close

Вход

Забыли?

вход по аккаунту

?

DNA polymorphisms detect ancient barriers to gene flow in Basques.

код для вставкиСкачать
AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 122:73– 84 (2003)
DNA Polymorphisms Detect Ancient Barriers to Gene
Flow in Basques
M. Iriondo,1* M.C. Barbero,2 and C. Manzano1
1
Department of Animal Biology and Genetics, Faculty of Sciences, University of the Basque Country,
48080 Bilbao, Spain
2
Department of Zoology and Animal Cell Dynamics, Faculty of Sciences, University of the Basque Country,
48080 Bilbao, Spain
KEY WORDS
genetic heterogeneity; genetic structure; Basque Country; genetic
boundaries; STR; FST
ABSTRACT
This work features the first district-bydistrict analysis of all provinces in the Iberian Peninsula
with an autochthonous Basque population, and indicates
the existence of genetic heterogeneity. The populations
cluster in three groups arising from processes of genetic
drift which probably occurred in pre-Mesolithic times, and
were probably those which repopulated the southern areas of the Basque Country after the Last Glacial Maximum. It seems that from that period onwards, the population settled in three major groups (West Basques,
Central Basques, and East Basques), along geographical
axes which appear substantial in the maintaining of each
population unit. This genetic structure is probably reflected in other aspects such as the existence of ancient
tribes and the dialects of the Basque language, the boundaries of which may be related at origin and which are quite
similar to those detected in this work. Our results indicate
that the populations of the Basque Country are genetically close to other neighboring populations, such as that
of Aragon, which may indicate an outgoing gene flow from
the Basque area down the River Ebro towards the Mediterranean seaboard. While our short tandem repeat data
suggest that population structure within the Basques
dates back to the Mesolithic, our findings are also consistent with the hypothesis that patterns of modern European genetic diversity have been shaped mainly during
the Neolithic. Am J Phys Anthropol 122:73– 84, 2003.
The present-day Basque Country is made up administratively of several provinces in all or part of
which Basque is still spoken as a mother tongue.
Although Basque-related languages were once spoken over a wider area, the language is now native to
seven provinces (Fig. 1), four of them in Spain
(Alava, Biscay, Guipuzcoa, and Navarre), and the
others in the French Département des PyrénéesAtlantiques (Labourd, Basse-Navarre, and Soule).
The possibility that there might be genetic heterogeneity within the Basque population was suggested when the first genetic analyses were performed on that population, and discrepancies were
obtained in the results from different samples (Marquer, 1963; Goedde et al., 1972, 1973; Constans and
Viau, 1975). Genetic heterogeneity was subsequently observed in several studies based on the
analysis of classical polymorphisms, between provinces (Manzano et al., 1993a,b, 1996a,b), between
natural districts in Biscay (Aguirre et al., 1991), and
between districts in the provinces of Alava, Guipuzcoa, and Biscay (Manzano et al., 2002). This last
study analyzed 18 classical polymorphic systems in
14 natural districts in the provinces of Alava,
Biscay, and Guipuzcoa, using the same sampling
system and method of analysis throughout the territory studied. Geographically patterned genetic
heterogeneity was described, and various factors
were tested which might have contributed to the
genetic structure. Those factors included a moderate
influence on allele frequencies of the main travel
routes through the Basque Country, and a slight
influence of ancient tribal distribution.
Even so, the area analyzed in Manzano et al.
(2002) did not include all the areas with native
Basque speakers. The biggest such area in terms
of area and population is the province of Navarre,
which is also the territory where the ancient tribe
the Vascones, from whom the name “Basque” derives, once lived. It therefore seems of interest to
include this area in analyses of the genetic structure of the Basque population. The province of
©
2003 WILEY-LISS, INC.
©
2003 Wiley-Liss, Inc.
Grant sponsor: University of the Basque Country; Grant sponsor:
Ministerio de Educación y Ciencia.
*Correspondence to: Mikel Iriondo, Department of Animal Biology
and Genetics, Faculty of Sciences, University of the Basque Country,
P.O.B. 644, 48080 Bilbao, Spain. E-mail: ggpirorm@lg.ehu.es
Received 29 May 2002; accepted 8 October 2002.
DOI 10.1002/ajpa.10212
74
M. IRIONDO ET AL.
Fig. 1. Districts analyzed within Basque Country. Dotted line indicates watershed. Rivers mentioned in text are shown, as are the
Pyrenees and region of Aragon (see text). Districts: Uribe (URI), Gernika (GER), Markina (MAR), Lea (LEA), Arratia (ARR), Durango
(DUR), Deba (DEB), Tolosa (TOL), Urola (URO), Goierri (GOI), Llanada (LLA), Montaña (MON), Valles (VAL), Rioja (RIO), Bidasoa
(BID), Iruñea (IRU), and Ebro (EBR).
Navarre was recently analyzed for various genetic
polymorphisms, and it was observed that although
the allele frequencies found were significantly different from those published previously in Basque
series, they nevertheless fit into the cluster
formed by Basque series in multivariate analyses
(Iriondo et al., 1999).
Furthermore, as pointed out by Chikhi et al.
(1998), recent studies of the distribution of
mtDNA in Europe showed little or no geographic
structure, which conflicts with previous results on
protein studies. This raises the suspicion that
variation at the protein level may not faithfully
mirror DNA diversity. Although the analysis of Y
chromosome haplotypes in Europe (Semino et al.,
2000) does reveal geographic structures, it has
been used (like mtDNA analysis) as evidence of an
ancient Paleolithic origin of European popula-
tions, contradicting the interpretation of the results obtained with protein markers. In this context, we believe that it is interesting to analyze
DNA markers in order to look more deeply at the
genetic structure of the Basques and eliminate
any effects of mechanisms such as natural selection on classical polymorphisms. We therefore undertook a microevolutionary study of the Basques,
using the same sampling system and method of
analysis throughout the territory covered. We selected the natural district as population unit, and
six short tandem repeat (STR) systems as the
polymorphisms to be studied.
We believe that including the population of the
province of Navarre and using DNA markers will
provide clearer, more rigorous results on the overall
configuration of internal genetic variability in the
Basques of the Iberian Peninsula.
ANCIENT BARRIERS TO GENE FLOW IN BASQUES
MATERIALS AND METHODS
We analyzed the provinces of Alava, Biscay,
Guipuzcoa, and Navarre, which are home to the
entire native Basque-speaking population of the Iberian Peninsula. The area analyzed was broken down
into 17 natural districts (Fig. 1). In all, 797 unrelated autochthonous individuals of both sexes were
analyzed, all of them native to one of the said 17
natural districts. Autochthony was checked by confirming that all four grandparents of each individual
had been born in the same area. The sample size
used is sufficient for a genetic population analysis
using microsatellites (Shriver et al., 1995; Takezaki
and Nei, 1996; Pérez-Lezaun et al., 1997). All samples were obtained with appropriate informed consent.
DNA was extracted using either the phenol-chloroform method (Smith et al., 1990) or the Chelex 100
resin method (Walsh et al., 1991). The samples were
amplified in a GeneAmp PCR system 2400 (PerkinElmer), using CSF1PO-TH01-TPOX and F13A1FES/FPS-VWF GenePrint STR Multiplex Systems
(Promega) according to the manufacturer’s instructions. The products of the polymerase chain reaction
(PCR) were separated for typing, using a denaturing
PAGE (4% acrylamide-bisacrylamide 19:1, 7 M urea,
thickness 0.4 mm). The sequencing gels were electrophoresed at a constant 1,500 V for approximately
3 hr. Silver stain was applied using the method
proposed by Bassam et al. (1991).
