ArticlePDF Available

Genetic structure of the purebred domestic domestic dog


Abstract and Figures

We used molecular markers to study genetic relationships in a diverse collection of 85 domestic dog breeds. Differences among breeds accounted for approximately 30% of genetic variation. Microsatellite genotypes were used to correctly assign 99% of individual dogs to breeds. Phylogenetic analysis separated several breeds with ancient origins from the remaining breeds with modern European origins. We identified four genetic clusters, which predominantly contained breeds with similar geographic origin, morphology, or role in human activities. These results provide a genetic classification of dog breeds and will aid studies of the genetics of phenotypic breed differences.
Content may be subject to copyright.
DOI: 10.1126/science.1097406
, 1160 (2004); 304Science
et al.Heidi G. Parker,
Genetic Structure of the Purebred Domestic Dog (this information is current as of December 5, 2008 ):
The following resources related to this article are available online at
version of this article at:
including high-resolution figures, can be found in the onlineUpdated information and services,
can be found at: Supporting Online Material
found at:
can berelated to this articleA list of selected additional articles on the Science Web sites
, 14 of which can be accessed for free: cites 23 articlesThis article
105 article(s) on the ISI Web of Science. cited byThis article has been
24 articles hosted by HighWire Press; see: cited byThis article has been
: subject collectionsThis article appears in the following
in whole or in part can be found at: this article
permission to reproduce of this article or about obtaining reprintsInformation about obtaining
registered trademark of AAAS.
is aScience2004 by the American Association for the Advancement of Science; all rights reserved. The title
CopyrightAmerican Association for the Advancement of Science, 1200 New York Avenue NW, Washington, DC 20005.
(print ISSN 0036-8075; online ISSN 1095-9203) is published weekly, except the last week in December, by theScience
on December 5, 2008 www.sciencemag.orgDownloaded from
cotransfected with wild-type and mutant
PINK1 cDNA and a green fluorescent protein
(GFP) reporter plasmid and then stressed
with the peptide aldehyde Cbz-leu-leu-
leucinal (MG-132), which inhibits the protea-
some and induces apoptosis via distinct
mechanisms, including mitochondrial injury
(8). Analysis of TMRM fluorescence in GFP-
positive cells revealed that the PINK1 mu-
tation had no significant effect on ⌬␺m
under basal conditions (Fig. 3A). However,
after stress with MG-132, there was a sig-
nificant decrease in ⌬␺m from basal levels
in cells transfected with G309D PINK1 as
compared with wild-type PINK1 (G309D,
44.1% 8.1; versus wild-type PINK1,
13.0% 13.7; P 0.01; n 8 data sets)
(Fig. 3B and fig. S4).
We next studied apoptosis of MG-132–
stressed SH-SY5Y cells transfected with
either wild-type or G309D PINK1 by FACS
using annexin V conjugated to the fluoro-
chrome phycoerythrin (annexin V-PE). An-
nexin V has a high binding affinity for the
membrane phospholipid, phosphatidylser-
ine, that is exposed on the surface of apo-
ptotic cells. Consistent with the changes in
⌬␺m after stress, overexpression of wild-
type PINK1 but not mutant PINK1 signifi-
cantly reduced the level of apoptotic cell
death induced by MG-132 in GFP-positive
cells [vector, 45.4% 5.0; wild-type
PINK1, 32.7% 4.0; G309D, 45.8%
5.0; P 0.05; analysis of variance
(ANOVA), n 12 data sets] (Fig. 3C and
fig. S5). These preliminary findings sug-
gest that wild type PINK1 may protect
neurons from stress-induced mitochondrial
dysfunction and stress-induced apoptosis
and that this effect is abrogated by the
G309D mutation.
Several lines of evidence suggest that im-
pairment of mitochondrial activity could rep-
resent an early critical event in the pathogen-
esis of sporadic PD (2). Environmental toxins
such as 1-methyl-4-phenyl-1,2,3,6-tetrahy-
dro-pyridin and the pesticide rotenone induce
selective death of dopaminergic neurons
through inhibition of complex I activity (9,
10). Complex I deficiency and a variety of
markers of oxidative stress have been dem-
onstrated in postmortem brains of PD patients
(2, 11, 12). In addition, several reports have
shown that mitochondrial dysfunction associ-
ated with oxidative stress can trigger
-synuclein aggregation and accumulation,
although the exact mechanisms remain un-
clear (13).
The PINK1 mutations described here oc-
cur in the putative serine/threonine kinase
domain and thus conceivably could affect
kinase activity or substrate recognition. Al-
tered phosphorylation has been reported as a
pathogenetic mechanism in other neurode-
generative diseases, including Alzheimer’s
disease, tauopathy, and spinocerebellar ataxia
(14, 15 ). The recent demonstration that phos-
phorylation of -synuclein at serine 129 oc-
curs in Lewy bodies in a variety of brains
from humans with synucleinopathy (16) sug-
gests that altered phosphorylation may also
play a role in PD. We hypothesize that
PINK1 may phosphorylate mitochondrial
proteins in response to cellular stress, pro-
tecting against mitochondrial dysfunction.
PINK1 was originally shown to be up-
regulated by the tumor suppressor gene
PTEN in cancer cells (6 ). In neurons, the
PTEN signaling pathway is involved in cell
cycle regulation and cell migration and pro-
motes excitotoxin-induced apoptosis in the
hippocampus (17 ). However, PINK1 has
not been shown to have any effects on
PTEN-dependent cell phenotypes (6), and
its role in the PTEN pathway therefore
requires further investigation.
References and Notes
1. W. Dauer, S. Przedborski, Neuron 39, 889 (2003).
2. T. M. Dawson, V. L. Dawson, Science 302, 819 (2003).
3. E. M. Valente et al., Am. J. Hum. Genet. 68, 895
4. E. M. Valente et al., Ann. Neurol. 51, 14 (2002).
5. Single-letter abbreviations for the amino acid resi-
dues are as follows: A, Ala; D, Asp; G, Gly; W, Trp.
6. M. Unoki, Y. Nakamura, Oncogene 20, 4457 (2001).
7. M. R. Duchen, A. Surin, J. Jacobson, Methods Enzymol.
361, 353 (2003).
