- Open Access
TagSNP transferability and relative loss of variability prediction from HapMap to an admixed population
© Lins et al; licensee BioMed Central Ltd. 2009
- Received: 23 June 2009
- Accepted: 14 August 2009
- Published: 14 August 2009
The application of a subset of single nucleotide polymorphisms, the tagSNPs, can be useful in capturing untyped SNPs information in a genomic region. TagSNP transferability from the HapMap dataset to admixed populations is of uncertain value due population structure, admixture, drift and recombination effects. In this work an empirical dataset from a Brazilian admixed sample was evaluated against the HapMap population to measure tagSNP transferability and the relative loss of variability prediction.
The transferability study was carried out using SNPs dispersed over four genomic regions: the PTPN22, HMGCR, VDR and CETP genes. Variability coverage and the prediction accuracy for tagSNPs in the selected genomic regions of HapMap phase II were computed using a prediction accuracy algorithm. Transferability of tagSNPs and relative loss of prediction were evaluated according to the difference between the Brazilian sample and the pooled and single HapMap population estimates.
Each population presented different levels of prediction per gene. On average, the Brazilian (BRA) sample displayed a lower power of prediction when compared to HapMap and the pooled sample. There was a relative loss of prediction for BRA when using single HapMap populations, but a pooled HapMap dataset generated minor loss of variability prediction and lower standard deviations, except at the VDR locus at which loss was minor using CEU tagSNPs.
Studies that involve tagSNP selection for an admixed population should not be generally correlated with any specific HapMap population and can be better represented with a pooled dataset in most cases.
- Linkage Disequilibrium
- Relative Loss
- Linkage Disequilibrium Analysis
- Admix Population
- PTPN22 Gene
Since association studies were first introduced as a tool in understanding the genetic basis of complex phenotypes  an enormous methodological and analytical framework has been developed with regard to regions of high linkage disequilibrium (LD) and common haplotypes for genome-wide LD mapping [2, 3]. The extension and localization of those regions are the mainstream in developing a set of SNPs capable of statistically representing untyped markers - the tagSNPs - reducing the costs of medium and high throughput genotyping in association studies [2–6]. The application of public genome data brought about great advances in the understanding of genetic variability and helped design association studies for complex phenotypes among several human populations of different ethnic backgrounds [7, 8]. The three continental population samples in the HapMap project - Utah residents with northern and western European ancestry (CEU), East-Asians (Japanese from Tokyo and Han Chinese from Beijing) (CHB+JPT) and African Yoruba from Ibadan, Nigeria (YRI) - are used in experimental design as a reference for association studies in worldwide populations [6–9].
The challenge in establishing the HapMap as a standard for research is highlighted by the observation that the distribution of the haplotype blocks differs between population groups due to genetic and demographic effects . However, tagSNP sharing from the HapMap dataset is commonly described as appropriately applied in European and East Asian populations [11–17], but less effective in other structured or multi-ethnic populations [9, 10, 18, 19]. Such differences increase proportionally with the geographical distance between the HapMap data collection points and the actual sample collection [6, 9, 15, 17]. Although the project never stated that these samples were representative of global variation, the fact that the HapMap study was carried out using only these ethno-geographic samples has been cited against the use of such data in populations that have a history of recent admixture [20–22].
Admixed populations can be useful in detecting genetic contributors to complex traits that differ in frequency between distinct populations. The admixture mapping approach has been proposed as an effective method for the identification of disease-susceptibility alleles with higher probability due to admixture-generated linkage disequilibrium extension . Considering that the Latin-American people are one of the most heterogeneous around the world [24–26] as a result of mating primarily amongst three ethnic groups - Europeans, Native (South) Americans and Africans - the admixture mapping should be used as an alternative approach for the identification of disease-susceptibility loci [21, 27].
Therefore, unintended use of tagging SNP data in admixed populations could lead to spurious results since there is evidence that admixture impacts the linkage disequilibrium structure, affecting the association of SNPs with etiological factors [28, 29]. Such issues could render HapMap-based tagSNP selection approaches for admixed populations inaccurate or even useless. Moreover, knowledge of the degree of portability of HapMap data to admixed populations is also needed in order to comprehend whether there is loss or gain of variability when using tagSNPs selected from the consortium populations. Thus, the aim of this work was to develop a first approach to evaluate the tagSNP transferability from HapMap to the Brazilian admixed population, using 37 SNPs distributed between four loci: VDR, PTPN22, HMGCR and CETP.