Allele frequencies were estimated by gene counting. Possible divergence from Hardy-Weinberg expectations was determined by calculating the exact
test proposed by Guo and Thompson (1992), running
the statistical package GENEPOP (Raymond and
Rousset, 1995). Heterozygosity was calculated following Nei (1978). The parameters F, ␪, and ƒ,
equivalent to FIT, FST, and FIS, respectively, were
calculated according to the formulae of Weir and
Cockerham (1984), running the FSTAT program
(Goudet, 1995). The confidence intervals of these
parameters were calculated by jackknife, and statistical significance was calculated using 15,000 permutations, all using FSTAT software. As indicated
by Weir and Cockerham (1984), the calculation of
these parameters by these formulae is unaffected by
the sampling procedure, the number of alleles per
locus, the number of individuals per population, or
the number of populations analyzed. When necessary, Bonferroni’s correction was performed (Weir,
1996). The R matrix (Harpending and Jenkins,
1973) was calculated on the basis of allele frequencies. FST genetic distance (Reynolds et al., 1983;
Cavalli-Sforza et al., 1994), neighbor-joining trees,
and bootstrap procedures (1,000 iterations) were
performed using the PHYLIP 3.5 statistical package
(Felsenstein, 1989). To check for correlation between
genetic and geographical distances, we drew up a
matrix with geographical distances between pairs of
districts. Geographical distances were measured as
75
the straight-line distance between the main towns of
the districts. Hierarchical analyses of molecular
variance (AMOVA; Excoffier et al., 1992) were performed using Arlequin version 1.1 software (Schneider et al., 1997). First-order correlations, partial
correlations (Smouse et al., 1986), and multiple correlations (Dow and Cheverud, 1985) were used to
explore the relationships between various types of
distance matrix. In these cases, SPSS 9.0 software
was used to obtain residual and prediction matrices.
Matrix comparison by Mantel’s method was carried
out using the NTSYS program (Rohlf, 1988), and
significance was obtained after 10,000 iterations
(Smouse et al., 1986). Spatial autocorrelation analysis (Sokal and Oden, 1978) and correlogram construction were carried out with SAAP software
(Wartenberg, 1989). Moran’s coefficient I (Moran,
1950) was chosen as the coefficient of spatial autocorrelation. The COCOPAN method (contiguity-constrained permutational ANOVA; Legendre et al.,
1990) and Delaunay’s triangulation method were
performed using R Package software (Legendre and
Vaudor, 1991).
The allele frequency data of the populations from
the Iberian Peninsula (Aragon, Andalusia, Central
Spain, Asturias, Pyrenees, Catalonia, North Portugal, and Central Portugal) were obtained from
the database http://www.ertzaintza.net/adn_nuclear,
and from Crespillo et al. (1996), Andrés et al.
(1996), Gamero et al. (1998), and Santos et al.
(1996). Allele frequencies from general samples
from European countries were obtained at http://
www.uni-duesseldorf.de/WWW/MedFak/Serology/
dna.html. The frequencies for North Poland and
South Poland were obtained from Pawlowski et al.
(1997) and Miscicka-Sliwka et al. (1998), respectively.
RESULTS
Table 1 shows the allele frequencies obtained for
the six STR systems analyzed in the 17 natural
districts studied. The results of the check of
Hardy-Weinberg proportions (not shown) indicate
that there is significant deviation only in the
HUMCSF1PO system in the Navarrese district of
IRU (P ⫽ 0.0139 ⫾ 0.0005) and in the HUMTPOX
system (in the Biscayne district of URI (P ⫽
0.0382 ⫾ 0.0006). These two results, which account
for 2% of the total (2 out of 102), may be attributed
to chance. Taken as a whole, once Bonferroni’s correction is applied, both the systems and the natural
districts are in Hardy-Weinberg equilibrium. Similarly, Table 2 shows that the parameters F (FIT) and
ƒ (FIS) evidence a slight, nonsignificant deficit of
homozygotes.
The districts analyzed do not form a homogenous
whole. Table 2 shows genetic heterogeneity after
Bonferroni’s correction for the HUMCSF1PO and
HUMTPOX systems, in both the exact test and the
calculation of the ␪ value (FST). In the exact test
overall significance was also calculated using ␹2,
TABLE 1. Allele frequencies for six STR polymorphisms analyzed in 17 districts of Basque Country1
GUIPUZCOA
CSF1PO*8
CSF1PO*8
CSF1PO*9
CSF1PO*10
CSF1PO*11
CSF1PO*12
CSF1PO*13
CSF1PO*14
CSF1PO*15
2n
TPOX*8
TPOX*9
TPOX*10
TPOX*11
TPOX*12
2n
TH01*6
TH01*7
TH01*8
TH01*9
TH01*9.3
TH01*10
2n
F13A01*3.2
F13A01*4
F13A01*5
F13A01*6
F13A01*7
F13A01*8
F13A01*11
F13A01*13
F13A01*14
F13A01*15
F13A01*16
F13A01*17
F13A01*18
2n
FES/FPS*8
FES/FPS*9
FES/FPS*10
FES/FPS*11
FES/FPS*12
FES/FPS*13
FES/FPS*14
2n
VWA31/A*13
VWA31/A*14
VWA31/A*15
VWA31/A*16
VWA31/A*17
VWA31/A*18
VWA31/A*19
VWA31/A*20
2n
1
BISCAY
ALAVA
NAVARRE
GOI
DEB
TOL
URO
URI
DUR
ARR
MAR
LEA
GER
LLA
RIO
MON
VAL
BID
IRU
EBR
0.000
0.000
0.030
0.290
0.280
0.370
0.020
0.010
0.000
100
0.600
0.050
0.120
0.210
0.020
100
0.163
0.122
0.194
0.153
0.367
0.000
98
0.050
0.050
0.160
0.230
0.490
0.000
0.000
0.000
0.000
0.000
0.020
0.000
0.000
100
0.000
0.000
0.390
0.370
0.180
0.050
0.010
100
0.000
0.074
0.128
0.170
0.394
0.160
0.074
0.000
94
0.000
0.000
0.010
0.250
0.410
0.290
0.030
0.000
0.010
100
0.430
0.090
0.070
0.350
0.060
100
0.240
0.160
0.150
0.200
0.250
0.000
100
0.040
0.050
0.150
0.310
0.420
0.000
0.000
0.010
0.000
0.000
0.010
0.010
0.000
100
0.010
0.000
0.290
0.340
0.250
0.110
0.000
100
0.010
0.050
0.180
0.200