8. J. H. Qiu et al., J. Neurosci. 20, 259 (2000).
9. W. J. Nicklas, I. Vyas, R. E. Heikkila, Life Sci. 36, 2503 (1985).
10. R. Betarbet et al., Nature Neurosci. 3, 1301 (2000).
11. A. H. Schapira et al., Lancet 2, 1269 (1989).
12. P. Jenner, C. W. Olanow, Ann. Neurol. 44, S72 (1998).
13. T. B. Sherer et al., J. Neurosci. 22, 7006 (2002).
14. L. Buee, T. Bussiere, V. Buee-Scherrer, A. Delacourte,
P. R. Hof, Brain Res. Rev. 33, 95 (2000).
15. HK. Chen et al., Cell 113, 457 (2003).
16. H. Fujiwara et al., Nature Cell Biol. 4, 160 (2002).
17. D. S. Gary, M. P. Mattson, Neuromol. Med. 2, 261 (2002).
18. We thank the patients and families who participated
in this study, J. Sinclair for technical assistance with
FACS experiments, G. Howell for bioinformatic sup-
port, M. Duchen for useful discussion, S. Eaton for
assistance with mitochondrial fractionation, and Y.
Nakamura and M. Unoki for the PINK1 plasmid. Sup-
ported by grants from Telethon, Italy (E.M.V.); the
Italian Ministry of Health (E.M.V. and B.D.); MURST
(B.D.); the Parkinson’s Disease Society, UK (N.W.W.,
D.S.L., R.J.H., and D.G.H.); the Brain Research Trust
(P.M.A.S. and N.W.W.); and the Deutsche Forschungs-
gemeinschaft (G.A. and S.G.). M.M.K.M. is a Medical
Research Council Clinical Research Training Fellow.
GenBank accession numbers are as follows: PINK1
genomic sequence, AL391357; PINK1 mRNAs,
AB053323, AF316873, AK075225, BC009534, and
BC028215; and PINK1 protein, BAB55647,
AAK28062, BAC11484, AAH09534, and AAH28215.
Supporting Online Material
Materials and Methods
SOM Text
Figs. S1 to S5
Tables S1 and S2
2 February 2004; accepted 1 April 2004
Published online 15 April 2004;
Include this information when citing this paper.
Genetic Structure of the
Purebred Domestic Dog
Heidi G. Parker,
Lisa V. Kim,
Nathan B. Sutter,
Scott Carlson,
Travis D. Lorentzen,
Tiffany B. Malek,
Gary S. Johnson,
Hawkins B. DeFrance,
Elaine A. Ostrander,
* Leonid Kruglyak
We used molecular markers to study genetic relationships in a diverse collection
of 85 domestic dog breeds. Differences among breeds accounted for 30% of
genetic variation. Microsatellite genotypes were used to correctly assign 99%
of individual dogs to breeds. Phylogenetic analysis separated several breeds
with ancient origins from the remaining breeds with modern European origins.
We identified four genetic clusters, which predominantly contained breeds with
similar geographic origin, morphology, or role in human activities. These results
provide a genetic classification of dog breeds and will aid studies of the genetics
of phenotypic breed differences.
The domestic dog is a genetic enterprise unique
in human history. No other mammal has en-
joyed such a close association with humans
over so many centuries, nor been so substan-
tially shaped as a result. A variety of dog mor-
phologies have existed for millennia, and repro-
ductive isolation between them was formalized
with the advent of breed clubs and breed stan-
dards in the mid–19th century. Since that time,
the promulgation of the “breed barrier” rule—
no dog may become a registered member of a
breed unless both its dam and sire are registered
members— has ensured a relatively closed ge-
netic pool among dogs of each breed. At
present, there are more than 400 described
breeds, 152 of which are recognized by the
American Kennel Club (AKC) in the United
States (1). Over 350 inherited disorders have
been described in the purebred dog population
(2). Many of these mimic common human dis-
orders and are restricted to particular breeds or
groups of breeds as a result of aggressive in-
21 MAY 2004 VOL 304 SCIENCE www.sciencemag.org1160
on December 5, 2008 www.sciencemag.orgDownloaded from
breeding programs used to generate specific
morphologies. We have previously argued that
mapping genes associated with common dis-
eases, including cancer, heart disease, epilepsy,
blindness, and deafness, as well as genes un-
derlying the striking diversity among breeds in
morphology and behavior, will best be accom-
plished through elucidating and taking advan-
tage of the population structure of modern
breeds (3). Understanding the genetic relation-
ships among breeds will also provide insight
into the directed evolution of our closest animal
Mitochondrial DNA analyses have been
used to elucidate the relationship between the
domestic dog and the wolf (46 ), but the
evolution of mitochondrial DNA is too slow
to allow inferences about relationships
among modern dog breeds, most of which
have existed for fewer than 400 years (1, 7,
8). One previous study showed that nuclear
microsatellite loci could be used to assign
dogs from five breeds to their breed of origin,
demonstrating large genetic distances among
these breeds (9). Another study used micro-
satellites to detect the relatedness of two
breed pairs in a collection of 28 breeds but
could not establish broader phylogenetic re-
lationships among the breeds (10). The fail-
ure to find such relationships could reflect the
properties of microsatellite loci (10), the lim-
ited number of breeds examined, or the ana-
lytical methods used in the study. Alterna-
tively, it may reflect the complex structure in
purebred dog populations, resulting from the
recent origin of most breeds and the mixing
of ancestral types in their creation. Here, we
show that microsatellite typing of a diverse
collection of 85 breeds, combined with phy-
logenetic analysis and modern genetic clus-
tering methods (11, 12), allows the definition
of related groups of breeds and that genetic
relatedness among breeds often correlates
with morphological similarity and shared
geographic origin.
To assess the amount of sequence varia-
tion in purebred dogs, we first resequenced
19,867 base pairs of noncontiguous genomic
sequence in 120 dogs representing 60 breeds.
We identified 75 single nucleotide polymor-
phisms (SNPs), with minor allele frequencies
ranging from 0.4 to 48% (table S1). Fourteen
of the SNPs were breed specific. When all
dogs were considered as a single population,
the observed nucleotide heterozygosity (13)
was 8 10
, essentially the same as that
found for the human population (14, 15).
To further characterize genetic variation
within and among breeds, we genotyped 96
microsatellite loci in 414 purebred dogs rep-
resenting 85 breeds (five unrelated dogs that
lacked any common grandparents were sam-
pled from most breeds; table S2). We predict-
ed that, because of the existence of breed
barriers, dogs from the same breed would be
more similar genetically than dogs from dif-
ferent breeds. To test this prediction, we es-
timated the proportion of genetic variation
among individual dogs that could be attribut-
ed to breed membership. An analysis of mo-
lecular variance (16 ) in the microsatellite
data showed that variation among breeds ac-
counts for more than 27% of total genetic
variation. Similarly, the average genetic dis-
tance between breeds calculated from the
SNP data is F
0.33. These observations
are consistent with previous reports that an-
alyzed fewer dog breeds (9, 10), confirming
the prediction that breed barriers have led to
strong genetic isolation among breeds, and
are in marked contrast to the much lower
genetic differentiation (typically in the range
of 5 to 10%) found among human popula-
tions (17, 18). Variation among breeds in
dogs is on the high end of the range reported
for domestic livestock populations (19, 20).
