IT IT
FR
IT
IT IT FR
IT
FRFR
FR
YG
CH−I
FR
RO
MK HU HU YG
IT
CH−I
FR
YG
FR
RO
FR
IT
FR
CH−I ITITITCH−G
RO YG YG
RO
FR
ES
RO
BGRO
CH−I
ITITIT
FR
ES
IT
ITCH−I
IT
FR
ROYG RO
ITIT ITIT
ES
IT
CY
YG
CH−I
ITIT
ITCH−I
CY CY
ITITIT
MK
IT
BG
IT
IT
IT
IT
IT
IT
ES ES
ES
RO
ALMK
IT IT IT
IT
CH−I
ITIT
IT
ITIT ITITIT
ITITIT
IT
IT ITCH−I
ES ESES
ES
ES
YG YG
MK
IT
IT
YG
IT
IT
IT ITIT
GBIT
YG
FR
IT
ES
YGYGYG
IT
ITIT
IT
FR
ITIT
PT
PT ES
GR
CH−I
YG
IT
IT
ES
ITITIT
PT PT
IT
ES
ES ES
IT
IT
IT
IT
IT
ITIT
IT
IT
IT
AL
PT
IT
IT
ITIT
IT
IT
ITIT IT
IT
IT
GR
IT
ITIT
SK
IT
IT
IT
IT
IT
IT
IT
IT
IT
ITIT
IT
IT
GR
IT
IT
ESES
IT
IT
IT
YGYG
IT
IT
IT
IT
IT
IT
IT
YG YG
IT
ITGR
IT
CH−I
IT
IT
GB
ESES
IT
IT
IT GR GR
IT
IT
IT
IT
IT
YG
IT
ESPTES
PT PT ES FR
IT
IT
YG
IT
IT
DE
YGAL
YG
YG
YG
IT
IT
GR
IT
IT
IT
ITIT
IT
KS
GR
ROYG
PT
PT
YG
ES
ES
HR
IT
ES
IT
ES
ES
PT
PT
ES
IT IT
KS
ES ES
ES
PT PT
PT
ES
ES
ES
ES
ESESES
PT
TR
FR
ES PT
PT
ES
ES
ES
PT
ES
PT
ESPT
ES
PT
ES
ES
ES
PT
ES
ES
ES
ES ES
ES
ES
ES
PT
TR
ES
ES
PT
PT ESPT
PT
PT
PT
ESES
ES
PT
ES
ES
PT
PT
ES
PT
ES
PT
ES
PT
ES
PT
PT
PT
ES
ES
ES
ES
ES
PT
PT
PT
ES
PT
PT
ESESESES
ES PT
ES
PT
TR
PT
ES
PT
ES
PT
PT
PT
PT
PT
PT
ES
PT
PT
ES
ES
PT
ES
ES
ES
PT
ESES
PT
ES
PT
PT
ES
ES
ES
PT
PT
ES
ES
PT
PT
PT
ES ES
ESPT
PT
PT
PT
ES
PT
ES
PT
PT
PT
ES
PT
PT
ES
ES
PT
PT
PTPT
ES PT ES
PT
PT
ES
ES
ES
ES
PT
ES
PT
PT
ESES
ES
PT
PT
ES
PT
PT
PT
ES
PT
ESES
ES
ES
ES
PT
ES
PTES PT
PT
PT
PT
PT
PT
PT
ES
TR
PT
PT
PT
ES
ES
PT
PT ES PT PT
PT
ES
PT
PT
IT
PT
PT PT
PT
PT PT
PT PT
ITIT
PT
PT
ES
IT
IE
IE
SE
●
●
●
●
●
●
●
●
●
●
400 km
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
IT
b)
(b)
1.0
Prediction Accuracy 0.8
1200-2500 km
0.6
800-1200 km
400-800 km
0.4
0-400 km
0.2
Austria [AT]
Belgium [BE]
Bosnia-Herzegovina [BA]
Croatia [HR]
Czech Republic [CZ]
France [FR]
Germany [DE]
Greece [GR]
Hungary [HU]
Ireland [IE]
Italy [IT]
Netherlands [NL]
Poland [PL]
Portugal [PT]
Romania [RO]
Spain [ES]
Sweden [SE]
Swiss-French [CH-F]
Swiss-German [CH-G]
Swiss-Italian [CH-I]
United Kingdom [GB]
Yugoslavia [YG]
AVERAGE
0.0
Figure 1. (a) Predicted locations for each of 1,387 European individuals based on genotypes at over half
a million DNA sites. The codes indicate the stated country of origin of the individuals’ four grandparents
(individuals with grandparents of mixed national origin or of non-European origin were excluded; the country of
the individual’s birth was used if the grandparental information was unavailable). The filled circles use the same
colour scheme and indicate the locations used to train the assignment method. (b) Distribution of prediction
accuracy by country. Reproduced by permission from Macmillan Publishers Ltd: Novembre, J. et al., Genes mirror
geography within Europe, Nature, 456, 98–101. (Copyright 2008)14
60
june2010
genetic homogeneity between culturally distinct neighbouring groups10.