0.310
0.190
0.060
0.000
100
0.000
0.000
0.020
0.320
0.360
0.270
0.030
0.000
0.000
100
0.520
0.110
0.080
0.270
0.020
100
0.180
0.140
0.110
0.140
0.430
0.000
100
0.020
0.020
0.180
0.300
0.420
0.010
0.000
0.000
0.000
0.010
0.030
0.000
0.010
100
0.000
0.000
0.400
0.310
0.210
0.070
0.010
100
0.000
0.143
0.082
0.286
0.316
0.112
0.051
0.010
98
0.000
0.000
0.030
0.270
0.340
0.320
0.010
0.000
0.030
100
0.460
0.070
0.110
0.300
0.060
100
0.270
0.120
0.110
0.180
0.320
0.000
100
0.010
0.040
0.140
0.280
0.500
0.010
0.000
0.000
0.000
0.000
0.020
0.000
0.000
100
0.000
0.000
0.430
0.290
0.190
0.090
0.000
100
0.000
0.140
0.110
0.210
0.330
0.170
0.040
0.000
100
0.000
0.000
0.070
0.290
0.270
0.320
0.050
0.000
0.000
100
0.510
0.100
0.010
0.310
0.070
100
0.300
0.150
0.120
0.170
0.260
0.000
100
0.040
0.050
0.210
0.280
0.350
0.000
0.010
0.010
0.010
0.010
0.030
0.000
0.000
100
0.030
0.000
0.290
0.430
0.200
0.050
0.000
100
0.010
0.130
0.140
0.270
0.250
0.150
0.030
0.020
100
0.000
0.000
0.000
0.210
0.400
0.310
0.070
0.010
0.000
100
0.450
0.110
0.050
0.310
0.080
100
0.270
0.180
0.110
0.210
0.230
0.000
100
0.028
0.047
0.142
0.377
0.340
0.019
0.000
0.000
0.009
0.009
0.009
0.019
0.000
106
0.009
0.000
0.349
0.321
0.255
0.066
0.000
106
0.000
0.132
0.170
0.292
0.217
0.123
0.066
0.000
106
0.009
0.010
0.000
0.140
0.290
0.480
0.080
0.000
0.000
100
0.430
0.190
0.060
0.270
0.050
100
0.360
0.090
0.100
0.170
0.270
0.010
100
0.030
0.040
0.220
0.330
0.360
0.000
0.000
0.000
0.010
0.010
0.000
0.000
0.000
100
0.020
0.000
0.380
0.380
0.130
0.090
0.000
100
0.000
0.080
0.180
0.180
0.310
0.150
0.100
0.000
100
0.000
0.000
0.009
0.250
0.268
0.438
0.027
0.000
0.000
112
0.420
0.107
0.196
0.223
0.054
112
0.313
0.089
0.116
0.232
0.241
0.009
112
0.038
0.058
0.115
0.394
0.375
0.000
0.000
0.000
0.000
0.000
0.019
0.000
0.000
104
0.019
0.000
0.377
0.274
0.264
0.066
0.000
106
0.000
0.142
0.094
0.321
0.283
0.075
0.085
0.000
106
0.000
0.000
0.010
0.320
0.280
0.350
0.030
0.000
0.010
100
0.450
0.120
0.080
0.240
0.110
100
0.260
0.100
0.100
0.210
0.300
0.030
100
0.028
0.019
0.132
0.321
0.491
0.009
0.000
0.000
0.000
0.000
0.000
0.000
0.000
106
0.010
0.000
0.356
0.298
0.212
0.125
0.000
104
0.000
0.106
0.115
0.212
0.337
0.125
0.106
0.000
104
0.000
0.010
0.010
0.180
0.350
0.430
0.020
0.000
0.000
100
0.550
0.090
0.040
0.240
0.080
100
0.300
0.140
0.080
0.150
0.320
0.010
100
0.020
0.040
0.230
0.360
0.320
0.010
0.000
0.000
0.000
0.000
0.020
0.000
0.000
100
0.000
0.000
0.340
0.420
0.230
0.010
0.000
100
0.000
0.130
0.090
0.310
0.260
0.160
0.050
0.000
100
0.000
0.000
0.029
0.231
0.404
0.298
0.038
0.000
0.000
104
0.509
0.028
0.066
0.349
0.047
106
0.202
0.087
0.173
0.183
0.346
0.010
104
0.051
0.031
0.112
0.367
0.378
0.000
0.000
0.010
0.010
0.020
0.010
0.010
0.000
98
0.020
0.010
0.408
0.245
0.235
0.071
0.010
98
0.000
0.098
0.127
0.235
0.324
0.157
0.049
0.010
102
0.000
0.000
0.019
0.255
0.472
0.170
0.085
0.000
0.000
106
0.548
0.038
0.067
0.279
0.067
104
0.224
0.122
0.133
0.214
0.306
0.000
98
0.031
0.061
0.112
0.316
0.439
0.000
0.000
0.010
0.000
0.020
0.010
0.000
0.000
98
0.000
0.000
0.367
0.418
0.133
0.071
0.010
98
0.000
0.173
0.092
0.224
0.276
0.143
0.082
0.010
98
0.000
0.000
0.024
0.256
0.317
0.305
0.098
0.000
0.000
82
0.575
0.063
0.087
0.262
0.013
80
0.256
0.128
0.115
0.179
0.321
0.000
78
0.092
0.000
0.132
0.368
0.382
0.000
0.000
0.000
0.013
0.013
0.000
0.000
0.000
76
0.000
0.000
0.372
0.282
0.256
0.077
0.013
78
0.000
0.077
0.192
0.218
0.244
0.128
0.128
0.013
78
0.000
0.013
0.000
0.158
0.461
0.329
0.039
0.000
0.000
76
0.527
0.081
0.054
0.297
0.041
74
0.257
0.081
0.041
0.162
0.432
0.027
74
0.044
0.015
0.132
0.382
0.382
0.000
0.000
0.000
0.000
0.015
0.000
0.029
0.000
68
0.000
0.000
0.441
0.250
0.294
0.015
0.000
68
0.000
0.114
0.157
0.229
0.357
0.100
0.043
0.000
70
0.000
0.000
0.016
0.323
0.355
0.258
0.048
0.000
0.000
62
0.677
0.048
0.048
0.210
0.016
62
0.242
0.145
0.097
0.226
0.290
0.000
62
0.000
0.048
0.113
0.419
0.403
0.000
0.000
0.016
0.000
0.000
0.000
0.000
0.000
62
0.000
0.000
0.371
0.371
0.145
0.113
0.000
62
0.000
0.032
0.097
0.274
0.323
0.210
0.048
0.016
62
0.000
0.000
0.000
0.347
0.333
0.264
0.056
0.000
0.000
72
0.542
0.097
0.111
0.208
0.042
72
0.306
0.111
0.069
0.306
0.208
0.000
72
0.014
0.042
0.125
0.333
0.431
0.000
0.000
0.028
0.014
0.014
0.000
0.000
0.000
72
0.000
0.000
0.278
0.347
0.292
0.083
0.000
72
0.000
0.069
0.222
0.278
0.222
0.139
0.069
0.000
72
0.000
0.000
0.000
0.355
0.403
0.177
0.065
0.000
0.000
62
0.500
0.097
0.032
0.371
0.000
62
0.210
0.113
0.145
0.226
0.306
0.000
62
0.032
0.000
0.210
0.290
0.435
0.000
0.000
0.000
0.016
0.016
0.000
0.000
0.000
62
0.000
0.000
0.290
0.339
0.226
0.145
0.000
62
0.000
0.048
0.161
0.306
0.258
0.210
0.016
0.000
62
2n, chromosome number.
77
ANCIENT BARRIERS TO GENE FLOW IN BASQUES
TABLE 2. Exact test probabilities for population differentiation and F (FIT),␪ (FST), and ƒ (FIS) values for each system1
System
Exact test probabilities
F ⫾ SE
P
␪ ⫾ SE
P
f ⫾ SE
P
HUMCSF1PO
HUMTPOX
HUMTH01
HUMF13A01
HUMFES/FPS
HUMVWA31/A
Total
0.0001 ⴞ 0.0001
0.0003 ⴞ 0.0001
0.2252 ⫾ 0.0047
0.5375 ⫾ 0.0084
0.0630 ⫾ 0.0031
0.1550 ⫾ 0.0045
Significant
⫺0.022 ⫾ 0.017
0.024 ⫾ 0.019
⫺0.023 ⫾ 0.015
0.045 ⴞ 0.020
⫺0.010 ⫾ 0.024
⫺0.013 ⫾ 0.014
⫺0.001 ⫾ 0.011
0.831
0.135
0.894
0.021
0.659
0.769
N.S.
0.013 ⴞ 0.020
0.006 ⴞ 0.003
0.003 ⫾ 0.003
0.000 ⫾ 0.002
0.003 ⫾ 0.002
0.002 ⫾ 0.001
0.004 ⴞ 0.002
<0.001
0.013
0.093
0.415
0.150
0.194
Significant
⫺0.035 ⫾ 0.018
0.018 ⫾ 0.018
⫺0.026 ⫾ 0.016
0.045 ⴞ 0.020
⫺0.012 ⫾ 0.025
⫺0.015 ⫾ 0.014
⫺0.005 ⫾ 0.012
0.942–0.930
0.225–0.198
0.925–0.910
0.028–0.023
0.722–0.695
0.812–0.785
N.S.
