Human Molecular Genetics, 2001, Vol. 10, No. 5 545-551
© 2001 Oxford University Press
The genome-wide distribution of background linkage disequilibrium in a population isolate
Center for Neurobehavioral Genetics, UCLA, Gonda Center, Room 3506, 695 Charles E. Young Drive South, Box 951761, Los Angeles, CA 90095-1761, USA
Received 26 December 2000; Revised and Accepted 8 January 2000.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Recent interest in using association studies to investigate complex traits has focused attention on understanding linkage disequilibrium (LD) in the human genome. We examined the genome-wide distribution and magnitude of such background LD (BLD) using 1036 densely spaced microsatellites, in a sample from the demographically well characterized population of the Central Valley of Costa Rica. High levels of BLD were found between linked markers several centiMorgans apart, and although BLD was significantly related to genetic distance between markers it was not spread uniformly throughout the genome. Understanding the forces governing the distribution of BLD in the genome will require similar investigations using a standard set of markers in other populations.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
The success of population level approaches to genetic mapping hinges on the maintenance of linkage disequilibrium (LD)—the non-random association of alleles at different genetic loci—between disease and nearby marker alleles. LD between disease and markers in population samples has been used successfully as an adjunct to pedigree analyses to identify or narrow the location of disease genes (1). Few studies have used LD to initially localize a disease gene (2,3). It has been proposed, however, that given the expected availability of very dense genetic maps, whole genome LD scanning is a powerful strategy to map complex diseases (4), especially in population isolates where it is hypothesized that affected individuals may descend from a few common founders (5,6). Not all researchers, however, are sanguine about the possible success of LD genome screening for complex disorders. For example, it has been suggested on theoretical grounds that LD will extend over such a short interval (in the order of a few kilobases), that adequate genome screens for association will require the use of up to 500 000 mapped, single nucleotide polymorphisms (SNPs) and that even in population isolates one would not expect to observe substantially more LD under most scenarios of population history (7). This view is supported by empirical evidence from examination of a limited number of genetic regions (8,9); however, other empirical evidence suggests LD to be more extensive (10,11). The number of SNPs for a genome screen is a topic of active debate, as the extent of LD in the human genome has not been well characterized (12).
For genome screens in population samples to be successful, the LD signal due to association with a shared disease allele must stand out from the signal from background LD (BLD), that is from allelic associations between marker loci. Theoretical studies have suggested that such BLD is highly dependent on population history (13), with rapidly growing populations showing less BLD than constant sized populations. This hypothesis was examined empirically by Laan and Pääbo (14) in a study of seven microsatellite markers spanning ~4 cM on chromosome X. These microsatellites were genotyped in four populations; a constant sized population showed more BLD than three rapidly expanding populations (14), in agreement with population genetic theory (13). The demographic history of a population may also predict the relative utility of that population for LD mapping of disease loci. In samples of affected individuals sharing a phenotype, and possibly sharing a susceptibility allele at the same disease locus, the amount of LD around the shared disease locus should be greater in a younger population, as there has been less time for recombination to erode associations. It has been suggested that if such a young population has also undergone rapid growth it should be ideal for LD mapping of disease loci (15). In this case there would be, on average, fewer generations to a common disease-bearing ancestor and LD should be more extensive around the disease locus (6); at the same time, as noted above, BLD should be less extensive in the population. However, these predictions have not yet been tested empirically.
