Estimated Haplotype Frequencies Research Articles

Haplotype association analysis of discrete and continuous traits using mixture of regression models.

We present a regression-based method of haplotype association analysis for quantitative and dichotomous traits in samples consisting of unrelated individuals. The method takes account of uncertain phase by initially estimating haplotype frequencies and obtaining the posterior probabilities of all possible haplotype combinations in each individual, then using these as weights in a finite mixture of regression models. Using this method, different combinations of marker loci can be modeled, to find a parsimonious set of marker loci that are most predictive and therefore most likely to be closely associated with the a quantitative trait locus. The method has the additional advantage of being able to use individuals with some missing genotype data, by considering all possible genotypes at the missing markers. We have implemented this method using the SNPHAP and Mx programs and illustrated its use on published data on idiopathic generalized epilepsy.

Estrogen receptor-α gene haplotype is associated with primary knee osteoarthritis in Korean population

Estrogen and estrogen receptors (ERs) are known to play important roles in the pathophysiology of osteoarthritis (OA). To investigate ER-α gene polymorphisms for its associations with primary knee OA, we conducted a case–control association study in patients with primary knee OA (n = 151) and healthy individuals (n = 397) in the Korean population. Haplotyping analysis was used to determine the relationship between three polymorphisms in the ER-α gene (intron 1 T/C, intron 1 A/G and exon 8 G/A) and primary knee OA. Genotypes of the ER-α gene polymorphism were determined by PCR followed by restriction enzyme digestion (PvuII for intron 1 T/C, XbaI for intron 1 A/G, and BtgI for exon 8 G/A polymorphism). There was no significant difference between primary knee OA patients and healthy control individuals in the distribution of any of the genotypes evaluated. However, we found that the allele frequency for the exon 8 G/A BtgI polymorphism (codon 594) was significantly different between primary knee OA patients and control individuals (odds ratio = 1.38, 95% confidence interval = 1.01–1.88; P = 0.044). In haplotype frequency estimation analysis, there was a significant difference between primary knee OA patients and control individuals (degrees of freedom = 7, χ2 = 21.48; P = 0.003). Although the number OA patients studied is small, the present study shows that ER-α gene haplotype may be associated with primary knee OA, and genetic variations in the ER-α gene may be involved in OA.

Chromosomal region 11p15 is associated with male factor subfertility.

The molecular aetiology of male factor subfertility, due to impaired spermatogenesis, is still unknown in the majority of cases. It is thought to be a complex disorder in which multiple genes are implicated. Cryptorchidism and reduced fecundity are symptoms in male Beckwith-Wiedemann patients and the ZNF214 and ZNF215 genes, localized on chromosomal region 11p15, are associated with this syndrome. We hypothesized that the ZNF214 and ZNF215 genes, which are predominantly expressed in the testis, could be involved in male factor subfertility in patients with idiopathic impaired spermatogenesis or in patients with impaired spermatogenesis due to cryptorchidism. Male partners of subfertile couples with idiopathic azoo- or severe oligozoospermia, male partners with azoo- or severe oligozoospermia and cryptorchidism in their medical history and men with normozoospermia were screened for nine single nucleotide polymorphisms in the ZNF214 and ZNF215 genes. An association study was performed based on allele and estimated haplotype frequencies. Statistically significant differences in allele frequencies and in estimated haplotype frequencies were found in both patient groups compared with controls. Thereafter, both genes were screened for mutations in all patients by PCR and single strand conformation polymorphism analysis. Aberrant patterns were confirmed by DNA sequencing. Mutation analysis in ZNF214 and ZNF215 revealed five new variants in the patients that were not present in the controls. At least three of these mutations were inherited from the mother. Our results suggest that chromosomal region 11p15 is associated with male factor subfertility due to impaired spermatogenesis with and without cryptorchidism.

Estimating haplotype frequencies and standard errors for multiple single nucleotide polymorphisms.