The sample of Brazilian subjects (BRA) consisted of 200 unrelated parents randomly selected from paternity test trios. A stratified sampling approach was adopted to represent the five Brazilian geopolitical regions according to each individual's place of birth. Genetic ancestry coefficients were estimated [30, 31] so as to validate the admixture source of the population. All sampled individuals signed an informed consent allowing the use of their DNA for paternity testing and further anonymous population genetics research.
The genotypes of the HapMap population samples were retrieved from the database (Data Rel 21a/phaseII Jan07, on NCBI B35 assembly, dbSNP b125) consisting of 89 unrelated East Asian individuals (CHB+JPT) comprising 45 Han Chinese from Beijing and 44 Japanese from Tokyo; 90 individuals of northern and western European origin (CEU); and 90 Yoruba individuals (YRI) from Ibadan, Nigeria. All HapMap population genotypes for each gene were combined into a pooled sample (POOL; n = 269) in order to test a representative multi-ethnic population thereby resulting in a final set of five population samples: CHB+JPT, CEU, YRI, POOL and BRA. The research project was approved by the Universidade Católica de Brasília Ethics Review Board.
SNP selection and genotyping
Characteristics of genomic regions genotyped in this study
Average Distance (Kb)
Gene Extension (Kb)
TagSNP transferability and LD analysis
The tagSNP transferability study was conducted using the Stampa algorithm  implemented on the Gevalt package . This algorithm aims to maximize the expected accuracy of predicting untyped SNPs based on genotype data of the tagSNPs . To conduct this study, first the variability prediction accuracy for each gene was assessed to calculate the coverage of the HapMap phase II data in relation to the total number of available SNPs in each region: number of common SNPs - with minor allele frequency (MAF) > 0.05; number of SNPs required to capture 100% of SNP prediction; maximum prediction using the same number of SNPs as in the study; and the prediction for the selected set of SNPs. Then, the set of SNPs selected with average distances of 5 Kb had their variability prediction calculated based on two until the maximum number of tagSNPs for all five samples. Finally, the relative loss of variability prediction (in percentage points; pp) was calculated by subtracting the variability prediction of tagSNPs selected for BRA from the relative prediction obtained when using the tagSNPs selected for each of the HapMap populations and the pooled sample in the Brazilian group.
Measures of linkage disequilibrium (LD) between pairs of SNP loci (D' and r2) were calculated by the Gerbil algorithm , implemented in Gevalt, using the standard maximum-likelihood and expectation-maximization algorithm methods. Only the SNPs accounted for in all five populations were evaluated. A pairwise population LD analysis was carried out using a Spearman's correlation coefficient.
Variability coverage of HapMap
SNP prediction Coverage for HapMap population samples
n to 100% prediction
Prediction of selected SNPs (%)
Max. Prediction (%)
TagSNP transferability analysis
Loss of SNP prediction coverage in BRA using HapMap tagSNPs
A.V. ± S.D.
3.668 ± 2.671
4.481 ± 2.501
6.372 ± 4.438
3.437 ± 1.050
Pairwise LD analysis
Spearman's correlation coefficient (rho)
The success of a genetic association study is strongly affected by marker selection for a specific population. With regard to admixed populations this criterion is of fundamental concern due to the risk of spurious associations in the case of inefficient choice. The HapMap Consortium provided solutions for most cases by making available millions of markers genome-wide that were genotyped in each of the continental populations, although it did not address how markers selected in one or more HapMap samples will perform in studies with other populations . To date, several studies have evaluated tagSNPs portability in a range of worldwide populations, but none has assessed a heterogeneous admixed population. The present study indicates that tagSNP sets from HapMap population can be portable to admixed populations to a reasonable degree, however the results can also be uncertain and inaccurate if applied improperly. It also demonstrates the necessity for understanding the patterns of physical (gene extension and SNP density) and genetic (LD patterns) differences in every genomic region prior to determining the tagSNPs to be used, in order to make a reasonable prediction for untyped markers.