In the past, evaluation of BLD was limited by the available genetic map. For example, early surveys of BLD were conducted in the human HLA region (16) where many polymorphic loci were available. Several studies subsequently used restriction fragment length polymorphisms (RFLPs) around positionally cloned genes to examine the relationship between BLD and physical or genetic distance in a variety of genome regions. In general, an inverse relationship was found between BLD and physical distance, but this relationship disappeared at very close physical distances (<50–60 kb) between loci (17). Peterson et al. (18) were the first to study BLD using polymorphic microsatellite markers over a large genomic region, examining 32 microsatellite loci spanning several regions of chromosome 4 in a Finnish population sample. They found that BLD was inversely associated with genetic distance between loci, however the strength of this association varied depending on the genetic map employed. As they noted a stronger correlation of BLD with physical distance between markers than with genetic distance, they suggested that genetic maps may be too inaccurate over the small intervals where one expects to see the most LD to quantify precisely the relationship between BLD and genetic distance. Huttley et al. (19) were the first to compare BLD in different chromosomal regions with the same sample, scanning for BLD across the entire human genome, using all of the microsatellite genotype data from the CEPH database; they examined 54 independent chromosomes derived from eight Utah and Amish CEPH pedigrees. They found BLD to be negatively related to the genetic distance between the loci and found differences both within and between chromosomes in the magnitude and distribution of BLD, even when controlling for different marker density. CEPH families, however, do not represent a population sample and this limits the inferences that can be drawn from that study. Additionally, in the database used by Huttley et al., the loci are distributed unevenly across the genome.
Several recent studies have examined the extent of BLD in a limited number of genomic regions in one or more populations. Many of these studies find that BLD is not distributed uniformly across the genome (10,11,20). Some of the studies conclude that there are few differences among populations, and that population isolates may not offer an advantage in LD mapping (8,9), whereas others suggest that ‘subisolates’ have increased levels of LD (11). The extent of BLD in these studies is highly variable, ranging from 5 kb to 1 Mb. As the extent of LD in different populations is critical to the success of association mapping, a complete characterization of BLD across the genome in populations suitable for mapping is clearly needed (15,21).
The present study examines BLD at relatively even intervals across the entire genome in a sample from a demographically well characterized and genetically homogeneous population, that of the Central Valley of Costa Rica (CVCR). Historical information and the results of several genetic investigations indicate that the CVCR population constitutes a genetic isolate. The CVCR population had less than 100 Spanish and 300 Amerindian founding families in the 16th–18th centuries. Most of the admixture between the founding populations was completed before the end of the 18th century, and the current CVCR population of two to three million is almost entirely descended from about 4000 individuals alive in 1700. Because the CVCR is separated from Pacific and Atlantic coastal regions by mountain ranges its population was geographically isolated from the founding of the Costa Rican colony until the late 19th century. During this time the CVCR population grew almost exponentially in the absence of substantial migration (22). The CVCR displays similar features to other genetic isolates with few founders, notably a concentration of rare autosomal-recessive diseases (23). The genetic homogeneity of the CVCR population has facilitated genetic mapping of such disorders by LD analysis (24), and has encouraged application of LD mapping methods to more complex disorders, such as bipolar disorder (BP).
For the current study, we evaluated BLD in non-transmitted chromosomes from parents of individuals affected with BP who were sampled independently from one another from hospital and clinic populations of the CVCR (22). Genealogical evaluations indicate that the individuals used in the current study have predominantly CVCR ancestry and are not descended from common ancestors within the past four generations; the majority of these individuals have known shared ancestry from 5 to 20 generations ago (Ophoff et al., unpublished data). We report here the results of our investigation of the genome-wide magnitude and distribution of BLD in the 157 non-transmitted chromosomes in this sample, using genotypes for 1036 microsatellite markers.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
In the 1012 tests of adjacent markers, 159 (16%) were found to have a P value from Fisher’s Exact test (FET) of <0.01, with nearly 40% of markers <1 cM apart having a FET <0.01 (Fig. 1). Using 180 independent locus pairs, the distance between markers was found to significantly predict the probability of being in BLD, with markers
4 cM apart (the approximate average distance between markers) having 2.24 times the odds of a FET P value of <0.05 than markers >4 cM apart (P = 0.022). Thirty-three percent of markers within 4 cM and 17% of markers >4 cM apart had a FET P value of <0.05, respectively. This relationship was not particularly tight however as strong BLD was observed in numerous markers as far apart as 7 cM (the correlation in Fig. 2 with FET is only 0.18). The number of alleles at each marker also predicted BLD, with r x c tables with a dimension >64 (the average dimension of the 1012 r x c tables) having 2.1 times the odds of a FET P value of <0.05 than markers with dimension <64 (P = 0.016). There was no significant interaction between distance and dimension.
