Different patterns of variation at the X- and Y-chromosome-linked microsatellite loci DXYS156X and DXYS156Y in human populations. Tatiana Karafet; Peter De Knijff; Elizabeth Wood; Jared Ragland; Andrew Clark; Michael F. Hammer.

Abstract: 

We compared the global pattern of variation at two homologous microsatellites mapping to the long arm of the X chromosome (DXYS156X) and to the short arm of the Y chromosome (DXYS156Y) in humans. A single pair of oligonucleotide primers amplifies these two nonallelic loci, each of which contains polymorphism in the number of pentanucleotide units. We observed 11 alleles in a sample of 2290 X chromosomes and 2006 Y chromosomes from 50 populations representing 6 major geographic regions. The overlapping size range of the X- and Y-chromosome alleles indicated a more complex distribution of alleles at these two loci than previously reported. Contrasting patterns of X-chromosome-linked and Y-chromosome-linked variation were reflected in statistically significant differences in genetic diversity values among geographic regions and between X and Y chromosomes. Higher levels of diversity characterized the DXYS156X locus in Africa (0.799 [+ or -] 0.004) and the DXYS156Y locus in East Asia (0.700 [+ or -] 0.006) compared with populations from other regions. These different patterns of variation can be explained by a combination of processes at both the molecular and population levels, including variable mutation rates, different effective population sizes, and genetic drift.

Full Text:

The last several years have witnessed an increasing number of human genetic studies using short tandem repeats (STRs), or microsatellites. Because of the high level of variability in repeat number at most loci, microsatellites are now widely used in forensics, linkage analysis, and population genetics. However, the specific molecular mechanisms involved in generating microsatellite variation are still enigmatic. Ultimately, we need to understand the rules governing mutation at microsatellite loci to be able to interpret patterns of variation in human populations. In this regard, a number of key observations should be considered: (1) Mutation occurs by means of a stepwise mechanism that favors very small (usually 1 unit) changes in repeat number; (2) estimated mutation rates for different microsatellite loci range from 2 x [10.sup.-5] to 1.4 x [10.sup.-2] per gamete per generation; (3) on average, mutation rates for non-disease-causing microsatellites are inversely related to their motif sizes; and (4) mutation rates can vary widely even among alleles within a single locus (Weber and Wong 1993; Talbot et al. 1995; Chakraborty et al. 1997; Heyer et al. 1997). These results suggest that sets of microsatellites with different properties may be useful for investigating population fissions of differing time depths over the course of human evolution (Hammer and Zegura 1996).

Studies of STRs on the nonrecombining portion of the Y chromosome offer the opportunity to study the effect of mutation in the absence of recombination. To date, 23 microsatellite loci have been mapped to the Y chromosome (Hammer and Zegura 1996; Jobling et al. 1997), and levels of variation appear to be similar to those described for autosomal STRs (Roewer et al. 1992; Ciminelli et al. 1995; Cooper et al. 1996). Y-chromosome-linked di-, tri-, and tetranucleotide STRs demonstrate considerable heterogeneity on global and regional scales, even between closely related populations (Ciminelli et al. 1995; Roewer et al. 1996; Deka et al. 1996). Thus far, only a single pentanucleotide STR has been found on the human Y chromosome. This polymorphic pentanucleotide repeat [(TAAAA).sub.n] occurs within a long interspersed repetitive element (LINE) that maps to a region of the short arm of the Y chromosome (Yp) that has sequence homology to a region on the long arm of the X chromosome (Xq) (Chen et al. 1994). The high degree of X-Y sequence similarity in this strictly sex-linked region is the result of a relatively recent transposition from Xq to Yp during hominid evolution (Schwartz et al. 1998). The sequence similarity is high enough that a single pair of polymerase chain reaction (PCR) primers flanking the [(TAAAA).sub.n] pentanucleotide repeat amplifies target sequences from both the X and Y chromosomes (loci DXYS156X and DXYS156Y) (Chen et al. 1994). In fact, we report here that the PCR-amplified sequences flanking the pentanucleotide repeat are identical on the X and Y chromosomes. As a result, the potential effects of differences in the flanking DNA on the mutability of the pentanucleotide array is minimized. Therefore the DXYS156 system presents an excellent opportunity to investigate the causes of differences in genetic variability associated with homologous regions on the X and Y chromosomes.