Measures of LD and SNP density vary across the genome and can be critical points when selecting a set of tagSNPs. A study by Tantoso and colleagues  showed that SNPs can be transferred from HapMap to other populations of the same ethnic and continental origin. Even so, tagSNP coverage increases along with the SNP density due to the high LD in European and the Asian populations. Hence, coverage of many untyped variants, especially the rare ones (MAF <0.05), drops from 50% to 30% depending on the population used . Another study  showed that the SNP density has a major effect on tag selection, proposing denser sets (i.e., one SNP every 1.3 kb) to improve the tagSNP performance. In the present study the SNPs were selected with SNP density that was approximately equal in the four regions studied (one SNP every 5 kb), to reduce or eliminate such an effect. Using the same genotyped SNP density at two regions with physically different densities - CETP (30 kb and 3.3 SNP/kb) and HMGCR (28 kb and 0.8 SNP/kb) - demonstrates that either maximum or minimum prediction among regions and within the population provided no more than 10 percentage points of loss in prediction (Table 2). Though, the fact that the prediction becomes stable or decreases as the number of tagSNPs increases is evidence that SNP density can be a critical point in tagSNP selection in larger genome-wide sets [15, 37], as well as in low-throughput region analysis, emphasizing that for an admixed population it is necessary to use, in a reduced panel, as many SNPs as possible.
The SNP prediction and tagSNP transferability are also dependent on the linkage disequilibrium patterns and hence in admixed populations they can be influenced both by the demographic events and by genetic factors. Generally, tagSNP sets selected for similar populations with similar haplotype block structures have better performance but differ if the block structures and boundaries also differ [6, 9–12, 38]. For instance, CEU tags are useful for populations with European ancestry and tagSNPs selected for YRI perform well in Sub-Saharan Africans, but require larger genotype densities due to lower LD among markers [11, 12, 37].
The linkage disequilibrium measures could be evidence leading to the belief that one could use tagSNPs directly transferred from CEU to BRA without great loss of variability, since the greatest ancestral contribution in the Brazilian sample is European [24, 25, 30, 31]. Considering all SNP pairs in the current dataset the pairwise LD had the highest correlation between BRA and CEU, followed by POOL, CHB+JPT and YRI, which had the lowest average LD and was less correlated. However when genes were analyzed individually, except for the VDR gene, the POOL data had the highest correlation compared to the other populations.
Although using tagSNPs directly form CEU worked with great efficiency in some cases, as in the case of VDR gene, in others this type of selection provided greater loss of variability, as in the specific case of the PTPN22 gene, reinforcing the idea that each genomic region will perform according to gene and population structure . Linkage disequilibrium arising from the recent admixture of genetically distinct populations can be categorized as a genome-wide effect and thus selecting markers from representative parental populations offers analytical risks due to the fact that in some genomic regions, particularly those with high LD, ancestral haplotype-block structures at the individual level are not always eliminated by recent admixture.
Population stratification along the Latin American populations varies extensively as consequence of their history of immigration and colonization over the last five centuries. In Brazil there is a major contribution from the European ancestry followed by African and Amerindian [24, 25, 30, 31]. In the present data the pooled sample tagSNP performance had a relative loss of prediction smaller than any other population sample. Although the relative loss of prediction among CHB+JPT and POOL were very close, the fact that standard deviation in the pooled sample was lower demonstrated that, in a study with multiple gene analysis, it can be a safe alternative to choose tagSNPs from the pooled samples, because different LD patterns at different genes can have different SNP coverage depending on each of the HapMap populations .
In other Latin-American populations, such as those from Mexico or Argentina, the contributions of the Amerindian proportion at population level are usually higher than in Brazil, and African ancestry is higher in Caribbean populations than in any other [39–41]. Such population structure difference should be considered when applying a tagSNP selection method depending on each specific case of admixture. It is possible that for Mexicans or Argentineans a combination of the CEU, CHB and JPT HapMap samples would perform better than the whole HapMap pool, as was the case for South Asian populations such as the Indian population  and Hazara, Kalash and Uygur populations . The combinations of HapMap panels were also effective at representing other populations, such as the Philippines , for which CHB samples and the combined CHB+JPT samples were most transferable to Cebu Filipino samples, indicating that different pools of HapMap panels should be tested and used as an alternative in many situations.
However, it is noteworthy that the SNP coverage in HapMap is not complete and tagging strategies critically depend on the investigation of other population polymorphisms . The project is now overcoming the representative world-wide population issue with the Phase III release, which includes Amerindian and Mexican ancestral populations among others. This will certainly improve the methods of tagSNP selection for admixed populations but a comprehensive study using high-throughput genome-wide SNPs in assorted admixed populations will be required to reduce confounding effects caused by population stratification and to enhance the tagSNP performance. Identification, re-sequencing, and genotyping of large-scale and high-throughput SNP data were beyond the scope of this study. Further analysis will be necessary to assess if such techniques will attain the same level of efficiency in other admixed populations in which a history of admixture processes differs from the Brazilian sample, known for being recent and continuous, as opposed to populations which have undergone well defined time limited admixture processes in the past.