Estimating haplotype frequencies becomes increasingly important in the mapping of complex disease genes, as millions of single nucleotide polymorphisms (SNPs) are being identified and genotyped. When genotypes at multiple SNP loci are gathered from unrelated individuals, haplotype frequencies can be accurately estimated using expectation-maximization (EM) algorithms (Excoffier and Slatkin, 1995; Hawley and Kidd, 1995; Long et al., 1995), with standard errors estimated using bootstraps. However, because the number of possible haplotypes increases exponentially with the number of SNPs, handling data with a large number of SNPs poses a computational challenge for the EM methods and for other haplotype inference methods. To solve this problem, Niu and colleagues, in their Bayesian haplotype inference paper (Niu et al., 2002), introduced a computational algorithm called progressive ligation (PL). But their Bayesian method has a limitation on the number of subjects (no more than 100 subjects in the current implementation of the method). In this paper, we propose a new method in which we use the same likelihood formulation as in Excoffier and Slatkin's EM algorithm and apply the estimating equation idea and the PL computational algorithm with some modifications. Our proposed method can handle data sets with large number of SNPs as well as large numbers of subjects. Simultaneously, our method estimates standard errors efficiently, using the sandwich-estimate from the estimating equation, rather than the bootstrap method. Additionally, our method admits missing data and produces valid estimates of parameters and their standard errors under the assumption that the missing genotypes are missing at random in the sense defined by Rubin (1976).

Association of polymorphisms in the promoter region of the PNMT gene with essential hypertension in African Americans but not in whites

BackgroundSeveral studies have indicated that a region on human chromosome 17 may influence blood pressure. Our group reported positive linkage for hypertension to the region on human chromosome 17, between D17S1814 and D17S800 in white sibling pairs. In this study, we further investigated this result by examining the phenylethanolamine N-methyltransferase (PNMT) gene, which is located at 17q21 within the region where we found linkage. MethodsA case/control association study was conducted to evaluate the relationship between genetic variants of the PNMT gene and risk for essential hypertension. Two single nucleotide polymorphisms (SNPs) in the promoter region of the gene were genotyped, PNMT-148 and PNMT-353, in three ethnic samples: African American (117 hypertensive, 96 normotensive), American white (91 hypertensive, 80 normotensive), and Greek white (99 hypertensive, 90 normotensive), using the homogeneous mass extend reaction (Sequenom) and RFLP for genotyping. ResultsA significant difference in allelic frequency of SNP-353 between hypertensives (38.02%) and normotensives (27.35%) in African Americans (P = .019) was found; however, no significant differences were observed for this SNP for the other ethnic groups. No association was found with SNP PNMT-148 in any of the ethnic groups. Frequencies of haplotypes based on the two SNPs were also compared between hypertensive and normotensive individuals. No significant difference was found in estimated haplotype frequencies between hypertensive and control subjects in the three ethnic groups. ConclusionsThese results suggest that genetic variants of PNMT may play a role in the development of essential hypertension.

Methods to correct the HLA selection bias in haplotype frequency estimations using donor registry data

Methods to correct the HLA selection bias in haplotype frequency estimations using donor registry data

Analysis of the French National Registry of unrelated bone marrow donors, using surnames as a tool for improving geographical localisation of HLA haplotypes.

The first statistical analysis of the French National Registry of volunteer bone marrow donors estimated the probabilities of haplotype frequencies separately for each of the 20 administrative regions of France. Here we propose to use donors' surnames to increase the accuracy of location of the donor's geographical origin. This approach allows us to estimate haplotype frequencies for administrative entities (90 departments) smaller than regions and to correct for bias resulting from recent mobility. We analysed 30,777 donors typed for HLA-A,B and 17,745 donors typed for HLA-A,B,DR,DQ. By using the donors' surnames, we identified common and rare haplotypes (those found in only one department) and estimated the degree of HLA polymorphism at the department level. We also identified departments with a distinctive genetic structure (for example, Paris, Corsica, Pyrenees and Meurthe-et-Moselle). By providing a more accurate geographical distribution of HLA polymorphisms in France, this study will enable us to optimise policies for recruiting bone marrow donors and to improve the fit between the donor file and patients' needs.

Linkage disequilibrium assessment via log-linear modeling of SNP haplotype frequencies.

Analyses of high-density single-nucleotide polymorphism (SNP) data, such as genetic mapping and linkage disequilibrium (LD) studies, require phase-known haplotypes to allow for the correlation between tightly linked loci. However, current SNP genotyping technology cannot determine phase, which must be inferred statistically. In this paper, we present a new Bayesian Markov chain Monte Carlo (MCMC) algorithm for population haplotype frequency estimation, particularly in the context of LD assessment. The novel feature of the method is the incorporation of a log-linear prior model for population haplotype frequencies. We present simulations to suggest that 1) the log-linear prior model is more appropriate than the standard coalescent process in the presence of recombination (>0.02 cM between adjacent loci), and 2) there is substantial inflation in measures of LD obtained by a "two-stage" approach to the analysis by treating the "best" haplotype configuration as correct, without regard to uncertainty in the recombination process.