Chen et al. (1994) found 2 Y-chromosome-specific alleles (160 bp and 165 bp) in 36 unrelated CEPH (Centre d'Etudes du Polymorphisme Humaine) males (35 of 36 males had the 165-bp allele) and 5 X-chromosome-linked alleles (125 bp, 140 bp, 145 bp, 150 bp, and 155 bp) in 72 unrelated CEPH individuals. In subsequent surveys of variation in human populations from Europe, North Africa, the Middle East, and North America 3 DXYS156Y alleles (160 bp, 165 bp, and 170 bp) were reported (Salem et al. 1996; Sajantila et al. 1996; Kayser et al. 1997). To obtain a more complete picture of variation at these X- and Y-chromosome-linked pentanucleotide repeats, we performed a global survey of DXYS156X and DXYS156Y variation in 50 human populations.

Materials and Methods

Samples. The 50 human populations surveyed in this study belong to 6 major groups: Africans (Khoisan, East Bantus, Gambians, Pygmies, Egyptians), Europeans (British, Germans, Dutch, Italians, Greeks), North Asians (Altais, Forest and Tundra Nentsi, Selkups, Kets, Komi, Buryats, Evenks, Evens, Yukaghirs, Koryaks, Yakuts, Chukchi, Eskimos, Manchurian Evenks, Manchurian Oroqens), East Asians (Mongolians, Koreans, Japanese, Tibetans, Chinese, Taiwanese), Australasians (Australian Aboriginal People, Papua New Guineans), and native Americans (Eskimos, Tanana, Navajo, Cheyenne, Havasupai, Pima, Pueblo, Zapotec, Ngobe, Kuna, Embera, Wounan, Mixtec, Mixe, Inuit, Wayu). Descriptions of these samples are presented elsewhere (Kolman et al., 1995; Hammer et al. 1997; Karafet et al. 1997; Kayser et al. 1997).

PCR Analysis and DNA Sequencing. The DXYS156 locus was amplified according to the procedure of Chen et al. (1994) with the following modifications: Oligonucleotide primers were not end-labeled with 32p, and only 10 ng of genomic DNA was used as template in a 15-[[micro]liter] final volume. PCR products were mixed with loading dye and resolved on 8% polyacrylamide gels in TBE buffer for 8 hr at 300-350 V. The gel was stained with ethidium bromide, and the fragments were visualized with ultraviolet light [ILLUSTRATION FOR FIGURE 1 OMITTED].

To obtain the DNA sequences of X- and Y-chromosome-linked DXYS156 alleles, we performed PCR amplification on genomic DNAs from homozygous females and from somatic cell hybrids containing either a single human X chromosome or a single human Y chromosome. After electrophoresis, PCR fragments were isolated from polyacrylamide gels and ethanol-precipitated. DNA sequencing was performed using the AmpliCycle Sequencing Kit (Perkin-Elmer, Norwalk, Connecticut), according to the manufacturer's instructions.

Data Analysis. The band sizes of DXYS156 alleles in males and females overlapped in their distributions in some populations (e.g., Africa and East Asia) (see [ILLUSTRATION FOR FIGURE 1 OMITTED] and Table 1). Thus it was possible that some phenotypes represented more than one genotype. We employed the EM (estimation-maximization) algorithm for computing maximum-likelihood estimates from incomplete data to count DXYS156X and DXYS156Y allele frequencies (Dempster et al. 1977). The EM algorithm consists of iterations of two steps. The first is an estimation step, where genotype frequencies are estimated from the observed phenotype frequencies and the current estimates of allele frequencies. For females the estimated frequency of genotype [X.sub.i][X.sub.j] is the observed phenotype frequency. Males that are [X.sub.i][Y.sub.i] also have an unambiguous estimate of genotype frequency. The frequency of male genotype [X.sub.i][Y.sub.j] is

G([X.sub.i][Y.sub.j]) = P([X.sub.I][Y.sub.j] ([x.sub.i][y.sub.j]/[x.sub.i]/[y.sub.j] + [x.sub.j] [y.sub.i]), (1)

where P([X.sub.i][Y.sub.j]) is the frequency of the male phenotype with i and j repeats, [x.sub.i] is the allele frequency of an X chromosome with i repeats, and [y.sub.j] is the allele frequency of a Y chromosome with j repeats.