The pooled HapMap sample provided the minimum loss of prediction in admixed population and therefore, combined with the SNP selection spaced at most every 5.0 kb may represent an efficient alternative. The present findings will be useful for the future design and analysis of genetic studies using other admixed populations, suggesting that on such occasions the selection of markers should not be generalized according to the tagSNPs of one or other current HapMap populations due to genetic and demographic effects.
This work was supported by Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES), Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), and Universidade Católica de Brasília (PRPGP-UCB). TCL and BSA were supported by a CAPES Master's scholarship. We thank Dr. Dario Grattapaglia for the ABI3100 sharing and providing the Brazilian samples and Rodrigo G Vieira for extensive work of genotyping and estimating the individual ancestry of the Brazilian samples. We are grateful to Robert Pogue for English version review of the final manuscript.
- Risch N, Merikangas K: The future of genetic studies of complex human diseases. Science. 1996, 273: 1516-10.1126/science.273.5281.1516.View ArticlePubMedGoogle Scholar
- Johnson GC, Esposito L, Barratt BJ, Smith AN, Heward J, Di Genova G, Ueda H, Cordell HJ, Eaves IA, Dudbridge F: Haplotype tagging for the identification of common disease genes. Nat Genet. 2001, 29: 233-237. 10.1038/ng1001-233.View ArticlePubMedGoogle Scholar
- Sachidanandam R, Weissman D, Schmidt SC, Kakol JM, Stein LD, Marth G, Sherry S, Mullikin JC, Mortimore BJ, Willey DL: A map of human genome sequence variation containing 1.42 million single nucleotide polymorphisms. Nature. 2001, 409: 928-933. 10.1038/35057149.View ArticlePubMedGoogle Scholar
- Carlson CS, Eberle MA, Rieder MJ, Yi Q, Kruglyak L, Nickerson DA: Selecting a maximally informative set of single-nucleotide polymorphisms for association analyses using linkage disequilibrium. Am J Hum Genet. 2004, 74: 106-120. 10.1086/381000.PubMed CentralView ArticlePubMedGoogle Scholar
- Ke X, Cardon LR: Efficient selective screening of haplotype tag SNPs. Bioinformatics. 2003, 19: 287-288. 10.1093/bioinformatics/19.2.287.View ArticlePubMedGoogle Scholar
- Xing J, Witherspoon DJ, Watkins WS, Zhang Y, Tolpinrud W, Jorde LB: HapMap tagSNP transferability in multiple populations: general guidelines. Genomics. 2008, 92: 41-51. 10.1016/j.ygeno.2008.03.011.PubMed CentralView ArticlePubMedGoogle Scholar
- Andrawiss M: First phase of HapMap project already helping drug discovery. Nat Rev Drug Discov. 2005, 4: 947-10.1038/nrd1918.View ArticlePubMedGoogle Scholar
- The International HapMap Consortium: A haplotype map of the human genome. Nature. 2005, 437: 1299-1320. 10.1038/nature04226.PubMed CentralView ArticleGoogle Scholar
- Gonzalez-Neira A, Ke X, Lao O, Calafell F, Navarro A, Comas D, Cann H, Bumpstead S, Ghori J, Hunt S: The portability of tagSNPs across populations: a worldwide survey. Genome Res. 2006, 16: 323-330. 10.1101/gr.4138406.PubMed CentralView ArticlePubMedGoogle Scholar
- Sawyer SL, Mukherjee N, Pakstis AJ, Feuk L, Kidd JR, Brookes AJ, Kidd KK: Linkage disequilibrium patterns vary substantially among populations. Eur J Hum Genet. 2005, 13: 677-686. 10.1038/sj.ejhg.5201368.View ArticlePubMedGoogle Scholar
- Conrad DF, Jakobsson M, Coop G, Wen X, Wall JD, Rosenberg NA, Pritchard JK: A worldwide survey of haplotype variation and linkage disequilibrium in the human genome. Nat Genet. 2006, 38: 1251-1260. 10.1038/ng1911.View ArticlePubMedGoogle Scholar