High-throughput single-nucleotide polymorphism analysis of the IL1RN locus in patients with ankylosing spondylitis by matrix-assisted laser desorption ionization-time-of-flight mass spectrometry.

To examine single-nucleotide polymorphisms (SNPs) in the 3' region of the IL1RN gene in a large Caucasoid case-control series and in ankylosing spondylitis (AS) families from Western Canada by use of high-throughput MassArray matrix-assisted laser desorption ionization-time-of-flight (MALDI-TOF) mass spectrometry SNP typing. An association analysis was performed in a case-control cohort of 394 AS cases and 500 controls. Family-based association analysis was performed in 58 simplex and 13 multiplex families. Three SNPs located in the 3' region of the IL1RN gene (T/C at position 27810 in exon 4, T/C at position 30735 in exon 6, and G/C at position 31017 in exon 6) were examined by high-throughput MassArray MALDI-TOF mass spectrometry. Haplotype inference software programs were used to infer the most likely haplotypes and to compare haplotype frequencies, which were then further analyzed in family-based association studies by transmission disequilibrium tests. The frequency of allele C at SNP position 30735 in exon 6 was significantly increased in AS cases (35.1%) versus controls (27.8%), as was the phenotype frequency (61.7% versus 48.6%). A significantly increased frequency of SNP allele G at position 31017 in exon 6 in cases (32.9%) versus controls (28.3%) was also noted. A highly significant difference in the overall distribution of haplotype frequencies was evident between cases and controls, with significant increases in the frequencies of the 27810C/30735C/31017C and 27810C/30735T/31017G haplotypes, but a significant reduction in the estimated frequency of the 27810C/30735T/31017C haplotype, in the AS cases. Estimation of haplotype frequencies based on 2 SNP markers indicated a highly significant increase in the 30735C/31017C haplotype and a highly significant decrease in the 30735T/31017C haplotype in cases compared with controls. Preliminary evidence for reduced transmission of the 27810C/30735T/31017C 3-marker haplotype was also found in family-based association analyses. Our data establish a highly significant disease association with markers in the IL1RN gene. In the absence of nonsynonymous coding sequence substitutions, it is possible that the primary disease-associated locus regulates gene expression. Association with specific haplotypes raises the possibility that the primary disease locus is in linkage disequilibrium with a specific combination(s) of markers in the IL1RN gene.

Nonparametric disequilibrium mapping of functional sites using haplotypes of multiple tightly linked single-nucleotide polymorphism markers.

As the speed and efficiency of genotyping single-nucleotide polymorphisms (SNPs) increase, using the SNP map, it becomes possible to evaluate the extent to which a common haplotype contributes to the risk of disease. In this study we propose a new procedure for mapping functional sites or regions of a candidate gene of interest using multiple linked SNPs. Based on a case-parent trio family design, we use expectation-maximization (EM) algorithm-derived haplotype frequency estimates of multiple tightly linked SNPs from both unambiguous and ambiguous families to construct a contingency statistic S for linkage disequilibrium (LD) analysis. In the procedure, a moving-window scan for functional SNP sites or regions can cover an unlimited number of loci except for the limitation of computer storage. Within a window, all possible widths of haplotypes are utilized to find the maximum statistic S* for each site (or locus). Furthermore, this method can be applied to regional or genome-wide scanning for determining linkage disequilibrium using SNPs. The sensitivity of the proposed procedure was examined on the simulated data set from the Genetic Analysis Workshop (GAW) 12. Compared with the conventional and generalized TDT methods, our procedure is more flexible and powerful.

Evolutionary-based association analysis using haplotype data.

Association studies, both family-based and population-based, can be powerful means of detecting disease-liability alleles. To increase the information of the test, various researchers have proposed targeting haplotypes. The larger number of haplotypes, however, relative to alleles at individual loci, could decrease power because of the additional degrees of freedom required for the test. An optimal strategy would focus the test on particular haplotypes or groups of haplotypes, much as is done with cladistic-based association analysis. First suggested by Templeton et al. ([1987] Genetics 117:343-351), such analyses use the evolutionary relationships among haplotypes to produce a limited set of hypothesis tests and to increase the interpretability of these tests. To more fully utilize the information contained in the evolutionary relationships among haplotypes and in the sample, we propose generalized linear models (GLM) for the analysis of data from family-based and population-based studies. These models fully account for haplotype phase ambiguity and allow for covariates. The models are encoded into a software package (the Evolutionary-Based Haplotype Analysis Package, EHAP), which also provides for various kinds of exploratory data analysis. The exploratory analyses, such as error checking, estimation of haplotype frequencies, and tools for building cladograms, should facilitate the implementation of cladistic-based association analysis with haplotypes.