The maximization step entails calculating the maximum-likelihood estimates of allele frequencies given the estimates of genotype frequencies. Allele frequencies are calculated as though the genotype frequencies are known without ambiguity. The method of gene counting is used, and, as the name implies, the alleles are simply tallied from the genotype frequencies. Gene counting of the X and Y chromosomes gives

[TABULAR DATA FOR TABLE 1 OMITTED]

[Mathematical Expression Omitted], (2)

[y.sub.i] = [summation over j] G([X.sub.i][Y.sub.j]), (3)

where [N.sub.m] is the number of males in the sample, [N.sub.f] is the number of females in the sample, 2[N.sub.f] + [N.sub.m] is the number of X chromosomes in the sample, [1/2 [[Sigma].sub.j] G([X.sub.i][X.sub.j]) + G([X.sub.i][X.sub.i])] is the frequency of [X.sub.i] in females, and [[[Sigma].sub.j] G([X.sub.i][Y.sub.j])] is the frequency of [X.sub.i] in males.

These estimates of allele frequencies are then used to make updated estimates of genotype frequencies in the next iteration of the estimation step, and the whole iteration is repeated until convergence. All the populations converged to the maximum-likelihood estimates in 5 iterations except Africa, which required 10 iterations. Standard errors for allele frequencies were obtained by the standard method of inverting the information matrix (Elandt-Johnson 1971).

We calculated an unbiased estimate of haplotype diversity (h) equivalent to heterozygosity for each population and geographic grouping according to Eq. (8.5) of Nei (1987) with standard error as described by Nei (1978). We used Excoffier et al.'s (1992) program for analysis of molecular variance (AMOVA) and [Phi] statistics.

Results and Discussion

DXYS156 Alleles Associated with X and Y Chromosomes. In an analysis of 2290 X chromosomes and 2006 Y chromosomes from 50 populations, we observed 11 alleles ranging in size from 125 bp to 180 bp ([ILLUSTRATION FOR FIGURE 1 OMITTED] and Table 1). An allele with a size of 130 bp was not identified. Based on their initial survey of DXYS156 alleles in male and female CEPH individuals, Chen et al. (1994) found no reason to postulate overlapping sizes of X- and Y-chromosome-derived alleles. Subsequent surveys of variation in populations from Europe, the Middle East, and North Africa also assumed a nonoverlapping distribution of DXYS156X and DXYS156Y alleles (Sajantila et al. 1996; Salem et al. 1996; Kayser et al. 1997).

In contrast, our global survey revealed evidence of overlapping distributions of alleles amplified from the X and Y chromosomes. The three largest alleles (170-180 bp) were found only in males, suggesting that they were amplified only from the Y chromosome. Similarly, the three smallest alleles (125-140 bp) appeared to be X chromosome specific because they were usually present in two copies in females and in a single copy in males. Five alleles (with sizes ranging from 145 bp to 165 bp) could be derived from either the X or the Y chromosome. However, most 160-bp and 165-bp fragments appeared to be derived from Y chromosomes because they were not observed in females, and a single copy in males was usually associated with either an X-chromosome-specific allele or an allele smaller than 160 bp in length. There were only nine exceptions to this pattern: Eight males from Africa and one male from Mongolia had two copies of alleles [greater than or equal to] 160 bp in length (Table 1). Furthermore, alleles with sizes of 145 bp and 150 bp were found to be mainly derived from the X chromosome with only 10 exceptions: 9 from Africa and 1 from Mongolia.

[TABULAR DATA FOR TABLE 2 OMITTED]

DXYS156 Allele Sequences. We determined the DNA sequence of DXYS156 alleles from three X chromosomes and two Y chromosomes (data not shown). No sequence differences were observed in comparisons among all five alleles in the region flanking the pentanucleotide repeat array. Two 140-bp alleles and one 150-bp allele derived from X chromosomes contained perfect [(TAAAA).sub.7] and [(TAAAA).sub.9] repeat motifs, respectively. A 160-bp allele and a 165-bp allele derived from the Y chromosome contained 11 and 12 copies of the (TAAAA) motif, respectively. However, for the 160-bp allele the fourth repeat from the 5[prime] end of the array contained an extra A (i.e., TAAAAA). It is not known if this represents a polymorphism within 160-bp alleles.

We standardized the nomenclature for both the X and the Y alleles by naming them according to the number of copies of the pentanucleotide repeat (i.e., their size in 5-bp increments). The X-chromosome alleles were named X4 [i.e., for [(TAAAA).sub.4]] and X6-X12, and the Y-chromosome alleles were named Y8-Y15 (Tables 2 and 3).