- de Bakker PI, Burtt NP, Graham RR, Guiducci C, Yelensky R, Drake JA, Bersaglieri T, Penney KL, Butler J, Young S: Transferability of tag SNPs in genetic association studies in multiple populations. Nat Genet. 2006, 38: 1298-1303. 10.1038/ng1899.View ArticlePubMedGoogle Scholar
- Gu S, Pakstis AJ, Li H, Speed WC, Kidd JR, Kidd KK: Significant variation in haplotype block structure but conservation in tagSNP patterns among global populations. Eur J Hum Genet. 2007, 15: 302-312. 10.1038/sj.ejhg.5201751.View ArticlePubMedGoogle Scholar
- Huang W, He Y, Wang H, Wang Y, Liu Y, Chu X, Xu L, Shen Y, Xiong X, Li H: Linkage disequilibrium sharing and haplotype-tagged SNP portability between populations. Proc Natl Acad Sci USA. 2006, 103: 1418-1421. 10.1073/pnas.0510360103.PubMed CentralView ArticlePubMedGoogle Scholar
- Montpetit A, Nelis M, Laflamme P, Magi R, Ke X, Remm M, Cardon L, Hudson TJ, Metspalu A: An evaluation of the performance of tag SNPs derived from HapMap in a Caucasian population. PLoS Genet. 2006, 2: e27-10.1371/journal.pgen.0020027.PubMed CentralView ArticlePubMedGoogle Scholar
- Service S, Sabatti C, Freimer N: Tag SNPs chosen from HapMap perform well in several population isolates. Genet Epidemiol. 2007, 31: 189-194. 10.1002/gepi.20201.View ArticlePubMedGoogle Scholar
- Willer CJ, Scott LJ, Bonnycastle LL, Jackson AU, Chines P, Pruim R, Bark CW, Tsai YY, Pugh EW, Doheny KF: Tag SNP selection for Finnish individuals based on the CEPH Utah HapMap database. Genet Epidemiol. 2006, 30: 180-190. 10.1002/gepi.20131.View ArticlePubMedGoogle Scholar
- Barrett JC, Cardon LR: Evaluating coverage of genome-wide association studies. Nat Genet. 2006, 38: 659-662. 10.1038/ng1801.View ArticlePubMedGoogle Scholar
- Xu Z, Kaplan NL, Taylor JA: Tag SNP selection for candidate gene association studies using HapMap and gene resequencing data. Eur J Hum Genet. 2007, 15: 1063-1070. 10.1038/sj.ejhg.5201875.View ArticlePubMedGoogle Scholar
- Choudhry S, Coyle NE, Tang H, Salari K, Lind D, Clark SL, Tsai HJ, Naqvi M, Phong A, Ung N: Population stratification confounds genetic association studies among Latinos. Hum Genet. 2006, 118: 652-664. 10.1007/s00439-005-0071-3.View ArticlePubMedGoogle Scholar
- Hoggart CJ, Parra EJ, Shriver MD, Bonilla C, Kittles RA, Clayton DG, McKeigue PM: Control of confounding of genetic associations in stratified populations. Am J Hum Genet. 2003, 72: 1492-1504. 10.1086/375613.PubMed CentralView ArticlePubMedGoogle Scholar
- Ziv E, Burchard EG: Human population structure and genetic association studies. Pharmacogenomics. 2003, 4: 431-441. 10.1517/phgs.4.4.431.22758.View ArticlePubMedGoogle Scholar
- Patterson N, Hattangadi N, Lane B, Lohmueller KE, Hafler DA, Oksenberg JR, Hauser SL, Smith MW, OBrien SJ, Altshuler D: Methods for high-density admixture mapping of disease genes. Am J Hum Genet. 2004, 74: 979-1000. 10.1086/420871.PubMed CentralView ArticlePubMedGoogle Scholar
- Callegari-Jacques SM, Grattapaglia D, Salzano FM, Salamoni SP, Crossetti SG, Ferreira ME, Hutz MH: Historical genetics: spatiotemporal analysis of the formation of the Brazilian population. Am J Hum Biol. 2003, 15: 824-834. 10.1002/ajhb.10217.View ArticlePubMedGoogle Scholar
- Salzano FM: Interethnic variability and admixture in Latin America - social implications. Rev Biol Trop. 2004, 52: 405-415.PubMedGoogle Scholar
- Sans M: Admixture studies in Latin America: from the 20th to the 21st century. Hum Biol. 2000, 72: 155-177.PubMedGoogle Scholar
- McKeigue PM: Prospects for admixture mapping of complex traits. Am J Hum Genet. 2005, 76: 1-7. 10.1086/426949.PubMed CentralView ArticlePubMedGoogle Scholar