Genetic data does have one advantage over
isotopic data: while there are only a handful of
elements available for isotope analysis, there are
millions of variable sites in the human genome.
If a large proportion of these were analysed,
this would help to average out the uncertainties
introduced at a single locus by “genetic drift”:
the random walk in the frequency of a genetic
variant over space and time that is brought
about by the stochastic nature of reproduction.
From the documents they released, it appears
that at least initially the Agency proposed to
use only mtDNA and Y chromosome markers. The former trace only the maternal lineage
(your mother, your mother’s mother, your
mother’s mother’s mother and so on); the latter
trace only the paternal line. To appreciate the
loss of information arising from this, note that
your paternal and maternal lineages include
neither your mother’s father nor your father’s
mother, and at every generation further back
in time the proportion of ancestors in that
generation who are included in these lineages
approximately halves. You have up to 16 greatgreat-grandparents, but these two techniques
between them will provide information from
just two.
At a single locus, migration and genetic
drift means that the same mtDNA or Y chromosome types might be found at appreciable
frequencies in Norway and India, in Mongolia
and Turkey, in Africa and Yorkshire11. Furthermore, the frequency distributions within these
ranges are “lumpy”, with interspersed regions of
high frequency. For example, a Y chromosome
haplotype that is common in North Wales, but
rare or absent in the rest of the UK and indeed
most of Europe, occurs at appreciable frequency in the eastern Mediterranean12. Could
this have resulted from a male migrant from
the Mediterranean who arrived in North Wales
and pleased at least one and perhaps several of
the local ladies, and thereby passed on many of
his lady-pleasing genes to subsequent generations? Careful use of these genetic systems can
allow researchers to make inferences about past
demographic events at the population level,
but, for the reasons outlined above, their use in
most forms of individual ancestry testing can be
considered as little more than genetic astrology.
Neither a Y chromosome test nor an mtDNA
test will allow us to say with confidence that
somebody came from, for example, Somalia
instead of Kenya.
Recent studies have shown the potential
for genome-wide data to be more precise than
mtDNA or Y chromosome data in predicting
geographic origin13–15. These studies have used
multivariate statistical methods, typically principal components analysis, to map European
individuals to their country of origin based on
their DNA data at hundreds of thousands of
genetic markers. Most individuals can be assigned to within several hundred kilometres of
the country where they were actually sampled,
or which they gave as the origin of their immediate ancestors (see Figure 1). These impressive
results point to the potential of modern genetic
systems to infer relatedness of individuals, and
hence population structure, which in turn is
indicative of demographic history. However,
it is important to note that these studies were
not based on random samples of individuals
but on convenience samples, and some of the
samples were pre-filtered to remove individuals
of mixed ancestry. While genome-wide genetic
data does hold much promise for identifying
genetic relationships among groups, the underlying problems remain concerning incomplete
sampling and how to deal with individuals
resulting from recent migrations and mixed
ancestry. Even complete knowledge of an individual’s genes cannot tell you with certainty his
or her country of birth.
Why did the UK Border Agency apparently fail to grasp these points? We believe
that intense public and media interest has
led to a distorted view of the field of human
genetic history. DNA-based ancestry testing
companies and organisations make claims
about individual ancestries, or indeed the
nature of human ancestry in general, that are
misleading and misrepresent the underlying
uncertainties. By telling somebody that they
“come from” some particular ethnic, racial or
national group, they also tend to reinforce the
notion that these categories have long-term
genetic validity, when in fact each of us has
many ancestral lineages, and the prevalence
of migration means that they are likely to tell
many different stories. DNA tests, and claims
based on them, are sometimes used in documentaries to tell stories about the origins of
specific individuals. This panders to the public
desire for a good story but also to outdated
notions of race, ethnicity and ancestry, while
scholars who point out flaws and limitations
of the methods employed are usually ignored.
In the milieu of seemingly contradictory and
confusing facts arising from human population
genetics, it is perhaps unsurprising that people
and organisations have exploited misunderstandings for their own gain. But we believe
that with the UK Border Agency’s “Human
Provenance” pilot project, the consequences
of the bad science of individual DNA-based
ancestry testing could go beyond the telling
of unverifiable stories, such as that a client is
the descendant of a Celtic princess, or a Viking
warlord, or a Zulu chief.