1
Significant values according to Bonferroni’s criterion are in bold type.
TABLE 3. Genetic distance and covariation between districts1
GOI
DEB
TOL
URO
URI
DUR
ARR
MAR
LEA
GER
LLA
RIO
MON
VAL
BID
IRU
EBR
1
GOI
DEB
TOL
URO
URI
DUR
ARR
MAR
LEA
GER
LLA
RIO
MON
VAL
BID
IRU
EBR
140
⫺9
28
23
⫺16
⫺51
⫺25
⫺7
⫺6
⫺14
4
2
10
⫺30
11
⫺37
⫺24
172
81
⫺33
14
9
11
⫺5
⫺23
0
⫺27
0
⫺10
⫺19
⫺20
⫺7
10
27
108
135
113
18
⫺6
⫺22
⫺47
⫺14
⫺12
4
0
12
⫺7
1
⫺2
⫺42
8
100
82
69
116
⫺19
⫺22
⫺26
9
34
⫺8
⫺11
⫺4
⫺42
⫺19
⫺13
⫺30
⫺21
177
99
131
127
190
⫺4
6
⫺25
⫺32
23
⫺6
0
⫺19
⫺68
⫺9
⫺17
⫺10
257
064
139
121
89
105
10
4
⫺12
20
⫺12
⫺4
⫺10
16
⫺34
9
⫺5
233
163
241
161
140
149
161
25
23
36
⫺37
⫺46
0
8
⫺41
⫺14
⫺29
241
178
184
132
173
113
155
173
31
12
⫺16
⫺41
⫺12
⫺11
⫺43
14
⫺75
116
92
97
44
121
123
141
99
103
⫺10
⫺22
⫺25
⫺19
⫺5
⫺21
⫺3
⫺24
205
163
145
160
79
97
120
146
165
114
⫺33
⫺20
⫺38
30
⫺13
⫺31
⫺46
147
79
79
71
154
92
212
161
112
159
107
14
17
33
⫺7
⫺33
2
184
122
113
117
149
129
280
269
160
195
113
109
3
⫺20
28
⫺3
5
139
103
102
109
119
84
170
137
97
136
70
139
110
3
⫺2
13
11
228
170
117
133
241
141
227
209
177
158
72
193
118
182
⫺35
⫺37
⫺30
162
172
142
161
164
180
283
250
159
176
144
122
117
230
125
33
30
239
117
195
164
129
108
250
154
119
198
190
176
102
258
141
134
34
232
79
122
147
132
139
306
293
163
242
129
132
145
250
157
133
146
Genetic distance values ⫻ 104 are above diagonal; rij and rii ⫻ 104 are below and along diagonal, respectively. Diagonal, in bold.
combining probabilities obtained previously (Fisher,
1970), and a significant value was also obtained
(␹2n12 df ⫽ 48.2, P ⬍ 0.001). The ␪ value obtained in
this study for the districts analyzed (␪ ⫽ 0.004 ⫾
0.002) is also significant, and is slightly higher than
that obtained on the basis of classical polymorphisms in 14 Basque districts (FST ⫽ 0.003 ⫾ 0.001;
Manzano et al., 2002).
We analyzed the structure of the genetic heterogeneity observed by several methods. We calculated the
R matrix, performed hierarchical analyses of molecular variance (AMOVA), made comparisons between
matrices using simple, partial, and multiple correlations (with statistical significance obtained by means
of Mantel’s test), and used the COCOPAN method
(contiguity-constrained permutational ANOVA) to analyze the genetic structure of a population, eliminating
the effect of spatial autocorrelation.
To analyze the effect of each district on the observed genetic heterogeneity, we calculated the R
matrix (Table 3). The mean of the rii (RST) terms is
0.013. The districts with the greatest influence on
the genetic heterogeneity observed (with a higher
than average rii) are geographically scattered, and
include peripheral districts (EBR, VAL, URI, and
ARR) and also central districts of the area analyzed
(GOI and MAR). It is between these districts that
the most negative covariations are found (e.g., MAREBR, rii ⫽ ⫺0.0075 and URI-VAL, rij ⫽ ⫺0.0068),
while the most genetically similar districts are generally adjacent to one another, with the pairs GER-
ARR (rij ⫽ 0.0036), IRU-EBR (rij ⫽ 0.0034), and
LLA-VAL (rij ⫽ 0.0033) standing out, though there
are some exceptions, such as the coastal district of
LEA and URO (rij ⫽ 0.0034).
To analyze how the genetic heterogeneity observed is structured, we began by using the smallest
possible number of groups. After checking that when
all the districts are classed in two groups on any
basis the FST value of at least one of those groups is
always statistically significant (results not shown),
we distributed the districts in three groups. This
process was assessed using hierarchical analyses of
molecular variance (AMOVA), in order to obtain a
maximum variance between groups (FCT) and a minimum variance between districts within groups
(FSC). Of all possible combinations, the hierarchical
classification that best fits this criterion is the one
shown in Figure 2 (FST ⫽ 0.0050, P ⫽ 0.0000 ⫾
0.0000; FCT ⫽ 0.0045, P ⫽ 0.0000 ⫾ 0.0000; FSC ⫽
0.0005, P ⫽ 0.0003 ⫾ 0.0001). Although FSC is statistically significant, it must be pointed out that the
component FCT accounts for 90% of the total value of
FST. The classification obtained identifies one group
to the west which would include practically the
whole province of Biscay except district LEA (West
Basques; 2n ⫽ 518), a second group which includes
Guipuzcoa and almost all of Alava (Central Basques;
2n ⫽ 768), and a third group with Navarre plus the
southern Alava district of RIO (East Basques; 2n ⫽
302). The borders between these groups coincide
with the seven greatest FST genetic distances be-
78
M. IRIONDO ET AL.
Fig. 2. Three groups of Basques described in text, separated at boundaries obtained using AMOVA analysis. Thickest lines
indicate seven greatest genetic distances obtained by Delaunay’s triangulation.
tween the pairs of districts grouped in the network
built up on the basis of Delaunay’s triangulation
method (Table 3).
Since the genetic structure of the Basque population observed through AMOVA analysis may be influenced by geographical distance, in which case the
genetic heterogeneity could be artificial and caused
by the arbitrary creation of boundaries, we performed a correlation analysis and obtained statistical significance through a Mantel’s test. We compared the FST genetic distance matrix (GEN), the
geographical distance matrix (GEO), and a matrix of
distances representing the grouping described above
(GROUP). To create this last matrix we allocated a
distance of 0 to pairs of districts from the same
group, and 1 to pairs from different groups. Simple
matrix correlations are highly significant for
GEN,GEO (r ⫽ 0.384; P ⫽ 0.0024) and GEN,
GROUP (r ⫽ 0.427; P ⫽ 0.0001), so both matrices
can be seen as useful predictors of the genetic
distance matrix. Similarly, GROUP and GEO are
closely correlated (r ⫽ 0.375; P ⫽ 0.0008), so we
performed a partial correlation analysis to determine the level of correlation between the matrices
GROUP and GEO and the genetic distance matrix
when one of them remains constant. The GROUP
matrix is better at explaining the genetic distance
matrix (rGEN,(GROUP.GEO) ⫽ 0.3303; P ⫽ 0.0002)
than is geography (rGEN,(GEO.GROUP) ⫽ 0.2668; P ⫽
0.0286). Oden and Sokal (1992) recommended to
use in tests of this type a limit value of P ⫽ 0.001,
so it is only in the case of partial correlation between GROUP and GEN that we can assure statistical significance for the correlation between
residuals. Finally, multiple correlation analysis
showed that when the information from GROUP
was added to the geographical information, the
variability explained by geography alone increased by 39% (r2GEN,(GEO-GROUP) ⫽ 0.2405;
r2GEN,GEO ⫽ 0.1475).