Strong genetic differentiation among dog
breeds suggests that breed membership could
be determined from individual dog genotypes
(9). To test this hypothesis, we first applied a
Bayesian model based clustering algorithm,
implemented in the program structure (11, 12,
21), to the microsatellite data. The algorithm
attempts to identify genetically distinct sub-
populations on the basis of patterns of allele
frequencies. We applied structure to overlap-
ping subsets of 20 to 22 breeds at a time (22)
and observed that most breeds formed distinct
clusters consisting solely of all the dogs from
that breed (Fig. 1A). Dogs in only four breeds
failed to consistently cluster with others of the
same breed: Perro de Presa Canario, German
Shorthaired Pointer, Australian Shepherd, and
Chihuahua. In addition, six pairs of breeds clus-
tered together in the majority of runs. These
pairingsAlaskan Malamute and Siberian
Husky, Belgian Sheepdog and Belgian Ter-
vuren, Collie and Shetland Sheepdog, Grey-
hound and Whippet, Bernese Mountain Dog
and Greater Swiss Mountain Dog, and Bull-
mastiff and Mastiffare all expected on the
basis of known breed history. To test whether
these closely related breed pairs were nonethe-
less genetically distinct, we applied structure to
each of these clusters. In all but one case, the
clusters separated into two populations corre-
sponding to the individual breeds (Fig. 1B). The
single exception was the cluster containing Bel-
gian Sheepdogs and Belgian Tervurens. The
European and Japanese Kennel Clubs classify
these as coat color and length varieties of a
single breed (23, 24 ), and although the AKC
recognizes them as distinct breeds, the breed
barrier is apparently too recent or insufficiently
strict to have resulted in genetic differentiation.
We next examined whether a dog could be
assigned to its breed on the basis of genotype
data alone. Using the direct assignment method
(25) with a leave-one-out analysis, we were
able to assign 99% of individual dogs to the
correct breed. Only 4 dogs out of 414 were
assigned incorrectly: one Beagle as a Perro de
Presa Canario, one Chihuahua as a Cairn Ter-
rier, and two German Shorthaired Pointers as a
Kuvasz and a Standard Poodle. All four errors
involved breeds that did not form single-breed
clusters in the structure analysis.
Having demonstrated that modern dog
breeds are distinct genetic units, we next
sought to define broader genetic relationships
among the breeds. We first used standard
neighbor-joining methods to build a majority-
rule consensus tree of breeds (Fig. 2), with
distances calculated using the chord distance
measure (26 ), which does not assume a par-
ticular mutation model and is thought to per-
form well for closely related taxa (27 ). The
tree was rooted using wolf samples. The
deepest split in the tree separated four Asian
spitz-type breeds, and within this branch the
Shar-Pei split first, followed by the Shiba Inu,
with the Akita and Chow Chow grouping
together. The second split separated the
Basenji, an ancient African breed. The third
split separated two Arctic spitz-type breeds,
the Alaskan Malamute and Siberian Husky,
and the fourth split separated two Middle
Eastern sight hounds, the Afghan and Saluki,
from the remaining breeds.
The first four splits exceeded the majority-
rule criterion, appearing in more than half of the
bootstrap replicates. In contrast, the remaining
breeds showed few consistent phylogenetic rela-
tionships, except for close groupings of five
breed pairs that also clustered together in the
structure analysis, one new pairing of the closely
related West Highland White Terrier and Cairn
Terrier, and the significant grouping of three
Asian companion breeds of similar appearance,
the Lhasa Apso, Shih Tzu, and Pekingese (fig.
S1). A close relationship among these three
breeds was also observed in the structure anal-
ysis, with at least two of the three clustering
together in a majority of runs. The flat topology
of the tree likely reflects a largely common
founder stock and occurrence of extensive gene
flow between phenotypically dissimilar dogs be-
fore the advent of breed clubs and breed barrier
rules. In addition, it probably reflects the fact that
some historically older breeds that died out dur-
ing the famines, depressions, and wars of the
19th and 20th centuries have been recreated with
Division of Human Biology,
Division of Clinical Re-
search, Fred Hutchinson Cancer Research Center, Post
Office Box 19024, 1100 Fairview Avenue North, D4-
100, Seattle, WA 98109–1024, USA.
Molecular and
Cellular Biology Program, University of Washington, Box
357275, Seattle, WA 98195–7275, USA.
Department of
Genome Sciences, University of Washington, Box
351800, Seattle, WA 98195–7275, USA.
Department of
Veterinary Pathobiology, College of Veterinary Medi-
cine, University of Missouri, Columbia, MO 65211, USA.
Howard Hughes Medical Institute, 1100 Fairview Ave-
nue North, D4-100, Seattle, WA 98109 –1024, USA.
*To whom correspondence should be addressed. E-
R EPORTS SCIENCE VOL 304 21 MAY 2004 1161
on December 5, 2008 www.sciencemag.orgDownloaded from
the use of stock from phenotypically similar or
historically related dogs.
Whereas the phylogenetic analysis showed
separation of several breeds with ancient ori-
gins from a large group of breeds with pre-
sumed modern European origins, additional
subgroups may be present within the latter
group that are not detected by this approach for
at least two reasons (28). First, the true evolu-
tionary history of dog breeds is not well repre-
sented by the bifurcating tree model assumed
by the method because existing breeds were
mixed to create new breeds (a process that
continues today). Second, methods based on
genetic distance matrices lose information by
collapsing all genotype data for pairs of breeds
into a single number. The clustering algorithm
implemented in structure was explicitly de-
signed to overcome these limitations (11, 12,
28) and has been applied to infer the genetic
structure of several species (17, 28, 29). We
therefore ran structure on the entire data set
using increasing values of K (the number of
subpopulations the program attempts to find) to
identify ancestral source populations. In this
analysis, a modern breed could closely mirror a
single ancestral population or represent a mix-
ture of two or more ancestral types.