DXYS156 Allele Frequencies. Because of the overlapping size range of the X and Y chromosome alleles, it was not possible to assign alleles to chromosomes with 100% confidence in every case. Therefore we used a maximum-likelihood approach to estimate DXYS156X and DXYS156Y allele frequencies for each population.

DXYS156Y allele frequencies estimated by the maximum-likelihood method are represented in Table 2. Most geographic regions were characterized by a high frequency of either the Y11 or Y12 allele, For example, the Y12 allele was found at very high frequencies in all native American and European populations (92% and 94%, respectively) and at moderate frequencies in North Asian and Australasian populations (64% and 69%, respectively), whereas Y11 was the most common allele in African populations (72%). East Asian populations demonstrated a different pattern: Both the Y11 [TABULAR DATA FOR TABLE 3 OMITTED] and Y12 alleles were found at similar frequencies (42% and 30%, respectively), and the otherwise rare Y13 allele was found at relatively high frequencies (17%). Furthermore, the Y14 and Y15 alleles were found only in East Asian populations. The smallest Y-chromosome-linked alleles (Y8 and Y9) were found in only nine Africans and one Mongolian.

Examination of DXYS156X alleles revealed similar frequency distributions in all geographic regions except Africa (Table 3). Although five DXYS156X alleles were found outside Africa, allele X7 was by far the most frequent in all non-African populations ([greater than]79%), followed by alleles X8 and X9. Of the 8 alleles found in African populations, 4 were present at relatively high frequencies, ranging from 18.4% to 26.5% (X9, X4, X7, and X8), and 3 were not found outside Africa (X4, X6, and X12).

Mutation Rates and Homoplasy at DXYS156Y. No mutations were detected in 42 males from 12 ancestral founding fathers in a French Canadian pedigree representing 229 independent paternal meiotic events. Therefore it was not possible to estimate an absolute mutation rate for this microsatellite. However, surveys of variation in human populations indicated both fewer numbers of alleles and lower gene diversities associated with the DXYS156Y microsatellite compared with other Y-chromosome-linked STRs (Kayser et al. 1997). Thus it is possible that DXYS156 has a lower mutation rate than microsatellites containing di-, tri-, and tetranucleotide arrays.

Figure 2 shows an evolutionary tree for six Y-chromosome haplotypes based on five unique polymorphisms in the YAP and SRY regions (Altheide and Hammer 1997; Hammer et al. 1997). Superimposed on this tree is the distribution of DXYS156Y alleles associated with these six haplotypes. All eight Y-chromosome-linked alleles (Y8-Y15) were found to be associated with haplotype 1, whereas the other five haplotypes were associated with either one (Y12) or two (Y11 and Y12) DXYS156Y alleles. This pattern of association is best explained by recurrent mutation (homoplasy) at the DXYS156Y microsatellite: At least four parallel single-step mutations are required to account for the distribution of DXYS156Y alleles on the haplotype tree. In a similar comparison involving a larger number of Y-chromosome-linked DYS19 alleles associated with each of 5 YAP haplotypes, a minimum of 15 parallel changes in allele size were required to account for the distribution of tetranucleotide repeat alleles on the haplotype tree (Hammer et al. 1997). Therefore, despite the apparently lower rate of mutation at the DXYS156Y pentanucleotide array, the same pattern of recurrent, parallel mutation characterizes both of these Y-chromosome-linked microsatellites.

Analysis of Molecular Variance. For the X chromosome the method of AMOVA (Excoffier et al. 1992) indicated that 20.4% of the total variance occurred among populations when each allele was treated as equally divergent from every other allele. When linear and squared distance measures were used, attempting to account for the fact that more similar sized alleles have more recent common ancestry, the among-population component was 16.3% and 2.6%, respectively. For the Y chromosome the DXYS156Y data gave a slightly higher among-population component of 26.3% under the equally divergent allele model. Corresponding portions of among-population variance for the Y chromosomes with linear and squared distances were 21.2% and 14.9%, respectively. Thus, by any measure, the Y chromosome exhibits greater among-population divergence than the X chromosome. This is consistent with the threefold higher effective population size of X chromosomes relative to Y chromosomes. The overall [[Phi].sub.ST] value based on the DXYS156Y data (0.149), similar to the global value reported for DYS19 (0.163), is much lower than worldwide [F.sub.ST] values based on Y-chromosome-linked point mutational and insertional polymorphisms (Hammer et al. 1997). The lower than expected among-group variances probably reflect higher mutation and convergence rates at STR loci (Ciminelli et al. 1995; Hammer et al. 1997).