ANCIENT BARRIERS TO GENE FLOW IN BASQUES
In any event, is it possible that spatial autocorrelation may be influencing the results for genetic
structuring. Taking into account the analyses carried out as indicated above, it is noteworthy that the
Mantel tests are appropriate only when there is no
spatial autocorrelation within the groups (Legendre
et al., 1990). To tackle this matter, we obtained the
correlograms for the 29 alleles with allele frequencies greater than 2%. We divided the area under
study into six distance classes, and observed that
when Bonferroni’s correction was applied, only the
correlogram for allele FES/FPS*12 showed a significant probability (P ⫽ 0.008). In any event, the pattern of this allele is not similar to any of those
described by Sokal et al. (1989). Therefore, in the
face of the possibility that this slight spatial autocorrelation may have influenced the result which
indicated the existence of the three genetically distinct groups, we used the COCOPAN method (contiguity-constrained permutational ANOVA), which
checks for the existence of genetic heterogeneity,
eliminating any possible autocorrelation effects. We
tested for the existence of the groups West Basques,
Central Basques, and East Basques for all alleles
with a mean frequency greater than 2%, as alleles
with lower frequencies have a relatively high standard error. The individual probabilities of the 29
alleles considered were corrected using Bonferroni’s
procedure, and significant results were detected for
one allele (CSF1P0*12; P ⫽ 0.001), enabling us to
conclude that the allele frequencies of the groups
constructed (Fig. 2) after eliminating the effect of
spatial autocorrelation differed significantly from
one another.
The three groups set up show no within-group
genetic heterogeneity (West Basques, FST ⫽ 0.003 ⫾
0.002, P ⫽ 0.053, N.S.; Central Basques, FST ⫽
0.000 ⫾ 0.001, P ⫽ 0.332, N.S.; East Basques, FST ⫽
0.001 ⫾ 0.002, P ⫽ 0.348, N.S.), but two by two
comparisons between groups provided statistically
significant values after Bonferroni’s correction was
applied (West Basques vs. Central Basques, FST ⫽
0.004 ⫾ 0.001, P ⬍ 0.001, significant; West Basques
vs. East Basques, FST ⫽ 0.003 ⫾ 0.001, P ⫽ 0.002,
significant; Central Basques vs. East Basques, FST ⫽
0.007 ⫾ 0.005, P ⬍ 0.001, significant).
To check whether the main geographical barrier
in the Basque Country (the watershed; see Fig. 1)
influences allele frequency distribution, we performed an AMOVA analysis considering two groups,
one to the north and the other to the south of the
watershed. The watershed is the biggest geographical feature in the Basque Country, and is made up
of a mountain chain running parallel to the coast
which separates the coastal area, whose rivers run
into the Bay of Biscay, from the southern area,
whose rivers run into the River Ebro and then on to
the Mediterranean. We obtained nonsignificant
variance between groups (FCT ⫽ 0.0005, P ⫽
0.2244 ⫾ 0.0033; FSC ⫽ 0.0032, P ⫽ 0.0004 ⫾
0.0002). We also set up a distance matrix to repre-
79
sent the effect of the watershed, allocating a distance of 0 to pairs of districts within the same group
and a distance of 1 to pairs from different sides of
the watershed. In this case, the resulting matrix and
the genetic distance matrix were not significantly
correlated (r ⫽ 0.099; P ⫽ 0.1508). Moreover, this
separation of districts to the north and south of the
watershed coincides practically with the current linguistic division between Basque and Spanish: in the
districts to the north of the dividing line, Basque is
spoken as a mother tongue, while to the south,
Spanish is spoken. The only exception is the IRU
district of Navarre, where the linguistic frontier
runs through the district itself. Even if the IRU
district is not considered, the AMOVA analysis continues to show a nonsignificant FCT value (FCT ⫽
0.0005, P ⫽ 0.2464 ⫾ 0.0042; FSC ⫽ 0.0032, P ⫽
0.0013 ⫾ 0.0004), as does the correlation with the
genetic distance matrix (r ⫽ 0.102, P ⫽ 0.1917).
To compare the district groupings obtained in our
study with geographically neighboring samples and
analyze their performance within the European genetic context, we went to the bibliography and selected the samples from the Iberian Peninsula and
Europe that have been analyzed to date for the six
systems used in our study. We built up a neighborjoining tree based on FST genetic distances. Figure 3
shows that the East Basques group forms a node
with the Central Basques group (44.7%), which subsequently joins up with the sample from Aragon
(55.7%) and the group of districts in Biscay (West
Basques), though in this case the bootstrap values
are lower (22.5%). Then come most of the samples
from the Iberian Peninsula and the Mediterranean
region, more or less grouped, and at the other end of
the tree are the Slavic populations (North Poland,
Slovenia, and South Poland). We found that including the three Basque samples together did not condition the resulting graph. Using any one of the
Basque samples separately did not result in any
variation in the structure of the tree (results not
shown).
DISCUSSION
This work features the first district-by-district
analysis of all provinces in the Iberian Peninsula
with an autochthonous Basque population, and confirms the existence of significant genetic heterogeneity within Basque territory. The analysis by several statistical methods of the structure of this
diversity showed that the 17 districts clustered in
three major groups: West Basques, Central Basques
and East Basques (Fig. 2).
The West Basques group includes some districts
which are geographically linked mainly to the
Ibaizabal river valley, which runs through Biscay
from east to west (Fig. 1). This river seems to have
been highly influential in the maintaining of the
population unit described. It is noteworthy that the
districts in Guipuzcoa, LEA in Biscay and almost all
the districts in Alava are grouped together in Cen-
80
M. IRIONDO ET AL.
Fig. 3. Neighbor-joining tree for European populations. Numbers represent number of times each node appears in 1,000 bootstrap
iterations.
tral Basques, even though the provinces of Alava
and Guipuzcoa were previously considered on the
basis of classical polymorphisms as the two extremes of Basque genetic variation (Manzano et al.,
1996a; Calderón et al., 1998). The analyses presented in these studies took provinces as their sampling units, and this may be why they obtained these
results: when the same geographical area and the
same genetic polymorphisms as in Manzano et al.
(1996a) were used in multivariate analyses of the
district-based study by Manzano et al. (2002), no
significant genetic differences were detected between most districts in Alava and those in Guipuzcoa. Taken along with the results of our study here,
this indicates that the administrative boundary between these two provinces does not reflect the existence of a barrier to gene flow. There is actually a
natural pass between the lowland Llanada area of
Alava and the basin of the River Deba (Fig. 1), which
is one of the main routes for contact between the
provinces of Alava and Guipuzcoa. The main Roman
roads, and later the Medieval highroads, between
the two provinces ran through this natural pass,
with all the effects on gene flow that this entails. In
regard to the East Basques group, it is interesting
that the Alava district of RIO, located by the River
Ebro, is genetically linked to the districts in Navarre, including EBR, which also lies on the Ebro.
Here the Ebro itself and its main tributaries (Fig. 1)
could be working as routes for genetic exchange.
The heterogeneity observed has a clear, powerful
structure. These results for this area of the Iberian
Peninsula raise two questions: when did this heterogeneity appear, and how has it come to persist into
our time?