Two exonic single nucleotide polymorphisms in the microsomal epoxide hydrolase gene are associated with polycystic ovary syndrome

Objective To determine whether genetic variability in the gene encoding microsomal epoxide hydrolase (EPHX) contributes to individual differences in susceptibility to the development of polycystic ovary syndrome (PCOS). Design Retrospective case-control study. Setting University-based clinic. Patient(s) One hundred twelve white women with PCOS and 115 healthy controls. Intervention(s) None. Main outcome measure(s) The presence of two single nucleotide polymorphisms (SNPs), T→C (Tyr113His) in exon 3 and A→G (His139Arg) in exon 4, in the EPHX gene. Single point analysis was expanded to pair of loci haplotype analysis to examine the estimated haplotype frequencies of the two SNPs, of unknown phase, in the PCOS and control groups. Estimated haplotype frequencies were assessed using the maximum-likelihood method, using an expectation–maximization algorithm. Result(s) Single point allele and genotype distributions in exon 3 and exon 4 of the EPHX gene were not statistically different between the groups. However, according to the haplotype estimation analysis, we observed a significantly elevated frequency of haplotype C-G (His113–Arg139) in the PCOS group versus the control group. The odds ratio for PCOS associated with the low activity haplotype C-G (His113–Arg139) was 2.28 (95% confidence interval 1.1–4.8). Conclusion(s) The use of two intragenic single nucleotide polymorphisms jointly in haplotype analysis of association demonstrated that the genetically determined low activity haplotype C-G (His113–Arg139) was significantly associated with PCOS.

Efficiency of single-nucleotide polymorphism haplotype estimation from pooled DNA.

The efficiency of single-nucleotide polymorphism haplotype analysis may be increased by DNA pooling, which can dramatically reduce the number of genotyping assays. We develop a method for obtaining maximum likelihood estimates of haplotype frequencies for different pool sizes, assess the accuracy of these estimates, and show that pooling DNA samples is efficient in estimating haplotype frequencies. Although pooling K individuals increases ambiguities, at least for small pool size K and small numbers of loci, the uncertainty of estimation increases <K times that of unpooled DNA. We also develop the asymptotic variance-covariance of maximum likelihood estimates and evaluate the accuracy of variance estimates by Monte Carlo methods. When the sample size of pools is moderately large, the asymptotic variance estimates are rather accurate. Completely or partially missing genotyping information is allowed for in our analysis. Finally, our methods are applied to single-nucleotide polymorphisms in the angiotensinogen gene.

Serotonin transporter gene promoter (5-HTTLPR) and intron 2 (VNTR) polymorphisms in Croatian suicide victims

Background Disturbances in serotonin (5HT) transmission are the most frequently reported neurobiological substrates of suicidal behavior. Because 5HT transporter plays a central role in the regulation of 5HT synaptic function and its gene contains two functional polymorphisms (5-HTTLPR in the promoter region and VNTR in the second intron), it represents an interesting candidate for association studies in suicidal behavior. Methods In this study, a possible association of 5-HTTLPR and intron 2-VNTR polymorphisms of the 5HT transporter gene with suicidal behavior was investigated in a sample of 135 suicide victims and 299 healthy control subjects of Croatian/southern Slavic origin. Results There were no significant differences in 5-HTTLPR and intron 2-VNTR genotype- and allele- frequency distributions between suicide victims and healthy control subjects; however, a tendency toward an increase of 5-HTTLPR allele L and VNTR-allele 10 were observed in suicide group. Analysis of distribution of estimated haplotype frequencies revealed differences between suicide victims and control subjects, with an excess of haplotype L10 among suicide victims ( p = .0112). Conclusions Our results provide modest evidence for a possible association of the 5HT transporter gene with a completed suicide. Further studies are needed to determine whether alterations in 5HTt gene expression are involved in suicidal behavior.