Contrasting Patterns of DXYS156X and DXYS156Y Diversity. The differences in X- and Y-chromosome-linked allele frequencies were reflected in statistically significant differences in genetic diversity values, both in comparisons of each locus among geographic regions and between the X- and Y-chromosome-linked loci (Tables 2 and 3). Significantly higher levels of DXYS156Y diversity were observed in East Asian populations compared with populations from other regions (Bonferroni test, p [less than] 0.001). Unlike other global STR surveys (Bowcock et al. 1994; Deka et al. 1996; Tishkoff et al. 1996; Hammer et al. 1997), African populations were found to have relatively low levels of DXYS156Y diversity (0.492 [+ or -] 0.012), similar to values in North Asian (0.474 [+ or -] 0.001) and Australasian (0.457 [+ or -] 0.011) populations. Consistent with studies of other Y-chromosome-linked STRs (Pena et al. 1995; Santos, Gerelsaikhan et al. 1996; Hammer et al. 1997), the lowest DXYS156Y diversity values were found in native American and European populations (0.151 [+ or -] 0.011 and 0.108 [+ or -] 0.010, respectively).

In contrast to our Y-chromosome results, the diversity patterns observed at the DXYS156X locus are similar to patterns reported for autosomal microsatellites and other kinds of markers: Higher levels of within-group variation were found in African versus non-African populations (Bowcock et al. 1994; Relethford 1995; Armour et al. 1996; Tishkoff et al. 1996). This has generally been attributed to an African origin of modem humans (Armour et al. 1996; Tishkoff et al. 1996) or to larger effective population sizes in Africa (Relethford and Harpending 1995). In considering the ratio of diversity at the DXYS156X and DXYS156Y loci, we found that Europeans (3.2), Africans (1.6), and native Americans (1.1) had higher levels of X-chromosome-linked diversity, whereas East Asians (0.2), North Asians (0.4), and Australasians (0.1) had greater levels of Y-chromosome-linked diversity.

How can we explain the different patterns of variation at the DXYS156X and DXYS156Y loci? One explanation for the higher level of Y-chromosome-linked within-group variation in East Asian populations is greater antiquity of Y chromosomes in Asia (i.e., an Asian origin of human Y chromosomes). Alternatively, this pattern could be explained by a larger long-term effective population size of East Asian males. However, results of studies at other Y-chromosome-linked loci do not lend support to these hypotheses. For example, levels of within-group variation at both the DYS19 microsatellite locus and nucleotide sites within the YAP region were not found to be higher in East Asian populations (Santos, Bianchi et al. 1996; Hammer et al. 1997).

Another hypothesis posits an African-specific reduction in Y-chromosome-linked diversity resulting from reductions in effective male population size in those societies that practice high rates of polygyny. This hypothesis receives some support from restriction fragment length polymorphism (RFLP) studies at TaqI sites detected by the Y-chromosome-linked probe p49f: Lower levels of diversity were observed in Central and South African Bantu-speaking populations relative to European populations (Torroni et al. 1990; Spurdle and Jenkins 1992; Lucotte et al. 1994). However, this hypothesis does not explain the lower level of DXYS156Y diversity observed in those African populations that do not practice high rates of polygyny (data not shown) and for which high levels of within-group variation have been ascertained using other Y-chromosome-linked markers (Hammer et al. 1997).

A fourth hypothesis posits a higher mutation rate at DXYS156Y in East Asian populations. Because there is evidence of increasing mutation rates with increasing number of repeat units contained in the allele (Talbot et al. 1995; Primmer et al. 1996), this hypothesis receives support from the observations that (1) East Asian populations exhibit high frequencies of the longest DXYS156Y alleles and (2) higher frequencies of the shortest alleles are found in African populations. Under this hypothesis genetic drift may have played a role in increasing the frequency of DXYS156Y alleles with greater than 12 repeats in East Asia. A fifth hypothesis is that the DXYS156 loci exhibit different patterns because of sampling and/or evolutionary variance.