At K 2, one cluster was anchored by the
first seven breeds to split in the phylogenetic
analysis, whereas the other cluster contained
the large number of breeds with a flat phylo-
genetic topology (Fig. 3A). Five runs of the
program produced nearly identical results,
with a similarity coefficient (17 ) of 0.99
across runs. Seven other breeds share a size-
able fraction of their ancestry with the first
cluster. These fourteen breeds all date to an-
tiquity and trace their ancestry to Asia or
Africa. When a diverse set of wolves from
eight different countries was included in the
Fig. 1. Clustering assignment of 85 dog breeds. (A) Seventy-four breeds
are represented by five unrelated dogs each, and the remaining 11 breeds
are represented by four unrelated dogs each. Each individual dog is
represented on the graph by a vertical line divided into colored segments
corresponding to different genetic clusters. The length of each colored
segment is equal to the estimated proportion of the individual’s mem-
bership in the cluster of corresponding color (designated on the y axis as
a percentage). Breeds are labeled below the figure. (B) Six clusters
containing two breeds each are subdivided at K 2, with colors
representing the estimated proportion of individual membership in only
two possible clusters. Black lines separate individual dogs and the two
breeds are labeled below the figures.
Fig. 2. Consensus neighbor-joining tree of 85 dog breeds and the gray wolf. Nine breeds that form
branches with statistical support are shown. The remaining 76 breeds show little phylogenetic
structure and have been combined into one branch labeled “All other breeds” for simplification. The
entire tree is shown in fig. S1. The trees that formed the consensus are based on the chord distance
measure. Five hundred bootstrap replicates of the data were carried out, and the fraction of
bootstraps supporting each branch is indicated at the corresponding node as a percentage for those
branches supported in more than 50% of the replicates. The wolf population at the root of the tree
consists of eight individuals, one from each of the following countries: China, Oman, Iran, Sweden,
Italy, Mexico, Canada, and the United States. Branch lengths are proportional to bootstrap values.
21 MAY 2004 VOL 304 SCIENCE www.sciencemag.org1162
on December 5, 2008 www.sciencemag.orgDownloaded from
analysis, they fell entirely within this cluster
(Fig. 3B). The branch leading to the wolf
outgroup also fell within this group of breeds
in the phylogenetic analysis (Fig. 2).
At K 3, additional structure was detected
that was not readily apparent from the phylo-
genetic tree. The new third cluster consisted
primarily of breeds related in heritage and ap-
pearance to the Mastiff and is anchored by the
Mastiff, Bulldog, and Boxer, along with their
close relatives, the Bullmastiff, French Bulldog,
Miniature Bull Terrier, and Perro de Presa Ca-
nario. Also included in the cluster are the Rot-
tweiler, Newfoundland, and Bernese Mountain
Dog, large breeds that are reported to have
gained their size from ancient Mastiff-type an-
cestors. Less expected is the inclusion of the
German Shepherd Dog. The exact origins of
this breed are unknown, but our results suggest
that the years spent as a military and police dog
in the presence of working dog types, such as
the Boxer, are responsible for shaping the ge-
netic background of this popular breed. Three
other breeds showed partial and inconsistent
membership in this cluster across structure
runs (fig. S2), which lowered the similarity
coefficient to 0.84.
At K 4, a fourth cluster was observed,
which included several breeds used as herding
dogs: Belgian Sheepdog, Belgian Tervuren,
Collie, and Shetland Sheepdog. The Irish Wolf-
hound, Greyhound, Borzoi, and Saint Bernard
were also frequently assigned to this cluster.
Although historical records do not suggest that
these dogs were ever used to herd livestock, our
results suggest that these breeds are either pro-
genitors to or descendants of herding types. The
breeds in the remaining cluster are primarily of
relatively recent European origins and are
mainly different types of hunting dogs: scent
hounds, terriers, spaniels, pointers, and retriev-
ers. Clustering at K 4 showed a similarity
coefficient of 0.61, reflecting similar cluster
membership assignments for most breeds but
variable assignments for other breeds across
runs (fig. S2). At K 5, the similarity coeffi-
cient dropped to 0.26 and no additional consis-
tent subpopulations were inferred, suggesting a
lack of additional high-level substructure in the
sampled purebred dog population.
Our results paint the following picture of the
relationships among domestic dog breeds. Dif-
ferent breeds are genetically distinct, and indi-
viduals can be readily assigned to breeds on the
basis of their genotypes. This level of divergence
is surprising given the short time since the origin
of most breeds from mixed ancestral stocks and
supports strong reproductive isolation within
each breed as a result of the breed barrier rule.
Our results support at least four distinct breed
groupings representing separate adaptive radia-
tions. A subset of breeds with ancient Asian and
African origins splits off from the rest of the
breeds and shows shared patterns of allele fre-
quencies. At first glance, it is surprising that a
single genetic cluster includes breeds from Cen-
tral Africa (Basenji), the Middle East (Saluki and
Afghan), Tibet (Tibetan Terrier and Lhasa
Apso), China (Chow Chow, Pekingese, Shar-
Pei, and Shi Tzu), Japan (Akita and Shiba Inu),
and the Arctic (Alaskan Malamute, Siberian
Husky, and Samoyed). However, several re-
searchers have hypothesized that early pariah
dogs originated in Asia and migrated with no-
madic human groups both south to Africa and
north to the Arctic, with subsequent migrations
occurring throughout Asia (5, 6, 30). This cluster
includes Nordic breeds that phenotypically re-
semble the wolf, such as the Alaskan Malamute
and Siberian Husky, and shows the closest ge-
netic relationship to the wolf, which is the direct
Fig. 3. (A) Population structure of 85 domestic dog breeds. Each individual
dog is represented by a single vertical line divided into K colors, where K
is the number of clusters assumed. Each color represents one cluster, and
the length of the colored segment shows the individual’s estimated
proportion of membership in that cluster. Black lines separate the breeds
that are labeled below the figure. Representative breeds pictured above
the graph from left to right: Akita, Pekingese, Belgian Sheepdog, Collie,
Doberman Pinscher, Basset Hound, American Cocker Spaniel, Bedlington
Terrier, Flat-Coated Retriever, Newfoundland, and Mastiff. Results shown
are averages over 15 structure runs at each value of K.(B) Population
structure, as in (A), but with gray wolves included. Graph shown is
averaged over five structure runs at K 2.
R EPORTS SCIENCE VOL 304 21 MAY 2004 1163
on December 5, 2008 www.sciencemag.orgDownloaded from
ancestor of domestic dogs. Thus, dogs from
these breeds may be the best living representa-
tives of the ancestral dog gene pool. It is notable
that several breeds commonly believed to be of
ancient origin, such as the Pharaoh Hound and
Ibizan Hound, are not included in this group.