Systematic search for single nucleotide polymorphisms in the 5 ′ flanking region of the human phosphodiesterase 3B gene: absence of evidence for major effects of identified polymorphisms on susceptibility to Japanese type 2 diabetes

The activation of phosphodiesterase 3B (PDE3B) reduces free fatty acid output from adipocytes. A reduced PDE3B gene expression could lead to insulin resistance. To determine whether there are polymorphisms associated with type 2 diabetes in PDE3B gene promoter, this 5 ′ flanking region was isolated. The transcription initiation site was located 206 bp upstream from the translation start site. Sequences of 2 kb of the 5 ′ flanking region for 24 type 2 diabetic Japanese subjects were initially analyzed using PCR direct sequencing, and the regions including the identified polymorphisms were then examined. In 98 controls and 98 type 2 diabetic subjects, −1947T > C, −567G > A, −465G > T, −458T > C, and −1727_−1726insTCAATT were found. Only −465G > T and this insertion had more than 5% frequencies. Since a complete linkage disequilibrium existed between them, −465G > T was further analyzed, along with a previously identified +1389G > A in the coding region, in a total of 200 controls and 207 type 2 diabetic subjects. These allele frequencies were not significantly different between these two groups (controls vs. cases; −465G > T, 12.0% vs. 10.1%, P=0.435; +1389G > A, 30.3% vs. 33.3%, P=0.408). These genotype distributions were not significantly different between these two groups. The T/T genotype at −465 was rare although this frequency could be higher in type 2 diabetes (4/207 subjects) than controls (0/200 subjects). The linkage disequilibrium existed between −465G > T and +1389G > A, and the estimated haplotype frequencies defined by these SNPs were not significantly different between the cases and controls. Thus, the identified polymorphisms are unlikely to have major effects on susceptibility to Japanese type 2 diabetes.

Estimation of multilocus haplotype effects using weighted penalised log-likelihood: analysis of five sequence variations at the cholesteryl ester transfer protein gene locus.

Direct analyses of haplotype effects can be used to identify those specific combinations of alleles that are associated with a specific phenotype. We introduce a method for direct haplotype analysis that solves two problems that arise when haplotypes are analysed in populations of unrelated subjects. Instead of assigning a single, most likely, haplotype pair to multiple heterozygous subjects, all haplotype pairs compatible with their genotype were determined and the posterior probabilities of these pairs were calculated using Bayes' theorem and estimated haplotype frequencies. For the individual patients, all possible haplotype pairs were included in the statistical analysis using the posterior probabilities as weights, which were re-estimated in an iterative process together with the haplotype effects. The second problem of unstable haplotype effect estimates, due to the numerous haplotypes and the low frequency at which some occur, was solved by assuming that haplotypes sharing the same alleles show a similar effect and that the extent of this similarity relates to the number of alleles shared. These assumptions were incorporated in a weighted log-likelihood model by introducing a penalty, where differences in effects of similar haplotypes were penalised. Using CETP gene haplotypes, consisting of five closely linked polymorphisms, and baseline CETP and HDL-C concentrations from the REGRESS population, we demonstrated that the model resulted in more stable effects than estimates based on unambiguous patients only.

Extensive Linkage Disequilibrium, a Common 16.7-Kilobase Deletion, and Evidence of Balancing Selection in the Human Protocadherin α Cluster

Regions of extensive linkage disequilibrium (LD) appear to be a common feature of the human genome. However, the mechanisms that maintain these regions are unknown. In an effort to understand whether gene density contributes to LD, we determined the degree of promoter sequence variation in a large tandem-arrayed gene family, the human protocadherin alpha cluster, on chromosome 5. These genes are expressed at synaptic junctions in the developing brain and the adult brain and may be involved in the determination of synaptic complexity. We sequenced the promoters of all 13 alpha protocadherin genes in 96 European Americans and identified polymorphisms in the promoters alpha 1, alpha 3, alpha 4, alpha 5, alpha 7, alpha 9, alpha 11, and alpha 13. In these promoters, 11 common SNPs are in extensive LD, forming two 48-kb haplotypes of equal frequency, in this population, that extend from the alpha1 through alpha 7 genes. We sequenced these promoters in East Asians and African Americans, and we estimated haplotype frequencies and calculated LD statistics for all three populations. Our results indicate that, although extensive LD is an ancient feature of the alpha cluster, it has eroded over time. SNPs 3' of alpha 7 are involved in ancestral recombination events in all populations, and overall alpha-cluster LD is reduced in African Americans. We obtained significant positive values for Tajima's D test for all alpha promoter SNPs in Europeans (D=3.03) and East Asians (D=2.64), indicating an excess of intermediate-frequency variants, which is a signature of balancing selection. We also discovered a 16.7-kb deletion that truncates the alpha 8 gene and completely removes the alpha 9 and alpha 10 genes. This deletion appears in unaffected individuals from multiple populations, suggesting that a reduction in protocadherin gene number is not obviously deleterious.