The five hypotheses are not mutually exclusive, and it is possible that different combinations of genetic, demographic, and evolutionary factors have operated at different times in different populations. Given the range of X and Y chromosome diversity in different regions, it seems plausible that sex-related activities (such as war, hunting, mating strategies, and migration) may have been involved. Interestingly, Scozzari et al. (1997) observed similar patterns of discordance between homologous X- and Y-chromosome-linked dinucleotide microsatellite loci. In Scozzari's worldwide survey of variation at the DXS8174, DXS8175, and DYS413 loci Europeans exhibited the highest level of Y-chromosome-linked within-group variation and sub-Saharan Africans showed the lowest level. In contrast, data from the two X-chromosome-linked microsatellites were consistent in showing higher levels of within-group diversity for African versus non-African populations. Direct comparisons with our data were not possible because Scozzari et al. (1997) examined only a single East Asian population.

Further worldwide studies of other X- and Y-chromosome-linked polymorphisms may help to (1) distinguish among the five hypotheses, (2) decipher the relative roles of different evolutionary forces in shaping patterns of variation on the X and Y chromosome, and (3) disentangle the relative contributions of population history and population structure to these distributional patterns.

Acknowledgments We thank F.F. Bernini and J. Jespersen for providing DNA samples and Stephen Zegura for comments on the manuscript. This publication was made possible by the National Institute of General Medical Sciences through grant GM-53566 awarded to M,F. Hammer. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the National Institutes of Health. This research was also supported by the National Science Foundation through grant OPP-9423429 awarded to M.F. Hammer.

Literature Cited

Altheide, T.K., and M.F. Hammer. 1997. Evidence for a possible Asian origin of [YAP.sup.+] Y chromosomes. Am. J. Hum. Genet. 61:462-466.

Armour, J.A., T. Anttinen, C.A. May et al. 1996. Minisatellite diversity supports a recent African origin for modern humans. Natur. Genet. 13:154-160.

Bowcock, A.M., A. Ruiz-Linares, J. Tomfohrde et al. 1994. High resolution of human evolutionary trees with polymorphic microsatellites. Nature 368:455-457.

Chakraborty, R., M. Kimmel, D.N. Silvers et al. 1997. Relative mutation rates at di-, tri-, and tetranucleotide microsatellite loci. Proc. Natl. Acad. Sci. USA 94:1041-1046.

Chen, H., W. Lowther, D. Avramopoulos et al. 1994. Homologous loci DXYS156X and DXYS156 Y contain a polymorphic pentanucleotide repeat (TAAAA)n and map to human X and Y chromosomes. Hum. Mutation 4:208-211.

Ciminelli, B., F. Pompei, P. Malaspina et al. 1995. Recurrent simple tandem repeat mutations during human Y chromosome radiation in Caucasian subpopulations. J. Molec. Evol. 41:966-976.

Cooper, G., W. Amos, D. Hoffman et al. 1996. Network analysis of human Y microsatellite haplotypes. Hum. Molec. Genet. 5:1759-1766.

Deka, R., L. Jin, M.D. Shriver et al. 1996. Dispersion of human Y chromosome haplotypes based on five microsatellites in global populations. Genome Res. 6:1177-1184.

Dempster, A.P., N.M. Laird, and D.B. Rubin. 1977. Maximum likelihood from incomplete data via the EM algorithm. J. R. Star. Soc. 39:1-38.

Elandt-Johnson, R.C. 1971. Probability Models and Statistical Methods in Genetics. New York: Wiley.

Excoffier, L., P.E. Smouse, and J.M. Quattro. 1992. Analysis of molecular variance inferred from metric distances among DNA haplotypes: Application to human mitochondrial DNA restriction data. Genetics 131:479-491.

Hammer, M.F., and S.L. Zegura. 1996. The role of the Y chromosome in human evolutionary studies. Evol. Anthropol. 5:116-134.

Hammer, M.F., A.B. Spurdle, T. Karafet et al. 1997. The geographic distribution of human Y chromosome variation. Genetics 145:787-805.

Heyer, E., J. Puymirat, P. Dieltjes et al. 1997. Estimating Y-chromosome-specific microsatellite mutation frequencies using deep rooting pedigrees. Hum. Molec. Genet. 6:799-803.

Jobling, M.A., A. Pandya, and C. Tyler-Smith. 1997. The Y chromosome in forensic analysis and paternity testing. Int. J. Legal Med. 110:118-124.