|
|
In addition to evaluating BLD using the P value from FET, we also calculated D’ for all pairs of markers. The majority of marker pairs (57%) had a D’ value
0.30 (Fig. 1). Using the 180 independent locus pairs, the correlation between genetic distance and D’ (r = –0.11) was weaker than the correlation between genetic distance and FET (r = 0.18) (Fig. 2). The value of FET and D’ were significantly negatively correlated (r = –0.59, P < 0.0001), therefore additional analyses used only the FET P value. Using all 1012 FET P values we found chromosomes to have significantly different levels of BLD, even when controlling for the (significant) covariates, distance between markers and numbers of alleles at each marker. Chromosome X, with a very high degree of BLD, was set as the reference, and chromosomes 1, 6, 10, 15, 17, 19, 21 and 22 were all found to have significantly less BLD, as measured by an increased FET P value, than chromosome X (significance ranging from 0.006 to 0.045). The centromeric area was set as the reference for chromosomal region, and non-centromeric regions were found to have less BLD than the centromeric region, when controlling for the distance between markers and number of alleles at each marker; however, these regions were not significant at the 0.05 level. The interaction between chromosome and region was significant (at P < 0.05) only for two combinations, on chromosomes 14 and 22, both in mid-arm regions. For both of these chromosomes, in contrast to the pattern found on other chromosomes, the highest levels of BLD were found in the mid-arm regions, rather than in the centromeric region. Summaries of results by chromosome and region can be found in Table 1.
|
The FET P values from locus pairs on different chromosomes (88 110 of the 92 750 tests described in Materials and Methods) resulted in 5% of tests to be significant at the 0.05 level and 1% of tests to be significant at the 1% level. As this result matches that expected by chance, we have no reason to believe that factors other than genetic distance, such as recent admixture, contribute to the level of BLD we have described for linked markers.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
We found abundant BLD in our sample of 157 chromosomes from the CVCR, and observed such BLD over large genetic distances, in numerous cases over several centiMorgans. In agreement with theory, simulations and previous empirical work, BLD was significantly negatively related to the distance between the loci and significantly positively related to the number of alleles at the two loci. Controlling for genetic distance and number of alleles, the distribution of BLD was non-uniform both within and between chromosomes.
We found substantially more BLD than the single previous genome-wide survey (19). For example, we observed ~20% of locus pairs within 4 cM of each other to display a FET P value <0.01, compared with ~4% of such markers in the study of Huttley et al. (19). There are several important differences between the current study and that of Huttley et al. that probably contribute to this observation: (i) the present study had a larger sample size (157 versus 54 haplotypes); (ii) the individuals in the current study were drawn from a population sample in a genetically homogeneous, recently founded population, in contrast to those in the study of Huttley et al. who were drawn from a more heterogeneous pedigree-derived sample; (iii) in the current study the markers were chosen to be highly informative, wherease the previous study used the entire set of markers in the CEPH database. Factors (i) and (iii) directly affect the power of our sample to detect BLD. The sample used in the current study was more powerful not only because it was larger, but also because the polymorphic markers chosen for the genome screen would have greater ability to detect BLD than markers of lower heterozygosity. The relationship between marker heterozygosity and BLD has been shown by previous theoretical and empirical studies (13,18,19). Because the current sample was drawn from an isolated, recently founded population, we would expect BLD to be more widespread throughout the genome than in the CEPH samples.