There are two processes by which the genetic heterogeneity observed here could have arisen: 1)
genetic drift or 2) differential mixing with allochthonous populations. The genetic differentiation attributable to genetic drift is unlikely to have increased
ANCIENT BARRIERS TO GENE FLOW IN BASQUES
over the past 5,000 – 6,000 years, taking into account
the increase in population in the Basque Country
and in Europe as a whole during that time. Demographic data from the last few centuries show sufficient migration in areas of the territory analyzed to
prevent any diversification (Toja, 1987; Peña, 1988).
Genetic differentiation due to drift from one or more
populations in what is now known as the Basque
Country is more likely to have occurred in pre-Neolithic times.
In a study province by province using classical
polymorphisms, Manzano et al. (1996b) concluded
that most of the genetic differentiation of Basques
was between “Atlantic” and “Mediterranean” Basques,
with the two groups being separated by the watershed, which is the biggest geographical and cultural
barrier in the area analyzed. They suggested that
this differentiation could be due to different levels of
external admixture with Neolithic immigrants. Our
results here, from both AMOVA analyses and correlations, demonstrate that the watershed does not act
as a barrier to gene flow. Nor does the possibility
that differential mixing with allochthonous populations could explain the origin of the groups detected
in our study hold up in the light of the neighborjoining tree constructed (Fig. 3), at least for Central
Basques and East Basques, since these two groups
are clearly distinct from one another, but lie at a
similar distance from the remaining populations of
the Iberian Peninsula and Europe as a whole. Perhaps due to their location in the neighbor-joining
tree, West Basques could have received a greater
allochthonous gene flow, although the group’s distance from the remaining populations leads us to
doubt this: if it was true, the branch would not be as
long. It must also be said that there are no archaeological, linguistic, or cultural records in this area
which would support the idea of a greater mixing
with allochthonous populations than elsewhere in
the Basque Country. In short, the possibility of differential gene flow with allochthonous populations
being the cause of the differentiation within the
Basque population described here does not seem to
fit with the data observed.
The following is a plausible scenario: the differentiation between groups detected in this study, which
is reflected in the branch lengths of the neighborjoining tree, must have arisen at a time when the
effective population size was small enough to allow
genetic drift to have a significant effect. This would
date to pre-Mesolithic periods (11,000 YBP), because
from then onward there was an expansion of areas of
occupation in the Basque Country, and the land
south of the watershed (Fig. 1) was repopulated
after standing empty practically throughout the Upper Paleolithic. This process of expansion, the diversification of subsistence patterns, and improvements
in the climate led to a significant increase in population (de la Rúa, 1995). The data obtained in our
study suggest that Alava was populated from the
area which is now Guipuzcoa, and Navarre from the
81
northern area of the province. From then on, the
population of what is now the Basque Country became settled in three major groups (West Basques,
Central Basques, and East Basques), structured respectively around the River Ibaizabal, the River
Deba corridor, and the River Ebro and its tributaries, all of which appear to have played a substantial
role in maintaining each population unit. One factor
which could have contributed to the maintaining of
differences is the way in which the population analyzed was structured: in small population units typical of so-called “mountain cultures,” where territories are organized on the basis of small towns,
villages, and farmsteads scattered through the valleys (Comas d’Argemir, 1995).
Moreover, although the boundaries between
Basque tribes described (Caro-Baroja, 1990) (Fig. 4)
do not exactly coincide with those found in our
study, the structuring of the districts into three
groups could be related to the situation described in
Roman times, in which there were several tribes in
the area which is now Basque Country. In this regard, Manzano et al. (2002) indicated on the basis of
classical polymorphisms that they might have encountered the residue of an ancient tribal differentiation, although it was not statistically significant.
Exact boundaries between tribes in the Basque
Country in Roman times were proposed by SánchezAlbornoz (1929), but objective data (Roman writings) which led to the establishment of these exact
boundaries do not seem sufficient to obtain such
precision, and it is likely that Sánchez-Albornoz
(1929) used other data (geographical data such as
mountain ranges and rivers, cultural data such as
current dialect areas, and administrative data such
as boundaries of modern provinces and regions) to
draw up his map of the tribes of the 1st century AD.
Some authors have called into question the exact
boundaries between tribes proposed (Garcı́a de Cortazar and Lorenzo, 1994). It is possible that the
genetic heterogeneity detected in our study might be
related to the distribution of ancient tribes. The
same genetic heterogeneity could well be behind the
emergence and maintaining of Basque dialects (Fig.
4): some barrier to gene flow is clearly required in
both cases. The existence of tribes and dialects with
boundaries which coincide to some extent, which
may be related at origin and which are quite similar
to the boundaries detected in this work, could well
reflect the genetic structure of Basques since preMesolithic periods.
The STR systems used in this study reveal a major
differentiation of Basque populations from other European populations. The location on the neighborjoining tree of the population of Aragon (located
downstream on the River Ebro) among the Basque
series, with high bootstrap values (Fig. 3), seems to
show that the Ebro valley is a route for genetic
exchange. In this regard, Hurles et al. (1999) obtained the world’s highest frequencies of haplogroup
22 in populations of the Basque Country (f ⫽
82
M. IRIONDO ET AL.
Fig. 4. Tribal divisions (separated by thick lines) during Roman period (2000 YBP; Caro-Baroja, 1990) and geographical locations
of extant Basque dialects (Zuazo, 1998).
0.111 ⫾ 0.029) and Catalonia (f ⫽ 0.219 ⫾ 0.073),
and suggested that this was indicative of male gene
flow between the upper reaches of the Ebro and the
area of the coast where the river meets the Mediterranean, though they did not determine the point of
origin of the haplogroup. The point on the neighborjoining tree at which our study places the population
of Aragon (among the Basque populations) is congruent with a migration of haplogroup 22 from the
Basque Country to the Mediterranean coast along
the Ebro valley. Moreover, Aragon being an area
where there is no record of Basque or any related
dialect being spoken at any period of history, this
result also suggests that the population expansion
processes were not accompanied by the spread of
Basque cultural characteristics. Since there is no
record in any other area of science to indicate migratory movements from the Basque Country towards the Mediterranean, the process involved must
be one of continuous gene flow over at least the last
3,000 years, as that is the estimated age of haplogroup 22 (Hurles et al., 1999).
The results of our study in terms of intrapopulation heterogeneity and the relationship between the
Basque populations described and the rest of the
populations of Europe enable us to tackle the matter
of the origin of the current European gene pool. For
much of the Upper Paleolithic, the populations of
Europe must have been small and, in many cases,
isolated from one another, due to their form of subsistence or their being forced to fall back to glacial
refuges during the coldest periods (Otte, 1990). During that period there must have been widespread
genetic population diversification in Europe as a
result of processes such as genetic drift and bottlenecks. As indicated previously, the genetic diversification detected in the Basque population could be
an example of this.
ANCIENT BARRIERS TO GENE FLOW IN BASQUES
The clear differentiation between Basque populations and those of nearby areas such as Aragon (and
probably also other known genetic outliers such as
Lapps, Sardinians, or Icelanders, which have not
been considered in our analysis) from the genetically
more homogenous rest of Europe is not congruent
with the process of widespread genetic diversification in the Paleolithic described above, unless there
was some subsequent process which resulted in the
homogenization of the allele frequencies of most European populations. This process would have had to
be caused by populations with a mainly European
genetic origin, as indicated by Richards et al. (1996)
and Semino et al. (2000), among others. Moreover, it
would have had to be able to form the European
continental clines which are observed for numerous
alleles (Sokal et al., 1989; Semino et al., 1996;
Chikhi et al., 1998), and to maintain the genetic
idiosyncrasy of the current Basque population.
