正在加载图片...
letter Aa2000nAtureAmericaInc..http:/iGenetics.nature.com Y chromosome sequence variation and the history of human populations Peter A Underhill, Peidong Shen, Alice A Lin!, Li Jin, Giuseppe Passarino!, Wei H. Yang Erin Kauffman? Batsheva Bonne-Tamird, Jaume Bertranpetit, Paolo Francalacci, Muntaser Ibrahim, Trefor Jenkins, Judith R. Kidd9, S Qasim Mehdilo, Mark T Seielstad,R. Spencer Wells 2, Alberto Piazza, Ronald W. Davis Marcus W. feldman L. luca cavalli-Sforza& Peter J Oefner Binary polymorphisms associated with the non-recombining close patterns of population affinities apparent from a max nal genetic legacy of our species that has persisted to the pre. cies (table 1). To facilitate presentation, we grouped the 11 sent,permitting inference of human evolution, population lotypes into 10 haplogroups as defined by either the presence or inity and demographic history'. We used denaturing high- the absence of mutations occupying strategic internal positions performance liquid chromatography(DHPLC; ref. 2)to identify in the phylogeny. Haplogroups VI, VIll and X, although poly 60 of the 166 bi-allelic and 1 tri-allelic site that formed a parsi. phyletic, are distinguished by criteria(fable 2) monious genealogy of 116 haplotypes, several of which display Three mutually reinforcing mutations, M42, M94 and M139 distinct population affinities based on the analysis of 1062( two transversions and a 1-bp deletion), distinguish haplogroup globally representative individuals. A minority of contempo. I, which is represented today by a minority of Africans-mainly rary East Africans and Khoisan represent the descendants of Sudanese, Ethiopians and Khoisans (Table 1). All non-Africans, the most ancestral patrilineages of anatomically modern except a single Sardinian, and most African males sampled carr humans that left Africa between 35,000 and 89,000 years ago. only the derived alleles at the three sites. This implies that mod We deduced a phylogenetic tree from 167 NRY polymorphisms ern extant human Y chromosomes trace ancestry to Africa and on the principle of maximum parsimony(Fig. 1). Of the 167 that the descendants of the derived lineage left Africa and eventu polymorphisms, 7 had been detected by means other than ally replaced archaic human Y chromosomes in Eurasi DHPLC and were taken from the literature. Of the 160 polymor- An important property of a phylogeny is the randomness of phisms detected by DhPLC, 73 had been reported previously3.4. number of mutations per segment of the tree. Of the 166 seg- Of the remaining 87 unreported polymorphisms, 53 were discov- ments, 41 carry no mutation, whereas 98, 16, 8, 2 and 1 segment ered in a set of 53 individuals of diverse geographic origin during have 1, 2, 3, 4 and 8 mutations, respectively. The mean number of the screening of the unique sequences and repeat elements, other mutations per segment is 1.024 with a variance of 0.945. Apply than long interspersed elements, contained in 3 overlapping cos- ing the G-test for goodness of fit and william ns correction to the uences(GenBank accession numbers AC003032, observed G, the data do not fit a Poisson distribution 2z898 AC003095, AC003097)and a few small fragments scattered (Gadi34.98, d.f. =3, P-10-). This is due to an excess of segments throughout the NrY. Finally, we detected 34 during genotyping. with one mutation, as expected in an exponentially growing pop In total, the marker panel is composed of 91 transitions, 53 trans- ulation. Similar results were obtained recently for the separate versions, 22 small insertions or deletions, and 1 Alu insertion. All analysis of four Y chromosome genes!. Further support that the polymorphisms are bi-allelic, except a double transversion human population has undergone a major expansion comes (M116)that has three alleles, A, C or T, defining different haplo- from the consistently negative values of Tajima's D (ref 6)for not types. Two non-CpG associated transitions (M64 and M108)only the Y chromosome but also for mitochondrial DNA, X- considered in the context of other markers. We placed the root of dence of significantly reduced variability to the other genetic the phylogeny using sequence information generated from the systems, confirming a similar comparison of a smaller number three great ape species. The sequential succession of mutational of polymorphisms on