Strength Of Linkage Disequilibrium Research Articles

Interpretation of Manhattan Plots and Other Outputs of Genome-Wide Association Studies.

With increasing marker density, estimation of recombination rate between a marker and a causal mutation using linkage analysis becomes less important. Instead, linkage disequilibrium (LD) becomes the major indicator for gene mapping through genome-wide association studies (GWAS). In addition to the linkage between the marker and the causal mutation, many other factors may contribute to the LD, including population structure and cryptic relationships among individuals. As statistical methods and software evolve to improve statistical power and computing speed in GWAS, the corresponding outputs must also evolve to facilitate the interpretation of input data, the analytical process, and final association results. In this chapter, our descriptions focus on (1) considerations in creating a Manhattan plot displaying the strength of LD and locations of markers across a genome; (2) criteria for genome-wide significance threshold and the different appearance of Manhattan plots in single-locus and multiple-locus models; (3) exploration of population structure and kinship among individuals; (4) quantile-quantile (QQ) plot; (5) LD decay across the genome and LD between the associated markers and their neighbors; (6) exploration of individual and marker information on Manhattan and QQ plots via interactive visualization using HTML. The ultimate objective of this chapter is to help users to connect input data to GWAS outputs to balance power and false positives, and connect GWAS outputs to the selection of candidate genes using LD extent.

A Bivariate Hypothesis Testing Approach for Mapping the Trait-Influential Gene

The linkage disequilibrium (LD) based quantitative trait loci (QTL) model involves two indispensable hypothesis tests: the test of whether or not a QTL exists, and the test of the LD strength between the QTaL and the observed marker. The advantage of this two-test framework is to test whether there is an influential QTL around the observed marker instead of just having a QTL by random chance. There exist unsolved, open statistical questions about the inaccurate asymptotic distributions of the test statistics. We propose a bivariate null kernel (BNK) hypothesis testing method, which characterizes the joint distribution of the two test statistics in two-dimensional space. The power of this BNK approach is verified by three different simulation designs and one whole genome dataset. It solves a few challenging open statistical questions, closely separates the confounding between ‘linkage’ and ‘QTL effect’, makes a fine genome division, provides a comprehensive understanding of the entire genome, overcomes limitations of traditional QTL approaches, and connects traditional QTL mapping with the newest genotyping technologies. The proposed approach contributes to both the genetics literature and the statistics literature, and has a potential to be extended to broader fields where a bivariate test is needed.

Diversity and linkage disequilibrium in farmed Tasmanian Atlantic salmon.

Farmed Atlantic salmon (Salmo salar) is a globally important production species, including in Australia where breeding and selection has been in progress since the 1960s. The recent development of SNP genotyping platforms means genome-wide association and genomic prediction can now be implemented to speed genetic gain. As a precursor, this study collected genotypes at 218132 SNPs in 777 fish from a Tasmanian breeding population to assess levels of genetic diversity, the strength of linkage disequilibrium (LD) and imputation accuracy. Genetic diversity in Tasmanian Atlantic salmon was lower than observed within European populations when compared using four diversity metrics. The distribution of allele frequencies also showed a clear difference, with the Tasmanian animals carrying an excess of low minor allele frequency variants. The strength of observed LD was high at short distances (<25kb) and remained above background for marker pairs separated by large chromosomal distances (hundreds of kb), in sharp contrast to the European Atlantic salmon tested. Genotypes were used to evaluate the accuracy of imputation from low density (0.5 to 5K) up to increased density SNP sets (78K). This revealed high imputation accuracies (0.89-0.97), suggesting that the use of low density SNP sets will be a successful approach for genomic prediction in this population. The long-range LD, comparatively low genetic diversity and high imputation accuracy in Tasmanian salmon is consistent with known aspects of their population history, which involved a small founding population and an absence of subsequent introgression. The findings of this study represent an important first step towards the design of methods to apply genomics in this economically important population.

Approximated prediction of genomic selection accuracy when reference and candidate populations are related.