Despite the differences in study design some of our findings were similar to those of Huttley et al. (19). We also saw that BLD was not uniformly distributed within or between chromosomes. In particular, both studies found less BLD on chromosome 15 and more BLD in the mid-arm of chromosome 22 than on other chromosomes. There are several possible explanations for the observed non-uniformity of BLD distribution in different chromosomal regions. One possibility, previously suggested by Peterson et al. (18), is that such non-uniformity is an artifact of inaccurate designation of genetic map distances in particular genome regions. It is also conceivable that microsatellite mutation rates vary between chromosomal regions; a high mutation rate at a given locus will obscure LD between that locus and its neighbors.
The extent of BLD in our study, as in all previous empirical studies, is much greater than that predicted by Kruglyak (7), and greater than that observed in empirical studies in limited genomic regions (8,9). Kruglyak’s work was based on population simulations, the assumptions of which may or may not have been appropriate for the actual populations studied. One advantage of our study is that the known demographic history of the population that we used can provide information for developing more accurate models of LD distribution. Also, Kruglyak simulated biallelic markers that will have less power to detect LD than more polymorphic markers.
We show that in a single recently founded genetic isolate LD occurs extensively throughout the genome. It has been suggested that such isolates may be particularly valuable for LD-mapping studies of complex traits (25). Our results suggest that in recently founded isolates, genome-wide LD mapping will be feasible even without dense marker maps. Our results also show, however, that unless accounted for explicitly, BLD will confuse the interpretation of LD analysis for mapping disease loci, as most methods of LD analysis assume linkage equilibrium between markers in control chromosomes. For LD mapping to succeed it will be necessary to develop more refined statistical methods for distinguishing disease-associated LD from BLD, by either accounting for observed BLD or modeling the population history through coalescent methods. For example, McPeek and Strahs (26) have suggested including BLD between adjacent markers in haplotype tests of association. We examined BLD in the non-transmitted chromosomes of parents of individuals recruited for a genetic mapping study. Whereas non-transmitted chromosomes are often used as matched controls in genetic mapping studies, if there is assortative mating for the phenotype being studied, it is possible that these non-transmitted chromosomes may not be ideal controls in that they may harbor disease predisposing loci as well. If this is true in our study, it is possible that the LD that we observed includes both a disease-associated LD signal along with BLD.
Our observations were made in a single, young population. To develop general strategies to adjust for BLD in mapping studies, two questions must be answered. First, is the inter- and intra-chromosomal variation in the distribution of BLD population specific or reflective of features of genome structure that are general between populations? For example, it is possible that there are hot- or cold-spots of recombination that are essentially specific to a particular population isolate. Second, to what degree is the average extent of BLD across the genome correlated with the age of a population as estimated by demographic history? Although our observations are consistent with our a priori assumption that the CVCR population is young and rapidly growing, data from several populations are needed to reach a clear conclusion on this question.
Freimer et al. (15) suggested previously that BLD should be studied in a number of populations using a reference set of closely linked, highly polymorphic microsatellite markers. This study is a step in that direction, as the markers are available in pre-set panels for automated genotyping, evenly spaced and all of a similarly high heterozygosity. We suggest that similar studies with these same markers in other population isolates (with similar and differing population histories) or recently admixed populations (with known demographic histories) will clarify factors that influence the distribution and magnitude of BLD, and will aid in the dissection of BLD from LD associated with disease susceptibility.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Subjects
We used haplotype data from the non-transmitted chromosomes of parents of probands recruited from the CVCR for a genetic mapping study of BP (Ophoff et al., unpublished data). The genealogy of all probands was investigated back to the level of their great-grandparents, and all probands included in the study of Ophoff et al. had at least six of the eight great-grandparents born in the CVCR. The population history of the CVCR and the suitability of this population isolate for genetic mapping are reviewed in Escamilla et al. (22). A total of 157 parents were genotyped, resulting in 157 non-transmitted chromosomes for this investigation of BLD.
Markers and genotyping
We genotyped 1036 fluorescently-labeled microsatellite markers from the Généthon set (27). The order and sex-averaged distance of the markers were based on the Marshfield map (28). The average spacing of the markers in our data set was <4 cM.