previously reported NRY sequences with 8 events is unequivocal, except for those appearing in the same tree X-linked. and 16 autosomal human genesi. Possible explana- segment(for example, M42, M94, M139). The phylogeny is com- tions include positive selection on NRY (ref 9)and a difference posed of 116 haplotypes and their frequencies in 21 general pop- between male and female effective population sizes lations are given Table 1). Forty-two haplotypes(36.2%)are Assuming expansion, the age of the most recent common ances represented by just one individual. Several haplotypes, however, tor(Tmrca was previously estimated at 59,000 years, with a 95% have higher frequencies and/or geographic associations that dis- probability interval of 40,000-140,000 years. This value is similar Department of Genetics, Stanford University, Stanford, California, USA. Stanford DNA See University of Texas-Houston, Human Genetics Center, Texas, USA. Sackler Faculty of Medicine, Human Genetics, Tel-Aviv University, Tel-Aviv. Israel. Unitat de Biologia Evolutiva, Facultat de Ciencies de la salut i de la Vida, Universitat Pompeu Fabra, Barcelona, Catalonia, Spain.Dipartimento Engineering Laboratories, Islamabad Pakistan. " Harvard School of Public Health Program for Population Genetics, Boston, Massachusetts uy Genetic Department of Genetics, Yale University School of Medicine, New Haven, Connecticut, USA. Dr A.Q. Khan Research Laboratories, Wellcome Trust Centre for Human Genetics, University of Oxford, Headington, UK. Department of Genetics, Biology and Biochemistry, Department of enetics, University of Torina, Torino, Italy. Department of Biological Sciences, Herrin Laboratories, Stanfo uld be addressed to P.A. U (e-mail: under@stanford. edu 8letter 358 nature genetics • volume 26 • november 2000 Y chromosome sequence variation and the history of human populations Peter A. Underhill1, Peidong Shen2, Alice A. Lin1, Li Jin3, Giuseppe Passarino1, Wei H. Yang2, Erin Kauffman2, Batsheva Bonné-Tamir4, Jaume Bertranpetit5, Paolo Francalacci6, Muntaser Ibrahim7, Trefor Jenkins8, Judith R. Kidd9, S. Qasim Mehdi10, Mark T. Seielstad11, R. Spencer Wells12, Alberto Piazza13, Ronald W. Davis2, Marcus W. Feldman14, L. Luca Cavalli-Sforza1 & Peter. J. Oefner2 1Department of Genetics, Stanford University, Stanford, California, USA. 2Stanford DNA Sequencing and Technology Center, Palo Alto, California, USA. 3University of Texas-Houston, Human Genetics Center, Houston, Texas, USA. 4Sackler Faculty of Medicine, Human Genetics, Tel-Aviv University, Tel-Aviv, Israel. 5Unitat de Biologia Evolutiva, Facultat de Ciències de la Salut i de la Vida, Universitat Pompeu Fabra, Barcelona, Catalonia, Spain. 6Dipartimento di Zoologia e Antropologia Biologica, Università di Sassari, Sassari, Italy. 7Institute of Endemic Diseases, University of Khartoum, Sudan. 8Department of Human Genetics, School of Pathology, South African Institute for Medical Research and the University of Witwatersrand, Johannesburg, South Africa. 9Department of Genetics, Yale University School of Medicine, New Haven, Connecticut, USA. 10Dr. A. Q. Khan Research Laboratories, Biomedical & Genetic Engineering Laboratories, Islamabad, Pakistan. 11Harvard School of Public Health, Program for Population Genetics, Boston, Massachusetts, USA. 12Wellcome Trust Centre for Human Genetics, University of Oxford, Headington, UK. 13Department of Genetics, Biology and Biochemistry, Department of Genetics, University of Torino, Torino, Italy. 14Department of Biological Sciences, Herrin Laboratories, Stanford University, California, USA. Correspondence should be addressed to P.A.U. (e-mail: under@stanford.edu). Binary polymorphisms associated with the non-recombining region of the human Y chromosome (NRY) preserve the pater￾nal genetic legacy of our species that has persisted to the pre￾sent, permitting inference of human evolution, population affinity and demographic history1. We used denaturing high￾performance liquid chromatography (DHPLC; ref. 2) to identify 160 of the 166 bi-allelic and 1 tri-allelic site that formed a parsi￾monious genealogy of 116 haplotypes, several of which display distinct population affinities based on the analysis of 1062 globally representative individuals. A minority of contempo￾rary East Africans and Khoisan represent the descendants of the most ancestral patrilineages of