A worrying feature of the Border Agency’s
proposals was the apparent lack of engagement
with the scientific research community. From
the letter referred to above it seems clear that
implementation of genetic and isotope analysis
was imminent, without any indication of trials
or mention of underpinning research. We are
aware of no university researchers who were
consulted, and none have subsequently come
forward.
The bad science of DNA
ancestry tests has gone far
beyond unverifiable stories
that a client is descended from
a Celtic princess
Thanks to the recognition of the dangers
by the journal Science16, the Home Office has
moved quickly to distance itself from the implications of its letter, saying that while this trial
is being undertaken, “no decisions on individual
cases will be made using these techniques, and
they will not be used for evidential purposes”,
and that “Only after review by the Forensic Science Regulator will the techniques be considered for use in asylum investigations”17.
Of course, isotope and DNA analysis
can convey information about an individual’s
origin. The problem is whether it is possible to
evaluate this information in a way that is useful
to Border Agency staff and fair to asylum seekers. In view of the problems outlined above,
we do not think this is achievable now or in
the near future. Although the draft guidelines
stated that case managers “must not rely solely
on the isotope and DNA test results” and that
“The results from the language analysis must
be afforded appropriate weight in relation to …
the isotope and DNA results”, no guidance was
given as to what was “appropriate” – nor does
it seem apparent to us what guidance could
usefully be given.
References
1. h t t p : / / w w w . t h e g e n e t i c g e nealogist.com/2007/11/07/thegenetic-genealogy-market-part-ii/
2. Bolnick, D. A., Fullwiley, D., Duster, T.,
et al. (2007) The science and business of genetic
ancestry testing. Science, 318, 399–400.
3. Davies, N. (1999) The Isles – A History.
Oxford: Oxford University Press, p. 4.
4. Tremlett G. (2003) Tracing Adam. The
Guardian, August 7. http://www.guardian.co.uk/science/2003/aug/07/
forensicscience
5. Garrigan, D. and Hammer, M. F. (2006)
Reconstructing human origins in the genomic era.
Nature Review Genetics, 7, 669–680.
6. Hellenthal, G., Auton, A., and Falush, D.
(2008) Inferring human colonization history using
a copying model. PLoS Genetics, 4, e1000078.
7. Bramanti, B., Thomas, M.G., Haak, W.,
et al. (2009) Genetic discontinuity between local
hunter gatherers and Central Europe’s first farmers.
Science, 326, 137–140.
8. Collard, M., Edinborough, K., Shennan, S.,
and Thomas, M.G. (2010) Radiocarbon evidence
indicates that migrants introduced farming to Britain. Journal of Archaeological Science, 37, 866–870.
9. Becquet, C., Patterson, N., Stone, A. C.,
Przeworski, M., and Reich, D. (2007) Genetic
structure of chimpanzee populations. PLoS Genetics, 3, e66.
10. Tishkoff, S., Reed, F. A., Friedlaender, F.
R., et al. (2009) The genetic structure and history
of Africans and African Americans. Science, 324,
1035–1044.
11. King, T. E., Parkin, E. J., Swinfield, G.,
et al. (2007). Africans in Yorkshire? The deepestrooting clade of the Y phylogeny within an English
genealogy. European Journal of Human Genetics, 15,
288–293
12. Weale, M.E., Weiss, D.A., Jager R.F., Bradman, N., and Thomas M.G. (2002) Y chromosome
evidence for Anglo-Saxon mass migration. Molecular Biology and Evolution 19(7): 1008–1021.
13. Lao, O., Lu, T., Nothnagel, M., et al. (2008)
Correlation between genetic and geographic structure in Europe. Current Biology, 18, 1241–1248.
14. Novembre, J. Johnson, T., Bryc, K., et al.
(2008) Genes mirror geography within Europe,
Nature, 456, 98–101.
15. Wang, C., Szpiech, Z. A., Degnan, J. H.,
et al. (2010) Comparing spatial maps of human
population-genetic variation using Procrustes
analysis. Statistical Applications in Genetics and
Molecular Biology, 9, Article 13.
16. Travis, J. (2009) Scientists decry isotope,
DNA testing of “Nationality”. Science, 326, 30–31.
17. http://news.sciencemag.org/
scienceinsider/2009/10/uk-backing-away.html
David J. Balding is at University College London Genetics Institute. Michael Weale is at the Department
of Medical and Molecular Genetics, Kings College London. Michael Richards is at the Department of Anthropology, University of British Columbia and at the
Max Planck Institute for Evolutionary Anthropology,
Leipzig; and Mark G. Thomas is at the Evolutionary
Biology Centre, Uppsala University, and the Research
Department of Genetics, Evolution and Environment
at University College London.
june2010
61