PCR was done using standard conditions using PE9700 instruments and PCR products were detected by using an ABI 377 sequencer and analyzed by Genescan and Genotyper software. All genotypes were independently double scored and consensus tables were created for each marker. The data were further checked for Mendelian inheritance using the UNKNOWN program.
Statistical methods
BLD between loci was evaluated using a Monte Carlo approximation to FET (29,30). This same procedure has been employed by others investigating the distribution of BLD in anonymous regions of the genome (11,14,18,19). The P value from the FET of BLD was not used to demonstrate statistical significance per se, but as an indication of the strength of BLD between loci. The P value from FET can be influenced by sample size, with smaller P values more likely with larger samples. As all of our pairwise tests used approximately the same sample size, this is not a confounding issue when interpreting our results, however, it limits comparisons with other datasets. We also calculated the multiallelic version of the disequilibrium statistic D’ (31) for comparison with other studies.
Tests of BLD were performed between all pairs of adjacent markers on all 23 chromosomes (1012 tests). That is, if marker order on a chromosome was A–B–C–D, BLD was assessed between markers A and B, B and C, C and D and so on. From these 1012 tests we first investigated the relationship between BLD and (i) the distance between the loci and (ii) the number of alleles at the two loci (measured as the dimension of the r x c table used in the FET). We constructed a binary random variable to identify locus pairs in BLD as pairs of loci for which the P value from FET was <0.05. As the average distance between markers in the genome screen was <4 cM, distance was dichotomized to be less or greater than 4 cM. The average number of alleles per marker in the screen was eight, therefore we made a dichotomous random variable that took the value 0 if the table dimension (the multiplication of the number of alleles at each locus) was <64 and 1 if table dimension was >64. We then used logistic regression techniques to model the probability of being in BLD as a function of distance between loci and number of alleles at each of the two loci. This analysis was performed using 180 loci pairs spread throughout the genome, with the requirement that pairs were 20 cM from each other, so that they would be roughly independent. These 180 pairs were selected by randomly choosing one of the first five marker pairs on each chromosome, and then selecting marker pairs every 20 cM over the length of the chromosome.
To assess whether there are differences between chromosomes or within chromosomes in the extent and distribution of BLD we employed the following procedure. Each chromosome was divided into three distinct regions. The centromeric region was the 
20 cM region around the centromere, as identified at http://www.ncbi.nlm.nih.gov/genemap99/. The telomeric region was the 15 cM region from pter and from qter (30 cM total). The rest of the chromosome comprised the third region, which we call mid-arm. We modeled the P value from the FET as a function of chromosome, region and their interaction. We also included terms for the distance between the loci and the number of alleles at each locus, to control for these factors known from previous studies to influence the probability of being in BLD. Because the P value is restricted to lie in the range 0–1, we did the analysis on the logit transformation of the P value. As data from all 1012 locus pairs were used in this analysis, P values from the same chromosomal region were not independent, especially since a given marker is used in the test of BLD with both markers on either side. This potential non-independence of P values from the same chromosome was initially addressed by using a model of exponential decay of the spatial dependence of the P values. We found, however, that the likelihood of the model when accounting for non-independence was not significantly greater than the likelihood when the 1012 locus pairs were treated independently, indicating very low correlation of adjacent FET P values.
Additionally, to test for BLD between unlinked loci we chose 100 autosomal loci at random, and examined BLD between these markers and all other autosomal markers (92 750 tests). As BLD between unlinked loci can occur in samples with significant admixture, these tests enable us to evaluate this potential in our sample and allow us to determine a baseline level of FET significance in our sample.
|
ACKNOWLEDGEMENTS |
|---|
We wish to thank the patients and their families who participated in this study, as well as the staff and administration of the National Psychiatric Hospital of Costa Rica and the Calderon Guardia Hospital. We would also like to thank Montgomery Slatkin, Lodewijk Sandkuijl and Chad Garner for comments and helpful discussion. This work was supported by grants MH49499 and MH01375 from the National Institutes of Health (to N.B.F.). Additional financial support was provided by Millennium Pharmaceuticals.