Development of rheumatoid arthritis is not associated with two polymorphisms in the Crohn's disease gene CARD15.

It has been proposed that genetic susceptibility loci for rheumatoid arthritis (RA) may be shared with other autoimmune/inflammatory diseases. Recently, common variation in the CARD15 (NOD2) gene on chromosome 16q12 has been associated with Crohn's disease (CD) in several independent populations. CARD15 is an excellent functional and positional candidate gene for RA. Genomic DNA was obtained from 392 RA cases and 471 ethnically matched healthy controls. All samples were genotyped for two polymorphisms in CARD15, 1007fs and R702W, using 5' nuclease reporter assays. Allele frequencies were compared between cases and controls using the chi(2) test. Estimated haplotype frequencies across the two mutations were determined using the EH program. The allele frequency of the 1007fs variant in RA cases was 1.8% compared with 1.6% in normal controls (not significant). The frequency of the R702W variant was 4.0% in both cases and controls. Haplotypes carrying either of the two mutations accounted for 5.6% of possible haplotypes. A haplotype carrying both mutations was rare, with estimated frequency <0.01%. This study provided high power to detect an association of similar magnitude to that in Crohn's disease. These data therefore exclude the possibility that the contribution of these mutations to RA is comparable to that seen in CD. Within defined statistical parameters, we excluded a role for the CARD15 1007fs and R702W variants in RA susceptibility. These data do not preclude a role for other polymorphisms in the CARD15 gene in RA susceptibility. Results from other autoimmune and inflammatory diseases will reveal whether the CARD15 gene is in fact a common autoimmune susceptibility locus.

Estimation of Haplotype Frequencies, Linkage-Disequilibrium Measures, and Combination of Haplotype Copies in Each Pool by Use of Pooled DNA Data

Inference of haplotypes is important for many genetic approaches, including the process of assigning a phenotype to a genetic region. Usually, the population frequencies of haplotypes, as well as the diplotype configuration of each subject, are estimated from a set of genotypes of the subjects in a sample from the population. We have developed an algorithm to infer haplotype frequencies and the combination of haplotype copies in each pool by using pooled DNA data. The input data are the genotypes in pooled DNA samples, each of which contains the quantitative genotype data from one to six subjects. The algorithm infers by the maximum-likelihood method both frequencies of the haplotypes in the population and the combination of haplotype copies in each pool by an expectation-maximization algorithm. The algorithm was implemented in the computer program LDPooled. We also used the bootstrap method to calculate the standard errors of the estimated haplotype frequencies. Using this program, we analyzed the published genotype data for the SAA (n=156), MTHFR (n=80), and NAT2 (n=116) genes, as well as the smoothelin gene (n=102). Our study has shown that the frequencies of major (frequency >0.1 in a population) haplotypes can be inferred rather accurately from the pooled DNA data by the maximum-likelihood method, although with some limitations. The estimated D and D' values had large variations except when the /D/ values were >0.1. The estimated linkage-disequilibrium measure rho2 for 36 linked loci of the smoothelin gene when one- and two-subject pool protocols were used suggested that the gross pattern of the distribution of the measure can be reproduced using the two-subject pool data.

Choosing Haplotype-Tagging SNPS Based on Unphased Genotype Data Using a Preliminary Sample of Unrelated Subjects with an Example from the Multiethnic Cohort Study

We describe an approach for picking haplotype-tagging single nucleotide polymorphisms (htSNPs) that is presently being taken in two large nested case-control studies within a multiethnic cohort (MEC), which are engaged in a search for associations between risk of prostate and breast cancer and common genetic variations in candidate genes. Based on a preliminary sample of 70 control subjects chosen at random from each of the 5 ethnic groups in the MEC we estimate haplotype frequencies using a variant of the Excoffier-Slatkin E-M algorithm after genotyping a high density of SNPs selected every 3–5 kb in and surrounding a candidate gene. In order to evaluate the performance of a candidate set of htSNPS (which will be genotyped in the much larger case-control sample) we treat the haplotype frequencies estimate above as known, and carry out a formal calculation of the uncertainty of the number of copies of common haplotypes carried by an individual, summarizing this calculation as a coefficient of determination, R²_h. A candidate set of htSNPS of a given size is chosen so as to maximize the minimum value of R²_h over the common haplotypes, h.