Karafet, T., S.L. Zegura, J. Vuturo-Brady et al. 1997. Y-chromosome markers and trans-Bering Strait dispersals. Am. J. Phys. Anthropol. 102:301-314.

Kayser, M., A. Caglia, D. Corach et al. 1997. Evaluation of Y-chromosomal STRs: A multi-center study. Int. J. Legal Med. 110:125-133.

Kolman, C.J., E. Bermingham, R. Cooke et al. 1995. Reduced mtDNA diversity in the Ngobe Amerinds of Panama. Genetics 140:275-283.

Lucotte, G., N. Gerard, R. Krishnamoorthy et al. 1994. Reduced variability in Y-chromosome-specific haplotypes for some Central African populations. Hum. Biol. 66:519-526.

Nei, M. 1978. Estimation of average heterozygosity and genetic distance from a small number of individuals. Genetics 89:583-590.

Nei, M. 1987. Molecular Evolutionary, Genetics. New York: Columbia University Press.

Pena, S.D.J., F.R. Santos, N.O. Bianchi et al. 1995. A major founder Y-chromosome haplotype in Amerindians. Natur. Genet. 11:15-16.

Primmer, C.R., H. Ellegren, N. Saino et al. 1996. Directional evolution in germline microsatellite mutations. Natur. Genet. 13:391-393.

Relethford, J.H. 1995. Genetics and modern human origins. Evol. Anthropol. 4:53-63.

Relethford, J.H., and H.C. Harpending. 1995. Ancient differences in population size can mimic a recent African origin of modern humans. Curr. Anthropol. 36:667-674.

Roewer, L., J. Arnemann, N.K. Spurr et al. 1992. Simple repeat sequences on the human Y chromosome are equally polymorphic as their autosomal counterparts. Hum. Genet. 89:389-394.

Roewer, L., M. Kayser, P. Dieltjes et al. 1996. Analysis of molecular variance (AMOVA) of Y-chromosome-specific microsatellites in two closely related human populations. Hum. Molec. Genet. 5:1029-1033.

Sajantila, A., A.H. Salem, P. Savolainen et al. 1996. Paternal and maternal DNA lineages reveal a bottleneck in the founding of the Finnish population. Proc. Natl. Acad. Sci. USA 93:12,035-12,039.

Salem, A.H., F.M. Badr, M.F. Gaballah et al. 1996. The genetics of traditional living: Y-chromosomal and mitochondrial lineages in the Sinai Peninsula. Am. J. Hum. Genet. 59:741-743.

Santos, F.R., N.O. Bianchi, and S.D.J. Pena. 1996. Worldwide distribution of human Y-chromosome haplotypes. Genome Res. 6:601-611.

Santos, F.R., T. Gerelsaikhan, B. Munkhtuja et al. 1996. Geographic differences in the allele frequencies of the human Y-linked tetranucleotide polymorphism DYS19. Hum. Genet. 97:309-313.

Schwartz, A., D.C. Chan, L.G. Brown et al. 1998. Reconstructing hominid Y evolution: X-homologous block, created by X-Y transposition, was disrupted by Yp inversion through LINE-LINE recombination. Hum. Molec. Genet. 7:1-11.

Scozzari, R., F. Cruciani, P. Malaspina et al. 1997. Differential structuring of human populations for homologous X and Y microsatellite loci. Am. J. Hum. Genet. 61:719-733.

Spurdle, A., and T. Jenkins. 1992. Y-chromosome probe p49a detects complex PvuII haplotypes and many new TaqI haplotypes in southern African populations. Am. J. Hum. Genet. 50:107-125.

Talbot, C.C., D. Avramopoulos, S. Gerkin et al. 1995. The tetranucleotide repeat polymorphism D21S1245 demonstrates hypermutability in germline and somatic cells. Hum. Molec. Genet. 4:1193-1199.

Tishkoff, S.A., E. Dietzsch, W. Speed et al. 1996. Global patterns of linkage disequilibrium at the CD4 locus and modern human origins. Science 271:1380-1387.

Torroni, A., O. Semino, R. Scozzari et al. 1990. Y-chromosome DNA polymorphisms in human populations: Differences between Caucasoids and Africans detected by 49a and 49f probes. Ann. Hum. Genet. 54:287-296.

Weber, J.L., and C. Wong. 1993. Mutation in human short tandem repeats. Hum. Molec. Genet. 2:1123-1128.