|
FOOTNOTES |
|---|
+ To whom correspondence should be addressed. Tel: +1 310 794 9571; Fax: +1 310 794 9613; Email: nfreimer@mednet.ucla.edu
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
1 Sheffield, V.C., Stone, E.M. and Carmi, R. (1998) Use of isolated inbred human populations for identification of disease genes. Trends Genet., 14, 391–396.[Web of Science][Medline]
2 Houwen, R.H.J., Baharloo, S., Blankenship, K., Raeymaekers, P., Juyn, J., Sandkuijl, L.A. and Freimer, N.B. (1994) Genome screening by searching for shared segments: mapping a gene for benign recurrent intrahepatic cholestasis. Nature Genet., 8, 380–386.
3 Friedman, T.B., Liang, Y., Weber, J.L., Hinnant, J.T., Barber, T.D., Winata, S., Nyoman Arhya, I. and Asher, J.H. (1995) A gene for congenital, recessive deafness DFNB3 maps to the pericentromeric region of chromosome 17. Nature Genet., 9, 86–91.[Web of Science][Medline]
4 Risch, N. and Merikangas, K. (1996) The future of genetic studies of complex human diseases. Science, 273, 1516–1517.
5 Escamilla, M.A., McInnes, L.A., Spesny, M., Reus, V.I., Service, S.K., Shimayoshi, N., Tyler, D.J., Silva, S., Molina, J., Gallegos, A. et al. (1999) Assessing the feasibility of linkage disequilibrium methods for mapping complex traits: an initial screen for bipolar disorder loci on chromosome 18. Am. J. Hum. Genet., 64, 1670–1678.[Web of Science][Medline]
6 Wright, A.F., Carothers, A.D. and Pirastu M. (1999) Population choice in mapping genes for complex diseases. Nature Genet., 23, 397–404.[Web of Science][Medline]
7 Kruglyak, L. (1999) Prospects for whole-genome linkage disequilibrium mapping of common disease genes. Nature Genet., 22, 139–144.[Web of Science][Medline]
8 Eaves, I.A., Merriman, T.R., Barber, R.A., Nutland, S., Tuomilehto-Wolf, E., Tuomilehto, J., Cucca, F. and Todd, J.A. (2000) The genetically isolated populations of Finland and Sardinia may not be a panacea for linkage disequilibrium mapping of common disease genes. Nature Genet., 25, 320–323.[Web of Science][Medline]
9 Dunning, A.M., Durocher, F., Healey, C.S., Teare, M.D., McBride, S.E., Carlomagno, F., Xu, C.-F., Dawson, E., Rhodes, S., Ueda, S. et al. (2000) The extent of linkage disequilibrium in four populations with distinct demographic histories. Am. J. Hum. Genet., 67, 1544–1554.
10 Abecasis, G.R., Noguchi, E., Heinzmann, A., Traherne, J.A., Bhattacharyya, S., Leaves, N.I., Anderson, G.G., Zhang, Y., Lench, N.J., Carey, A. et al. (2001) Extent and distribution of linkage disequilibrium in three genomic regions. Am. J. Hum. Genet. 68, 191–197.[Web of Science][Medline]
11 Zavattari, P., Deidda, E., Whalen, M., Lampis, R., Mulargia, A., Loddo, M., Eaves, I., Mastio, G., Todd, J.A. and Cucca, F. (2000) Major factors influencing linkage disequilibrium by analysis of different chromosome regions in distinct populations: demography, chromosome recombination frequency and selection. Hum. Mol. Genet., 9, 2947–2957.