- Terwilliger JD, Hiekkalinna T: An utter refutation of the "Fundamental Theorem of the HapMap". Eur J Hum Genet. 2006, 14: 426-437. 10.1038/sj.ejhg.5201583.View ArticlePubMedGoogle Scholar
- Xu S, Huang W, Wang H, He Y, Wang Y, Qian J, Xiong M, Jin L: Dissecting linkage disequilibrium in African-American genomes: roles of markers and individuals. Mol Biol Evol. 2007, 24: 2049-2058. 10.1093/molbev/msm135.View ArticlePubMedGoogle Scholar
- Lins TC: [Impact of admixture on the performance of HapMap data in Brazilian population assessed in PTPN22 and VDR genes]. 2007, Universidade Católica de Brasília, Dissertation ThesisGoogle Scholar
- Lins TC, Vieira RG, Abreu BS, Grattapaglia D, Pereira RW: Genetic composition of Brazilian population samples based on a set of twenty eight ancestry informative SNPs. Am J Hum Biol. 2009, Epub:DOI 10.1002/ajhb.20976Google Scholar
- Phillips MS, Lawrence R, Sachidanandam R, Morris AP, Balding DJ, Donaldson MA, Studebaker JF, Ankener WM, Alfisi SV, Kuo FS: Chromosome-wide distribution of haplotype blocks and the role of recombination hot spots. Nat Genet. 2003, 33: 382-387. 10.1038/ng1100.View ArticlePubMedGoogle Scholar
- Lins TC, Nogueira LR, Lima RM, Gentil P, Oliveira RJ, Pereira RW: A multiplex single-base extension protocol for genotyping Cdx2, FokI, BsmI, ApaI, and TaqI polymorphisms of the vitamin D receptor gene. Genet Mol Res. 2007, 6: 316-324.PubMedGoogle Scholar
- Halperin E, Kimmel G, Shamir R: Tag SNP selection in genotype data for maximizing SNP prediction accuracy. Bioinformatics. 2005, 21 (Suppl 1): i195-203. 10.1093/bioinformatics/bti1021.View ArticlePubMedGoogle Scholar
- Davidovich O, Kimmel G, Shamir R: GEVALT: an integrated software tool for genotype analysis. BMC Bioinformatics. 2007, 8: 36-10.1186/1471-2105-8-36.PubMed CentralView ArticlePubMedGoogle Scholar
- Kimmel G, Shamir R: GERBIL: Genotype resolution and block identification using likelihood. Proc Natl Acad Sci USA. 2005, 102: 158-162. 10.1073/pnas.0404730102.PubMed CentralView ArticlePubMedGoogle Scholar
- Tantoso E, Yang Y, Li KB: How well do HapMap SNPs capture the untyped SNPs?. BMC Genomics. 2006, 7: 238-10.1186/1471-2164-7-238.PubMed CentralView ArticlePubMedGoogle Scholar
- Liu N, Sawyer SL, Mukherjee N, Pakstis AJ, Kidd JR, Kidd KK, Brookes AJ, Zhao H: Haplotype block structures show significant variation among populations. Genet Epidemiol. 2004, 27: 385-400. 10.1002/gepi.20026.View ArticlePubMedGoogle Scholar
- Benn-Torres J, Bonilla C, Robbins CM, Waterman L, Moses TY, Hernandez W, Santos ER, Bennett F, Aiken W, Tullock T: Admixture and population stratification in African Caribbean populations. Ann Hum Genet. 2008, 72: 90-98.PubMedGoogle Scholar
- Martinez-Marignac VL, Valladares A, Cameron E, Chan A, Perera A, Globus-Goldberg R, Wacher N, Kumate J, McKeigue P, O'Donnell D: Admixture in Mexico City: implications for admixture mapping of type 2 diabetes genetic risk factors. Hum Genet. 2007, 120: 807-819. 10.1007/s00439-006-0273-3.View ArticlePubMedGoogle Scholar
- Seldin MF, Tian C, Shigeta R, Scherbarth HR, Silva G, Belmont JW, Kittles R, Gamron S, Allevi A, Palatnik SA: Argentine population genetic structure: large variance in Amerindian contribution. Am J Phys Anthropol. 2007, 132: 455-462. 10.1002/ajpa.20534.PubMed CentralView ArticlePubMedGoogle Scholar
- Marvelle AF, Lange LA, Qin L, Wang Y, Lange EM, Adair LS, Mohlke KL: Comparison of ENCODE region SNPs between Cebu Filipino and Asian HapMap samples. J Hum Genet. 2007, 52: 729-737. 10.1007/s10038-007-0175-9.View ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.