The microevolutionary process required to meet
the above conditions is one of Neolithization by populations carrying mtDNA and Y-chromosome haplotypes of European Upper Paleolithic origin. A plausible scenario for this is the migration of the first
farmers from the Middle East to Greece around
9,000 YBP, travelling by sea and probably only in
small numbers. In this area, a slow increase is detected in the number of Neolithic settlements over
approximately 1,500 years (van Andel and Runnels,
1995). During that time it is highly likely that the
Mesolithic populations of the area would have mixed
with the Neolithics and become Neolithized. This
process would have resulted in a gene pool at the
time of population saturation which was very different from that which existed at the beginning of
colonization. At the time of saturation, the region
would then become a Neolithic jumping-off point
(van Andel and Runnels, 1995). As these authors
pointed out, similar processes seem to have taken
place in later stages of the migration, in the Balkans
and in Italy. The expansion of the Neolithic throughout Europe would thus have taken place later than
these initial states of colonization by immigrants
from the Middle East. Thus, our results, the European continent-wide allele gradients, and the Upper
Paleolithic origin of Y-chromosome and mtDNA haplotypes would be compatible with a process of demic
diffusion associated with agriculture.
The genetic results seem to indicate that the
Basque population would have been less affected by
these agriculture-related demic movements. This
must have been conditioned by various factors in the
natural environment which were highly influential
in the meeting of the basic needs of the autochthonous populations and, in short, in the balance between human groups and their environment (Cava,
1990). In the area analyzed in our study, archaeological information shows a long period of transition
between the Epi-Paleolithic and the full Neolithic
(Martı́, 1998). The fact that a genetic structure probably rooted in pre-Mesolithic times can still be seen
83
in the Basque population is indicative of the processes described here.
ACKNOWLEDGMENTS
We are indebted to those Basque students who
participated generously in the development of this
genetic study.
LITERATURE CITED
Aguirre A, Vicario A, Mazón LI, Estomba A, Martı́nez de
Pancorbo M, Arrieta Picó V, Perez Elortondo F, Lostao MC.
1991. Are the Basques a single and a unique population? Am J
Hum Genet 49:450 – 458.
Andrés MI, Prieto V, Flores IC, Sanz P. 1996. Population genetics
of three STRs: TH01, CSF1PO and TPOX in Southern Spain.
In: Carracedo A, Brinkmann B, Bär W, editors. Advances in
forensic haemogenetics, volume 6. Berlin: Springer. p 489 – 491.
Bassam BJ, Caetano-Anolles G, Gresshoff PM. 1991. Fast and
sensitive silver staining of DNA in polyacrylamide gels. Anal
Biochem 196:80 – 83.
Calderón R, Vidales C, Peña JA, Perez-Miranda A, Dugoujon JM.
1998. Immunoglobulin allotypes (GM and KM) in Basques from
Spain: approach to the origin of the Basque population. Hum
Biol 70:667– 698.
Caro-Baroja J. 1990. Materiales para una historia de la lengua
vasca en su relación con la latina. San Sebastián: Txertoa.
Cava A. 1990. El neolı́tico en el Paı́s Vasco. Munibe (Antropol
Arkeol) 42:97–106.
Cavalli-Sforza LL, Menozzi P, Piazza A. 1994. The history and
geography of human genes. Princeton, NJ: Princeton University Press.
Chikhi L, Destro-Bisol G, Pascali V, Baravelli V, Dobosz M,
Barbujani G. 1998. Clinal variation in the nuclear DNA of
Europeans. Hum Biol 70:643– 657.
Comas d’Argemir D. 1995. La definición de la cultura pirenaica.
In: Bertranpetit J, Vives E, editors. Muntanyes i població. El
passat del Pirineus des d’una prespectiva multidisciplinària.
Andorra la Vella: Ministeri de Relacions Exteriors, Centre de
Trobada de las Cultures Pirinenques. p 333–346.
Constans J, Viau M. 1975. Distribution of haptoglobin subtypes
in French Basques. Hum Hered 25:156 –159.
Crespillo M, Luque JA, Ramı́rez E, Garcı́a P, Fernández RM,
Valverde JL. 1996. Allele frequencies of four STRs (HUMTH01,
HUMVWFA31, HUMF13A01, HUMFESFPS) in north-east of
Spain. In: Carracedo A, Brinkmann B, Bär W, editors. Advances in forensic haemogenetics, volume 6. Berlin: Springer. p
520 –522.
de la Rúa C. 1995. La historia del poblamiento del Paı́s Vasco
desde una perspectiva antropológica. In: Bertranpetit J, Vives
E, editors. Muntanyes i població. El passat del Pirineus des
d’una prespectiva multidisciplinària. Andorra la Vella: Ministeri de Relacions Exteriors, Centre de Trobada de las Cultures
Pirinenques. p 301–316.
Dow MM, Cheverud JM. 1985. Comparison of distance matrices
in studies of population structure and genetic microdifferentiation: quadratic assignment. Am J Phys Anthropol 68:367–373.
Excoffier L, Smouse PE, Quattro JM. 1992. Analysis of molecular
variance inferred from metric distances among DNA haplotypes: application to human mitochondrial DNA restriction
data. Genetics 131:479 – 491.
Felsenstein J. 1989. PHYLIP—phylogeny inference package (version 3.2). Cladistics 5:164 –166.
Fisher RA. 1970. Statistical methods for research workers, 14th
ed. Edinburgh: Oliver & Boyd.
Gamero JJ, Romero JL, Souto L, Vizcaya MA, Arufe ML, Feliu
MM, Vieira DN, Vide MC. 1998. Population study of the HUMVWA, HUMTH01, HUMFES and HUMF13A1 STR polymorphisms in the south-west of spain. In: Olaisen B, Brinkmann B,
Lincoln PJ, editors. Progress in forensic genetics, volume 7.
Amsterdam: Elsevier Science. p 366 –368.
Garcı́a de Cortazar JA, Lorenzo JM. 1994. Historia del Paı́s
Vasco. Donostia: Editorial Txertoa.
84
M. IRIONDO ET AL.
Goedde HW, Hirth L, Benkmann H-G, Pellicer A, Pellicer T,
Stahn M, Singh S. 1972. Population genetic studies of red cell
enzyme polymorphisms in four Spanish populations. Hum
Hered 22:552–560.
Goedde HW, Hirth L, Benkmann H-G, Pellicer A, Pellicer T,
Stahn M, Singh S. 1973. Population genetic studies of serum
protein polymorphisms in four Spanish populations. Hum
Hered 23:135–146.
Goudet J. 1995. FSTAT (version 1.2): a computer program to
calculate F-statistics. J Hered 86:485– 486.
Guo SW, Thompson EA. 1992. Performing the exact test of Hardy-Weinberg proportion for multiple alleles. Biometrics
48:361–372.
Harpending H, Jenkins T. 1973. Genetic distance among southern African populations. In: Crawford MH, Workman PL, editors. Methods and theories of anthropological genetics. Albuquerque: University of New Mexico Press. p 177–199.
Hurles ME, Veitia R, Arroyo E, Armenteros M, Bertranpetit J,
Pérez-Lezaun A, Bosch E, Shlumukova M, Cambon-Thomsen
A, McElreavey K, López de Munain A, Röhl A, Wilson IJ, Singh
L, Pandya A, Santos FR, Tyler-Smith C, Jobling MA. 1999.
Recent male-mediated gene flow over a linguistic barrier in
Iberia, suggested by analysis of a Y-chromosomal DNA polymorphism. Am J Hum Genet 65:1437–1448.
Iriondo M, de la Rúa C, Barbero MC, Aguirre A, Manzano C.
1999. Analysis of 6 short tandem repeat loci in Navarre (northern Spain). Hum Biol 71:43–54.
Legendre P, Vaudor A. 1991. The R package: multidimensional
analysis, spatial analysis. Montreal: Département de Sciences
Biologiques, Université de Montréal. 142 p.