12 Risch, N.J. (2000) Searching for genetic determinants in the new millennium. Nature, 405, 847–856.[Medline]
13 Slatkin, M. (1994) Linkage disequilibrium in growing and stable populations. Genetics, 137, 331–336.[Abstract]
14 Laan, M. and Pääbo, S. (1997) Demographic history and linkage disequilibrium in human populations. Nature Genet., 17, 435–438.[Web of Science][Medline]
15 Freimer, N.B., Service, S.K. and Slatkin, M. (1997) Expanding on population studies. Nature Genet., 17, 371–373.[Web of Science][Medline]
16 Hedrick, P.W, Klitz, W., Robinson, W.P, Kuhner, M.K. and Thomson, G. (1991) Population genetics of HLA. In Selander, R.K., Clark, A.G. and Whitam, T.S. (eds), Evolution at the Molecular Level. Sinauer Associates, Sunderland, MA, pp. 248–271.
17 Jorde, L.B. (1995) Linkage disequilibrium as a gene-mapping tool. Am. J. Hum. Genet., 56, 11–14.[Web of Science][Medline]
18 Peterson, A.C., Di Rienzo, A., Lehesjoki, A.-E., de la Chapelle, A., Slatkin, M. and Freimer, N.B. (1995) The distribution of linkage disequilibrium over anonymous genome regions. Hum. Mol. Genet., 4, 887–894.
19 Huttley, G.A., Smith, M.W., Carrington, M. and O’Brien, S.J. (1999) A scan for linkage disequilibrium across the human genome. Genetics, 152, 1711–1722.
20 Taillon-Miller, P., Bauer-Sardiña, I., Saccone, N.L., Putzel, J., Laitinen, T., Cao, A., Kere, J., Pilia, G., Rice, J.P. and Kwok, P.-Y. (2000) Juxtaposed regions of extensive and minimal linkage disequilibrium in human Xq25 and Xq28. Nature Genet., 25, 324–328.[Web of Science][Medline]
21 Boehnke, M. (2000) A look at linkage disequilibrium. Nature Genet., 25, 246–247.[Web of Science][Medline]
22 Escamilla, M.A., Spesny, M., Reus, V.I., Gallegos, A., Meza, L., Molina, J., Sandkuijl, L.A., Fournier, E., Leon, P.E., Smith, L.B. et al. (1996) Use of linkage disequilibrium approaches to map genes for bipolar disorder in the Costa Rican population. Am. J. Med. Genet., 67, 244–253.[Web of Science][Medline]
23 Saborio, M. (1992) Experience in providing genetic services in Costa Rica. Birth Defects, 28, 96–102.
24 Uhrhammer, N., Lange, E., Porras, O., Naeim, A., Chen, X., Sheikhavandi, S., Chiplunkar, S., Yang, L., Dandekar, S. and Liang, T. (1995) Sublocalization of an ataxia-telangiectasia gene distal to D11S384 by ancestral haplotyping in Costa Rican families. Am. J. Hum. Genet., 57, 102–111.
25 Peltonen, L., Palotie, A. and Lange, K. (2000) Use of population isolates for mapping complex traits. Nature Reviews, 1, 182–190.