These are often thought to be the oldest of all
dog breeds, descending directly from the ancient
Egyptian dogs drawn on tomb walls more than
5000 years ago. Our results indicate, however,
that these two breeds have been recreated in
more recent times from combinations of other
breeds. Thus, although their appearance
matches the ancient Egyptian sight hounds,
their genomes do not. Similar conclusions ap-
ply to the Norwegian Elkhound, which clus-
ters with modern European breeds rather than
with the other Arctic dogs, despite reports of
direct descent from Scandinavian origins more
than 5000 years ago (1, 24 ).
The large majority of breeds appears to rep-
resent a more recent radiation from shared Eu-
ropean stock. Although the individual breeds are
genetically differentiated, they appear to have
diverged at essentially the same time. This radi-
ation probably reflects the proliferation of dis-
tinct breeds from less codified phenotypic vari-
eties after the introduction of the breed concept
and the creation of breed clubs in Europe in the
1800s. A more sensitive cluster analysis was
able to discern additional genetic structure of
three subpopulations within this group. One con-
tains Mastiff-like breeds and appears to reflect
shared morphology derived from a common an-
cestor. Another includes Shetland Sheepdog, the
two Belgian Sheepdogs, and Collie, and may
reflect shared ancestral herding behavior. The
remaining population is dominated by a prolif-
eration of breeds dedicated to various aspects of
the hunt. For these breeds, historical and breed
club records suggest highly intertwined blood-
lines, consistent with our results.
Dog breeds have traditionally been
grouped on the basis of their roles in human
activities, physical phenotypes, and histor-
ical records. Here, we defined an indepen-
dent classification based on patterns of ge-
netic variation. This classification supports
a subset of traditional groupings and also
reveals previously unrecognized connec-
tions among breeds. An accurate under-
standing of the genetic relationships among
breeds lays the foundation for studies
aimed at uncovering the complex genetic
basis of breed differences in morphology,
behavior, and disease susceptibility.
References and Notes
1. J. Crowley, B. Adelman, Eds., The Complete Dog Book;
Official Publication of the American Kennel Club
(Howell Book House, New York, ed. 19, 1998).
2. D. F. Patterson et al., J. Am. Vet. Med. Assoc. 193,
1131 (1988).
3. E. A. Ostrander, L. Kruglyak, Genome Res. 10, 1271 (2000).
4. C. Vila et al., Science 276, 1687 (1997).
5. P. Savolainen, Y. P. Zhang, J. Luo, J. Lundeberg, T.
Leitner, Science 298, 1610 (2002).
6. J. A. Leonard et al., Science 298, 1613 (2002).
7. C. A. Rogers, A. H. Brace, The International Encyclopedia of
Dogs (Howell Book House, New York, ed. 1, 1995).
8. B. Fogel, The Encyclopedia of the Dog (DK Publishing,
New York, 1995).
9. M. T. Koskinen, Anim. Genet. 34, 297 (2003).
10. D. N. Irion et al., J. Hered. 94, 81 (2003).
11. J. K. Pritchard, M. Stephens, P. Donnelly, Genetics
155, 945 (2000).
12. D. Falush, M. Stephens, J. K. Pritchard, Genetics 164,
1567 (2003).
13. F. Tajima, M. Nei, Mol. Biol. Evol. 1, 269 (1984).
14. R. Sachidanandam et al., Nature 409, 928 (2001).
15. J. C. Venter et al., Science 291, 1304 (2001).
16. L. Excoffier, P. E. Smouse, J. M. Quattro, Genetics 131,
479 (1992).
17. N. A. Rosenberg et al., Science 298, 2381 (2002).
18. L. L. Cavelli-Sforza, P. Menozzi, A. Piazza, The History
and Geography of Human Genes (Princeton Univ.
Press, Princeton, NJ, 1994).
19. D. E. MacHugh, R. T. Loftus, P. Cunningham, D. G.
Bradley, Anim. Genet. 29, 333 (1998).
20. G. Laval et al., Genet. Sel. Evol. 32, 187 (2000).
21. J. K. Pritchard, M. Stephens, N. A. Rosenberg, P. Don-
nelly, Am. J. Hum. Genet. 67, 170 (2000).
22. Materials and methods are available as supporting
material on Science Online.
23. T. Yamazaki, K. Yamazaki, Legacy of the Dog: The
Ultimate Illustrated Guide to Over 200 Breeds (Chron-
icle Books, San Francisco, CA, 1995).
24. B. Wilcox, C. Walkowicz, Atlas of Dog Breeds of the
World (TFH Publications, Neptune City, NJ, ed. 5,
25. D. Paetkau, W. Calvert, I. Stirling, C. Strobeck, Mol.
Ecol. 4, 347 (1995).
26. L. L. Cavalli-Sforza, A. W. Edwards, Evolution 32, 550
27. D. B. Goldstein, A. R. Linares, L. L. Cavalli-Sforza,
M. W. Feldman, Genetics 139, 463 (1995).
28. N. A. Rosenberg et al., Genetics 159, 699 (2001).
29. D. Falush et al., Science 299, 1582 (2003).
30. M. V. Sablin, G. A. Khlopachev, Curr. Anthropol. 43,
795 (2002).
31. Supported by the Burroughs Wellcome Innovation
Award (E.A.O. and L.K.), grants from the AKC–Canine
Health Foundation (G.S.J.), NIH training grant T32
HG00035 (H.G.P.), and a postdoctoral fellowship
from the Waltham Foundation (N.B.S.). E.A.O. also
acknowledges support from K05 CA90754. L.K. is a
James S. McDonnell Centennial Fellow. We thank the
many dog owners, researchers, and breeders who
provided DNA samples for this work, especially G.
Brewer, C. Gaiser, and K. Murphy; D. Lynch, A. Ziska,
C. Ramirez, M. Langlois, and D. Akey for assistance
with sample collection; M. Stephens for advice and
assistance with the program structure; K. Markianos
and J. Akey for helpful discussions; M. Eberle for
computing assistance; H. Coller, E. Giniger, R. Wayne,
and three anonymous reviewers for comments on the
manuscript; and the AKC and C. Jierski for use of the
canine artwork included in Fig. 3.