anatomically modern humans that left Africa between 35,000 and 89,000 years ago. We deduced a phylogenetic tree from 167 NRY polymorphisms on the principle of maximum parsimony (Fig. 1). Of the 167 polymorphisms, 7 had been detected by means other than DHPLC and were taken from the literature. Of the 160 polymor￾phisms detected by DHPLC, 73 had been reported previously3,4. Of the remaining 87 unreported polymorphisms, 53 were discov￾ered in a set of 53 individuals of diverse geographic origin during the screening of the unique sequences and repeat elements, other than long interspersed elements, contained in 3 overlapping cos￾mid sequences (GenBank accession numbers AC003032, AC003095, AC003097) and a few small fragments scattered throughout the NRY. Finally, we detected 34 during genotyping. In total, the marker panel is composed of 91 transitions, 53 trans￾versions, 22 small insertions or deletions, and 1 Alu insertion. All polymorphisms are bi-allelic, except a double transversion (M116) that has three alleles, A, C or T, defining different haplo￾types. Two non-CpG associated transitions (M64 and M108) show evidence of recurrence, but generate no ambiguities when considered in the context of other markers. We placed the root of the phylogeny using sequence information generated from the three great ape species. The sequential succession of mutational events is unequivocal, except for those appearing in the same tree segment (for example, M42, M94, M139). The phylogeny is com￾posed of 116 haplotypes and their frequencies in 21 general pop￾ulations are given (Table 1). Forty-two haplotypes (36.2%) are represented by just one individual. Several haplotypes, however, have higher frequencies and/or geographic associations that dis￾close patterns of population affinities apparent from a maximum likelihood analysis (Fig. 2) performed on the haplotype frequen￾cies (Table 1). To facilitate presentation, we grouped the 116 hap￾lotypes into 10 haplogroups as defined by either the presence or the absence of mutations occupying strategic internal positions in the phylogeny. Haplogroups VI, VIII and X, although poly￾phyletic, are distinguished by criteria (Table 2). Three mutually reinforcing mutations, M42, M94 and M139 (two transversions and a 1-bp deletion), distinguish haplogroup I, which is represented today by a minority of Africans—mainly Sudanese, Ethiopians and Khoisans (Table 1). All non-Africans, except a single Sardinian, and most African males sampled carry only the derived alleles at the three sites. This implies that mod￾ern extant human Y chromosomes trace ancestry to Africa and that the descendants of the derived lineage left Africa and eventu￾ally replaced archaic human Y chromosomes in Eurasia5. An important property of a phylogeny is the randomness of number of mutations per segment of the tree. Of the 166 seg￾ments, 41 carry no mutation, whereas 98, 16, 8, 2 and 1 segment have 1, 2, 3, 4 and 8 mutations, respectively. The mean number of mutations per segment is 1.024 with a variance of 0.945. Apply￾ing the G-test for goodness of fit and Williams’ correction to the observed G, the data do not fit a Poisson distribution (Gadj=34.98, d.f.=3, P∼10–7). This is due to an excess of segments with one mutation, as expected in an exponentially growing pop￾ulation. Similar results were obtained recently for the separate analysis of four Y chromosome genes4. Further support that the human population has undergone a major expansion comes from the consistently negative values of Tajima’s D (ref. 6) for not only the Y chromosome, but also for mitochondrial DNA, X￾chromosomal and autosomal genes4. Notably, NRY shows evi￾dence of significantly reduced variability to the other genetic systems4, confirming a similar comparison of a smaller number of polymorphisms on previously reported NRY sequences with 8 X-linked7,8 and 16 autosomal human genes4. Possible explana￾tions include positive selection on NRY (ref. 9) and a difference between male and female effective population sizes10. Assuming expansion, the age of the most recent common ances￾tor (TMRCA) was previously estimated at 59,000 years, with a 95% probability interval of 40,000–140,000 years11. This value is similar © 2000 Nature America Inc. • http://genetics.nature.com © 2000 Nature America Inc. • http://genetics.nature.com
向下翻页>>
©2008-现在 cucdc.com 高等教育资讯网 版权所有