26 McPeek, M.S. and Strahs, A. (1999) Assessment of linkage disequilibrium by the decay of haplotype sharing, with application to fine-scale genetic mapping. Am. J. Hum. Genet., 65, 858–875.[Web of Science][Medline]
27 Dib, C., Faure, S., Fizames, C., Samson, D., Drouot, N., Vignal, A., Millasseau, P., Marc, S., Hazan, J., Seboun, E. et al. (1996) A comprehensive genetic map of the human genome based on 5,264 microsatellites. Nature, 380, 152–154.[Medline]
28 Broman, K.W., Murray, J.C., Sheffield, V.C., White, R.L. and Weber, J.L. (1998) Comprehensive human genetic maps: individual and sex-specific variation in recombination. Am. J. Hum. Genet., 63, 861–869.[Web of Science][Medline]
29 Mehta, C.R. and Patel, N.R. (1983) A network algorithm for performing Fisher’s exact test in r x c contingency tables. J. Am. Stat. Assoc., 78, 427–434.[Web of Science]
30 Guo, S.W. and Thompson, E.A. (1992) Performing the exact test of Hardy-Weinberg proportion for multiple alleles. Biometrics, 48, 361–372.[Web of Science][Medline]
31 Hedrick, P.W. (1987) Gametic disequilibrium measures: proceed with caution. Genetics, 117, 331–341.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  B. A. Raby, M. E. Soto-Quiros, L. Avila, S. L. Lake, A. Murphy, C. Liang, E. Fournier, M. Spesny, J. S. Sylvia, A. Verner, et al. Sex-specific linkage to total serum immunoglobulin E in families of children with asthma in Costa Rica Hum. Mol. Genet., February 1, 2007; 16(3): 243 - 253. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. M. De La Vega, H. Isaac, A. Collins, C. R. Scafe, B. V. Halldorsson, X. Su, R. A. Lippert, Y. Wang, M. Laig-Webster, R. T. Koehler, et al. The linkage disequilibrium maps of three human chromosomes across four populations reflect their demographic history and a common underlying recombination pattern Genome Res., April 1, 2005; 15(4): 454 - 462. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. B. Sutter, M. A. Eberle, H. G. Parker, B. J. Pullar, E. F. Kirkness, L. Kruglyak, and E. A. Ostrander Extensive and breed-specific linkage disequilibrium in Canis familiaris Genome Res., December 1, 2004; 14(12): 2388 - 2396. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. L. Vallejo, Y. L. Li, G. W. Rogers, and M. S. Ashwell Genetic Diversity and Background Linkage Disequilibrium in the North American Holstein Cattle Population J Dairy Sci, December 1, 2003; 86(12): 4137 - 4147. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Rahman, A. Jones, J. Curtis, S. Bartlett, L. Peddle, B. A. Fernandez, and N. B. Freimer The Newfoundland population: a unique resource for genetic investigation of complex diseases Hum. Mol. Genet., October 15, 2003; 12(90002): R167 - 172. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Heutink and B. A. Oostra Gene finding in genetically isolated populations Hum. Mol. Genet., October 1, 2002; 11(20): 2507 - 2515. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. G. Schulze, Y.-S. Chen, N. Akula, K. Hennessy, J. A. Badner, M. G. McInnis, J. R. DePaulo, J. Schumacher, S. Cichon, P. Propping, et al. Can long-range microsatellite data be used to predict short-range linkage disequilibrium? Hum. Mol. Genet., June 1, 2002; 11(12): 1363 - 1372. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Menashe, O. Man, D. Lancet, and Y. Gilad Population differences in haplotype structure within a human olfactory receptor gene cluster Hum. Mol. Genet., June 1, 2002; 11(12): 1381 - 1390. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Tiret, O. Poirier, V. Nicaud, S. Barbaux, S.-M. Herrmann, C. Perret, S. Raoux, C. Francomme, G. Lebard, D. Tregouet, et al. Heterogeneity of linkage disequilibrium in human genes has implications for association studies of common diseases Hum. Mol. Genet., February 1, 2002; 11(4): 419 - 429. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L.B. Jorde, W.S. Watkins, and M.J. Bamshad Population genomics: a bridge from evolutionary history to genetic medicine Hum. Mol. Genet., October 1, 2001; 10(20): 2199 - 2207. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. A. McInnes, S. K. Service, V. I. Reus, G. Barnes, O. Charlat, S. Jawahar, S. Lewitzky, Q. Yang, Q. Duong, M. Spesny, et al. Fine-scale mapping of a locus for severe bipolar mood disorder on chromosome 18p11.3 in the Costa Rican population PNAS, September 25, 2001; 98(20): 11485 - 11490. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Sobacchi, A. Frattini, P. Orchard, O. Porras, I. Tezcan, M. Andolina, R. Babul-Hirji, I. Baric, N. Canham, D. Chitayat, et al. The mutational spectrum of human malignant autosomal recessive osteopetrosis Hum. Mol. Genet., August 1, 2001; 10(17): 1767 - 1773. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||