Supporting Online Material
Materials and Methods
Figs. S1 and S2
Tables S1 to S5
2 March 2004; accepted 21 April 2004
Mutational Analysis of the
Tyrosine Phosphatome in
Colorectal Cancers
Zhenghe Wang,
Dong Shen,
* D. Williams Parsons,
* Alberto
* Jason Sager,
Steve Szabo,
Janine Ptak,
, Brock A. Peters,
Michiel S. van der Heijden,
Hai Yan,
Tian-Li Wang,
Greg Riggins,
Steven M.
James K. V. Willson,
Sanford Markowitz,
Kenneth W.
Bert Vogelstein,
Victor E. Velculescu
Tyrosine phosphorylation, regulated by protein tyrosine phosphatases (PTPs)
and kinases (PTKs), is important in signaling pathways underlying tumorigen-
esis. A mutational analysis of the tyrosine phosphatase gene superfamily in
human cancers identified 83 somatic mutations in six PTPs (PTPRF, PTPRG,
PTPRT, PTPN3, PTPN13, PTPN14 ), affecting 26% of colorectal cancers and a
smaller fraction of lung, breast, and gastric cancers. Fifteen mutations were
nonsense, frameshift, or splice-site alterations predicted to result in truncated
proteins lacking phosphatase activity. Five missense mutations in the most
commonly altered PTP (PTPRT) were biochemically examined and found to
reduce phosphatase activity. Expression of wild-type but not a mutant PTPRT
in human cancer cells inhibited cell growth. These observations suggest that the
mutated tyrosine phosphatases are tumor suppressor genes, regulating cellular
pathways that may be amenable to therapeutic intervention.
Phosphorylation of tyrosine residues is a central
feature of many cellular signaling pathways,
including those affecting growth, differentia-
tion, cell cycle regulation, apoptosis, and inva-
sion (1, 2). This phosphorylation is coordinately
controlled by protein tyrosine kinases (PTKs)
and phosphatases (PTPs). Although a variety of
PTK genes have been directly linked to tumor-
igenesis through somatic activating mutations
(36), only a few PTP genes have been impli-
cated in cancer (710). Moreover, it is not
known how many or how frequently members
21 MAY 2004 VOL 304 SCIENCE www.sciencemag.org1164
on December 5, 2008 www.sciencemag.orgDownloaded from
... Such an event would have to have preceded the derivation of all domestic dogs, since the association between a y and A82 R83 extends across distantly related breeds. Recently, Parker et al. (2004) used 96 microsatellites to sort 85 dog breeds into four groupings: ancient Spitz type, giant Mastiff type, herding type, and the majority of other types. The S82 H83 variant was found in representatives of all four of these groups: the Chow, Shar-Pei, Pekingese, Akita, and Afghan Hound in the ancient group; the Mastiff, Bullmastiff, Boxer, and French Bulldog in the giant group; the Belgian Shepherds, Collie, and Shetland Sheepdog in the herding group; and the Dachshund, Great Dane, Pug, and Cairn Terrier in the "others" group (Table 2). ...
Association of SKG13 gene with hairlessness in domestic dogs
... In addition, there could be the influence of recent selection on the breed-typical behavior. However, the problem-solving test in our study, combined with the result of earlier genome-wide studies of the domestic dogs 7,25,26 , indicates that Japanese dogs which have relatively close DNAs to wolves show behaviour similar to wolves. The results of the two-way choice test were similar in both groups of dogs, showing that both groups have similar capabilities in understanding human gestures and adjusting their responses accordingly. ...
Full-text available
The dog ( Canis familiaris ) was the first domesticated animal and hundreds of breeds exist today. During domestication, dogs experienced strong selection for temperament, behaviour, and cognitive ability. However, the genetic basis of these abilities is not well-understood. We focused on ancient dog breeds to investigate breed-related differences in social cognitive abilities. In a problem-solving task, ancient breeds showed a lower tendency to look back at humans than other European breeds. In a two-way object choice task, they showed no differences in correct response rate or ability to read human communicative gestures. We examined gene polymorphisms in oxytocin, oxytocin receptor, melanocortin 2 receptor, and a Williams–Beuren syndrome-related gene (WBSCR17), as candidate genes of dog domestication. The single-nucleotide polymorphisms on melanocortin 2 receptor were related to both tasks, while other polymorphisms were associated with the unsolvable task. This indicates that glucocorticoid functions are involved in the cognitive skills acquired during dog domestication.
... Several studies in the field of AD research highlighted the importance of diagnostic and therapeutic interventions in the asymptomatic phase while the individuals are cognitively intact [14,105] . However, it is challenging to first identify and then track disease progression in humans without sensitive, specific, and cost-effective early diagnostic approaches and a lack of robust natural translational models. ...
Full-text available
Aims: Cerebral amyloid burdens may be found in otherwise cognitively intact adults, often not showing worsening deficits with passing years. Alzheimer’s transgenic rodents have been widely used to investigate this phenomenon, but a spontaneous disorder in other animals, such as dogs that cohabit with humans and thus may have some shared environmental risks, may contribute and offer opportunities not possible in the smaller laboratory animals. In animals, the spontaneous disorder most comparable to Alzheimer’s disease (AD) affects mature to aged dogs and is designated canine cognitive dysfunction. Motivated by AD, many studies have revealed that amyloid progressively accumulates in the canine central nervous system, including the retina. Here, we investigated whether deposits of amyloid and/or tau can be found in the canine retina of neurologically normal animals from the first year of life to the elderly. Suppose canine ocular amyloid and tau are present from early life. In that case, that raises the question of whether similar patterns of accumulation occur in man, whether as part of aging, AD, or other. Methods: This study used eye tissues from 30 dogs with a variety of ophthalmic or other orbital disorders, of which 7/30 were 1-2 years old. Tissues were subdivided into dogs of three different age groups: young (1-5 years old), middle (6-10 years old), and old (≥ 11 years old). Results: Following immunostaining of tissue sections with nanobodies against retinal Aβ1-40 and Aβ1-42 oligomers, and antibodies against Aβ plaques (Aβp) and hyperphosphorylated Tau (p-Tau), our investigations revealed that accumulation of Aβ1-40 and Aβ1-42 oligomers were widespread in the retina in all age groups. In contrast, Aβp were detected in the middle and old age groups but not in the young age group. Furthermore, p-Tau staining was observed in four old dogs only, while other dogs were p-Tau free. Interestingly, both Aβo and Aβp co-localized in the middle and old age groups of dogs. Moreover, diffuse granular p-Tau co-localized with intracellular Aβo in the old age group. Finally, we also observed co-localization of Aβo and Aβp in the retinal vasculature which might be similar to cerebral amyloid angiopathy associated with AD. Conclusion: As far as we know, the presence of amyloid and tau in the canine retina is hitherto unreported. If similar, early-in-life subclinical retinal deposits occur in a human cohort perhaps identified by AD genetic risk factors, following this group may offer the prospect of preclinical therapeutic intervention in imminent dementia, a strategy recognized as likely necessary to impact this burgeoning disorder.