Legendre P, Oden NL, Sokal RR, Vaudor A, Kim J. 1990. Approximate analysis of variance of spatially autocorrelated regional
data. J Class 7:53–75.
Manzano C, Aguirre AI, Madoz P, Ribó G, Osaba L, Moreno P, de
la Rúa C. 1993a. New contribution to the genetics of the
Basques: heterogeneity in the esterase D subtype distribution.
Hum Hered 43:219 –222.
Manzano C, Moral P, de la Rúa C, Moreno P. 1993b. Serum
protein polymorphisms (GC, TF, and PI subtypes) in the
Basque population of Alava. Hum Hered 43:121–125.
Manzano C, Aguirre AI, Iriondo M, Martı́n M, Osaba L, de la Rúa
C. 1996a. Genetic polymorphisms of the Basques from Gipuzkoa: genetic heterogeneity of the Basque population. Ann Hum
Biol 23:285–296.
Manzano C, Orue JM, de la Rúa C. 1996b. The “Basqueness” of
the Basques of Alava: a reappraisal from a multidisciplinary
perspective. Am J Phys Anthropol 99:249 –258.
Manzano C, de la Rúa C, Iriondo M, Mazón LI, Vicario A, Aguirre
A. 2002. Structuring the genetic heterogeneity of the Basque
population: a view from classical polymorphisms. Hum Biol
74:51–74.
Marquer P. 1963. Contribution a l’étude anthropologique du
peuple basque at au probleme de ses origines raciales. Bull Soc
Anthropol Paris 4: 240 p.
Martı́ B. 1998. El Neolı́tico. In: Barandiarán I, Martı́ B, del
Rincón MA, Maya JL, editors. Prehistoria de la Penı́nsula
Ibérica. Barcelona: Editorial Ariel. p 121–195.
Miscicka-Sliwka D, Czarny J, Grzybowski T, Wozniak M. 1998.
Population genetics of 14 STRs: vWA, TH01, TPOX, CSF1PO,
D5S818, D13S317, D7S820, D16S539, F13A1, FES/FPS, F13B,
LPL, D3S1358, and FGA in the Pomerania-Kujawy region of
Poland. In: Olaisen B, Brinkmann B, Lincoln PJ, editors.
Progress in forensic genetics, volume 7. Amsterdam: Elsevier
Science. p 261–263.
Moran PAP. 1950. Notes on continuous stochastic phenomena.
Biometrika 37:17–23.
Nei M. 1978. Estimation of average heterozygosity and genetic
distance from a small number of individuals. Genetics 89:583–
590.
Oden NL, Sokal RR. 1992. An investigation of 3-matrix permutation tests. J Class 9:275–290.
Otte M. 1990. In: Soffer O, Gamble C, editors. The world at 18.000
BP, volume 1. London: Unwin Hyman. p 54 – 68.
Pawlowski R, Maciejewska A, Paszkowska R, Welz A. 1997. Frequencies for five short tandem repeat (STR) systems in a population from north Poland. Int J Legal Med 110:10 –13.
Peña JA. 1988. Estructura demográfica y genética de la población
del Valle de Orozco (Vizcaya). Siglos XVI–XX. Ph.D. thesis,
University of the Basque Country.
Pérez-Lezaun A, Calafell F, Mateu E, Comas D, Ruiz-Pacheco R,
Bertranpetit J. 1997. Microsatellite variation and the differentiation of modern humans. Hum Genet 99:1–7.
Raymond M, Rousset F. 1995. GENEPOP (version 1.2): population genetics software for exact tests and ecumenicism. J Hered
86:248 –249.
Reynolds J, Weir BS, Cockerham CC. 1983. Estimation of the
coancestry coefficient: basis for a short-term genetic distance.
Genetics 105:767–779.
Richards M, Corte-Real H, Forster P, Macaulay V, WilkinsonHerbots H, Demaine A, Papiha S, Hedges R, Bandelt H-J,
Sykes B. 1996. Paleolithic and Neolithic lineages in the European mitochondrial gene pool. Am J Hum Genet 59:185–203.
Rohlf FJ. 1988. NTSYS-pc: numerical taxonomy and multivariate
analysis system for the IBM PC microcomputer (and compatible), version 1.40. New York: Oxford University Press.
Sánchez-Albornoz C. 1929. Divisiones tribales y administrativas
del solar del reino de Asturias en la época romana. Bol Real
Acad Hist 95:315–395.
Santos SMM, Budowle B, Smerick JB, Keys KM, Moretti TR.
1996. Portuguese population data on the six short tandem
repeat loci—CSF1PO, TPOX, TH01, D3S1358, VWA and FGA.
Forensic Sci Int 83:229 –235.
Schneider S, Kueffer J-M, Roessli D, Excoffier L. 1997. Arlequin
version 1.1: a software for population genetic data analysis. Geneva: Genetics and Biometry Laboratory, University of Geneva.
Semino O, Passarino G, Brega A, Fellous M, Santachiara-Benerecetti AS. 1996. A view of the Neolithic demic diffusion in
Europe through two Y chromosome-specific markers. Am J
Hum Genet 59:964 –968.
Semino O, Passarino G, Oefner PJ, Lin AA, Arbuzova S, Beckman
LE, De Benedictis G, Francalacci P, Kouvatsi A, Limborska S,
Marcikiæ M, Mika A, Mika B, Primorac D, Santachiara-Benerecetti AS, Cavalli-Sforza LL, Underhill PA. 2000. The genetic
legacy of Paleolithic Homo sapiens sapiens in extant Europeans: a Y chromosome perspective. Science 290:1155–1159.
Shriver MD, Jin L, Boerwinkle E, Deka R, Ferrell RE, Chakraborty
R. 1995. A novel measure of genetic distance for highly polymorphic tandem repeat loci. Mol Biol Evol 12:914–920.
Smith JC, Anwar R, Riley J, Jenner D, Markham AF. 1990.
Highly polymorphic minisatellite sequences: allele frequencies
and mutation rates for five locus-specific probes in a Caucasian
population. J Forensic Sci Soc 30:19 –32.
Smouse PE, Long JC, Sokal RR. 1986. Multiple regression and
correlation extensions of the Mantel test of matrix correspondence. Syst Zool 35:627– 632.
Sokal RR, Oden NL. 1978. Spatial autocorrelation in biology 1.
Methodology. Biol J Linn Soc 10:199 –228.
Sokal RR, Harding RM, Oden NL. 1989. Spatial patterns of
human gene frequencies in Europe. Am J Phys Anthropol 80:
267–294.
Takezaki N, Nei M. 1996. Genetic distances and reconstruction of
phylogenetic trees from microsatellite DNA. Genetics 144:389–399.
Toja DI. 1987. Estructura matrimonial de las poblaciones de los
valles pirenaicos. Ph.D. thesis, Universitat de Barcelona.
van Andel TH, Runnels CN. 1995. The earliest farmers in Europe.
Antiquity 69:481–500.
Walsh PS, Metzger DA, Higuchi R. 1991. Chelex 100 as a medium
for simple extraction of DNA for PCR-based typing from forensic material. Biotechniques 10:506 –513.
Wartenberg D. 1989. SAAP. A spatial autocorrelation analysis
program. Setauket, NY: Exeter.
Weir BS. 1996. Genetic data analysis II. Sunderland, MA: Sinauer Associates, Inc.
Weir BS, Cockerham CC. 1984. Estimating F-statistics for the
analysis of population structure. Evolution 38:1358 –1370.
Zuazo K. 1998. Euskalkiak gaur. Fontes Linguae Vasconum 78:
191–232.
Документ
Категория
Без категории
Просмотров
0
Размер файла
323 Кб
Теги
flow, detect, basques, polymorphism, dna, genes, barriers, ancient
1/--страниц
Пожаловаться на содержимое документа