... These results indicate that the NMR metabolomics method is a valuable tool for metabolic studies in basic physiology and form a solid foundation for canine metabolomics studies examining disease associations. Dog breeds differ in terms of genetics, morphology, physiology and behaviour [6][7][8][9], and are suggested to also vary in their metabolism [10,11]. In our study, breed was a powerful driver of variation in all studied measurands. ...
Full-text available
As an individual's metabolism reflects health and disease states well, metabolomics holds a vast potential in biomedical applications. However, normal physiological factors, such as age, can also influence metabolism, challenging the establishment of disease-specific metabolic aberrations. Here, we examined how physiological and diet-related factors drive variance in the metabolism of healthy pet dogs. We analysed 2068 serum samples using a canine nuclear magnetic resonance (NMR) spectroscopy-based metabolomics platform. With generalized linear models, we discovered that age, breed, sex, sterilization, diet type and fasting time significantly affected the canine metabolite profiles. Especially, breed and age caused considerable variation in the metabolite concentrations, and breeds with very different body conformations systematically differed in several lipid measurands. Our results enhance the understanding how normal physiological factors influence canine metabolism, aid accurate interpretation of the NMR results, and suggest the NMR platform might be applied in identifying aberrations in nutrient absorption and metabolism.
... Courrriel : L'origine des chiens de race est réputée polyphylétique, et beaucoup de croisements se sont produits entre différentes lignées dont quatre d'entre elles, chacune reliée au loup de façon indépendante, ont constitué, sur des critères morphologiques, des pôles de regroupements inégaux (Parker et al. 2004 ;Wang & Tedford, 2008) : quatre ensembles génétiques (A, B, C, D) rassemblent des populations d'origines géographiques identiques (Asie, Afrique), chacune étant composée de sujets aux morphologies semblables, ainsi que deux ensembles associés (E, F) et un plus récent (G), européen, composé de 3 groupes (a, b, c) ( Figure 1). Le plus important en effectif et en diversité génétique est le groupe A (chiens primitifs, asiatiques et certains européens) ; il est admis que des croisements entre chien et loup, et/ou une domestication renouvelée de certains loups auraient eu lieu dans ces populations (Wayne &Vilà, 2001 ;Lignereux, 2006). ...
La cohabitation du chien et de l’Homme s’est traduite par l’émergence d’un morphotype canin ancestral, puis de races primaires, à partir desquelles la sélection a créé les races modernes que nous connaissons actuellement. Les standards en définissent les phénotypes, mais leur interprétation peut privilégier la production de certains morphotypes raciaux déviants, inductrice d’un appauvrissement de la variabilité génétique par pression sélective inappropriée. Le vétérinaire a toute légitimité pour s’impliquer dans le maintien de la nécessaire diversité génétique des effectifs ; il a le devoir de veiller au bien-être et à la santé du chien de race, en rappelant qu’une dépendance interspécifique harmonieuse du chien et de l’Homme impose à ce dernier de pratiquer une sélection morphologique raisonnable et de favoriser l’épanouissement des aptitudes comportementales naturelles dans chaque groupe de races. Mots-clés :chien,éthique, race, sélection, standard de race.
Differences in the behavior of the domestic dog (Canis lupus familiaris) and its progenitor species, the gray wolf (Canis lupus), are well recognized but the mechanisms of the wolf to dog transformation remain an area of scientific debate. A view of dog domestication that is centered on genetic selection for behavioral traits receives support from the famous Russian farm-fox experiment that began in the 1950s. Selection of foxes (Vulpes vulpes), separately, for tame and for aggressive behavior, has yielded two strains with markedly different, genetically determined, behavioral phenotypes. Tame-strain foxes communicate with humans in a positive manner and are eager to establish human contact. Conversely, foxes from the aggressive strain are aggressive to humans and difficult to handle. Although selected solely for behavior, changes in physiology, morphology, and appearance with significant parallels to characteristics of the domestic dog were observed in the selected strains. The genetic analysis of the fox populations identified several genomic regions that are homologous to the regions in the dog genome that differentiate dogs from wolves. Although the genetic regulation of domesticated behavior is far from being completely understood enormous progress has been made in this field. This chapter reviews studies of behavior and genetics in dogs and foxes and highlights the role of selection for behavior in ancient and modern dog formation.
Canine behavior has been studied for decades, but not until 1998 was it discovered that dogs have human-like cooperative communication skills that rival those of even our closest primate relatives. Ever since, canines have become subjects of increased research into the genetic underpinnings of these abilities. Here, we posit that domestication has been a driving force in the evolution of dog cognition. The latest technological advances have been instrumental in allowing us to have a better understanding of the impact of domestication on the canine genome, as well as the role that genetics play in dog behavior and cognition. Finally, we explore the ways this knowledge can be applied to better the lives of dogs and that of their human companions.
Full-text available
GM1 gangliosidosis is a progressive, recessive, autosomal, neurodegenerative, lysosomal storage disorder that affects the brain and multiple systemic organs due to an acid β-galactosidase deficiency encoded by the GLB1 gene. This disease occurs in the Shiba Inu breed, which is one of the most popular traditional breeds in Japan, due to the GLB1:c.1649delC (p.P550Rfs*50) mutation. Previous surveys performed of the Shiba Inu population in Japan found a carrier rate of 1.02–2.94%. Currently, a miniature type of the Shiba Inu called “Mame Shiba”, bred via artificial selection to yield smaller individuals, is becoming more popular than the standard Shiba Inu and it is now one of the most popular breeds in Japan and China. The GM1 gangliosidosis mutation has yet to be surveyed in the Mame Shiba population. This study aimed to determine the frequency of the mutant allele and carrier rate of GM1 gangliosidosis in the Mame Shiba breed. Blood samples were collected from 1832 clinically healthy adult Mame Shiba Inus used for breeding across 143 Japanese kennels. The genotyping was performed using a real-time PCR assay. The survey found nine carriers among the Mame Shibas, indicating that the carrier rate and mutant allele frequency were 0.49% and 0.00246, respectively. This study demonstrated that the mutant allele has already been inherited by the Mame Shiba population. There is a risk of GM1 gangliosidosis occurrence in the Mame Shiba breed if breeders use carriers for mating. Further genotyping surveys are necessary for breeding Mame Shibas to prevent the inheritance of this disease.
Based on claims that dogs are less aggressive and show more sophisticated socio-cognitive skills compared with wolves, dog domestication has been invoked to support the idea that humans underwent a similar ‘self-domestication’ process. Here, we review studies on wolf–dog differences and conclude that results do not support such claims: dogs do not show increased socio-cognitive skills and they are not less aggressive than wolves. Rather, compared with wolves, dogs seek to avoid conflicts, specifically with higher ranking conspecifics and humans, and might have an increased inclination to follow rules, making them amenable social partners. These conclusions challenge the suitability of dog domestication as a model for human social evolution and suggest that dogs need to be acknowledged as animals adapted to a specific socio-ecological niche as well as being shaped by human selection for specific traits.
Dogs are remarkable, adaptable, and dependable creatures that have evolved alongside humans while contributing tremendously to our survival. Our canine companions share many similarities to human disease, particularly cancer. With the advancement of next-generation sequencing technology, we are beginning to unravel the complexity of cancer and the vast intra- and intertumoral heterogeneity that makes treatment difficult. Consequently, precision medicine has emerged as a therapeutic approach to improve patient survival by evaluating and classifying an individual tumor's molecular profile. Many canine and human cancers share striking similarities in terms of genotypic, phenotypic, clinical, and histological presentations. Dogs are superior to rodent models of cancer because they are a naturally heterogeneous population in which tumors occur spontaneously, are exposed to similar environmental conditions, and show more similarities in key modulators of tumorigenesis and clinical response, including the immune system, drug metabolism, and gut microbiome. In this chapter, we will explore various canine models of human cancers and emphasize the dog's critical role in advancing precision medicine and improving the survival of both man and man's best friend.
Full-text available
Abstract A set of eleven pig breeds originating from six European countries, and including a small sample of wild pigs, was chosen for this study of genetic diversity. Diversity was evaluated on the basis of 18 microsatellite markers typed over a total of 483 DNA samples collected. Average breed heterozygosity varied from 0.35 to 0.60. Genotypic frequencies generally agreed with Hardy-Weinberg expectations, apart from the German Landrace and Schwäbisch-Hällisches breeds, which showed significantly reduced heterozygosity. Breed differentiation was significant as shown by the high among-breed fixation index (overall FST = 0.27), and confirmed by the clustering based on the genetic distances between individuals, which grouped essentially all individuals in 11 clusters corresponding to the 11 breeds. The genetic distances between breeds were first used to construct phylogenetic trees. The trees indicated that a genetic drift model might explain the divergence of the two German breeds, but no reliable phylogeny could be inferred among the remaining breeds. The same distances were also used to measure the global diversity of the set of breeds considered, and to evaluate the marginal loss of diversity attached to each breed. In that respect, the French Basque breed appeared to be the most "unique" in the set considered. This study, which remains to be extended to a larger set of European breeds, indicates that using genetic distances between breeds of farm animals in a classical taxonomic approach may not give clear resolution, but points to their usefulness in a prospective evaluation of diversity.
Full-text available
We present here a framework for the study of molecular variation within a single species. Information on DNA haplotype divergence is incorporated into an analysis of variance format, derived from a matrix of squared-distances among all pairs of haplotypes. This analysis of molecular variance (AMOVA) produces estimates of variance components and F-statistic analogs, designated here as phi-statistics, reflecting the correlation of haplotypic diversity at different levels of hierarchical subdivision. The method is flexible enough to accommodate several alternative input matrices, corresponding to different types of molecular data, as well as different types of evolutionary assumptions, without modifying the basic structure of the analysis. The significance of the variance components and phi-statistics is tested using a permutational approach, eliminating the normality assumption that is conventional for analysis of variance but inappropriate for molecular data. Application of AMOVA to human mitochondrial DNA haplotype data shows that population subdivisions are better resolved when some measure of molecular differences among haplotypes is introduced into the analysis. At the intraspecific level, however, the additional information provided by knowing the exact phylogenetic relations among haplotypes or by a nonlinear translation of restriction-site change into nucleotide diversity does not significantly modify the inferred population genetic structure. Monte Carlo studies show that site sampling does not fundamentally affect the significance of the molecular variance components. The AMOVA treatment is easily extended in several different directions and it constitutes a coherent and flexible framework for the statistical analysis of molecular data.
We describe extensions to the method of Pritchard et al. for inferring population structure from multilocus genotype data. Most importantly, we develop methods that allow for linkage between loci. The new model accounts for the correlations between linked loci that arise in admixed populations (“admixture linkage disequilibium”). This modification has several advantages, allowing (1) detection of admixture events farther back into the past, (2) inference of the population of origin of chromosomal regions, and (3) more accurate estimates of statistical uncertainty when linked loci are used. It is also of potential use for admixture mapping. In addition, we describe a new prior model for the allele frequencies within each population, which allows identification of subtle population subdivisions that were not detectable using the existing method. We present results applying the new methods to study admixture in African-Americans, recombination in Helicobacter pylori, and drift in populations of Drosophila melanogaster. The methods are implemented in a program, structure, version 2.0, which is available at
We describe a model-based clustering method for using multilocus genotype data to infer population structure and assign individuals to populations. We assume a model in which there are K populations (where K may be unknown), each of which is characterized by a set of allele frequencies at each locus. Individuals in the sample are assigned (probabilistically) to populations, or jointly to two or more populations if their genotypes indicate that they are admixed. Our model does not assume a particular mutation process, and it can be applied to most of the commonly used genetic markers, provided that they are not closely linked. Applications of our method include demonstrating the presence of population structure, assigning individuals to populations, studying hybrid zones, and identifying migrants and admixed individuals. We show that the method can produce highly accurate assignments using modest numbers of loci—e.g., seven microsatellite loci in an example using genotype data from an endangered bird species. The software used for this article is available from
An attempt has been made to establish a procedure for estimating the course taken by evolution. The model used is that of a branching random walk, which is strictly valid only when the causes of divergence between populations are random genetic drift and variable selection. With suitable transformations of the data, evolution can then be considered as a branching Brownian-motion process. To keep the model as simple as possible it was supposed that no population becomes extinct and that each population splits, at a random time, into two daughter populations each identical to its parent. The problem was to estimate the form and dimensions of the most probable tree uniting the presently living populations. The ideal method of estimation, maximum likelihood, proved difficult and had to be replaced in part by alternative procedures. In addition to describing the available procedures in detail, a simple example is worked out fully, and the logical content and limitations of the methods are considered in depth.