BackgroundGenomic selection is still to be evaluated and optimized in many species. Mathematical modeling of selection schemes prior to their implementation is a classical and useful tool for that purpose. These models include formalization of a number of entities including the precision of the estimated breeding value. To model genomic selection schemes, equations that predict this reliability as a function of factors such as the size of the reference population, its diversity, its genetic distance from the group of selection candidates genotyped, number of markers and strength of linkage disequilibrium are needed. The present paper aims at exploring new approximations of this reliability.ResultsTwo alternative approximations are proposed for the estimation of the reliability of genomic estimated breeding values (GEBV) in the case of non-independence between candidate and reference populations. Both were derived from the Taylor series heuristic approach suggested by Goddard in 2009. A numerical exploration of their properties showed that the series were not equivalent in terms of convergence to the exact reliability, that the approximations may overestimate the precision of GEBV and that they converged towards their theoretical expectations. Formulae derived for these approximations were simple to handle in the case of independent markers. A few parameters that describe the markers’ genotypic variability (allele frequencies, linkage disequilibrium) can be estimated from genomic data corresponding to the population of interest or after making assumptions about their distribution. When markers are not in linkage equilibrium, replacing the real number of markers and QTL by the “effective number of independent loci”, as proposed earlier is a practical solution. In this paper, we considered an alternative, i.e. an “equivalent number of independent loci” which would give a GEBV reliability for unrelated individuals by considering a sub-set of independent markers that is identical to the reliability obtained by considering the full set of markers.ConclusionsThis paper is a further step towards the development of deterministic models that describe breeding plans based on the use of genomic information. Such deterministic models carry low computational burden, which allows design optimization through intensive numerical exploration.Electronic supplementary materialThe online version of this article (doi:10.1186/s12711-016-0183-3) contains supplementary material, which is available to authorized users.

A crossbred reference population can improve the response to genomic selection for crossbred performance.

BackgroundBreeding goals in a crossbreeding system should be defined at the commercial crossbred level. However, selection is often performed to improve purebred performance. A genomic selection (GS) model that includes dominance effects can be used to select purebreds for crossbred performance. Optimization of the GS model raises the question of whether marker effects should be estimated from data on the pure lines or crossbreds. Therefore, the first objective of this study was to compare response to selection of crossbreds by simulating a two-way crossbreeding program with either a purebred or a crossbred training population. We assumed a trait of interest that was controlled by loci with additive and dominance effects. Animals were selected on estimated breeding values for crossbred performance. There was no genotype by environment interaction. Linkage phase and strength of linkage disequilibrium between quantitative trait loci (QTL) and single nucleotide polymorphisms (SNPs) can differ between breeds, which causes apparent effects of SNPs to be line-dependent. Thus, our second objective was to compare response to GS based on crossbred phenotypes when the line origin of alleles was taken into account or not in the estimation of breeding values.ResultsTraining on crossbred animals yielded a larger response to selection in crossbred offspring compared to training on both pure lines separately or on both pure lines combined into a single reference population. Response to selection in crossbreds was larger if both phenotypes and genotypes were collected on crossbreds than if phenotypes were only recorded on crossbreds and genotypes on their parents. If both parental lines were distantly related, tracing the line origin of alleles improved genomic prediction, whereas if both parental lines were closely related and the reference population was small, it was better to ignore the line origin of alleles.ConclusionsResponse to selection in crossbreeding programs can be increased by training on crossbred genotypes and phenotypes. Moreover, if the reference population is sufficiently large and both pure lines are not very closely related, tracing the line origin of alleles in crossbreds improves genomic prediction.Electronic supplementary materialThe online version of this article (doi:10.1186/s12711-015-0155-z) contains supplementary material, which is available to authorized users.

Identification of QTL for seed coat colour and oil content inBrassica napusby association mapping using SSR markers

Qu, C., Hasan, M., Lu, K., Liu, L., Zhang, K., Fu, F., Wang, M., Liu, S., Bu, H., Wang, R., Xu, X., Chen, L. and Li, J. 2015. Identification of QTL for seed coat colour and oil content in Brassica napus by association mapping using SSR markers. Can. J. Plant Sci. 95: 387–395. Association mapping identifies quantitative trait loci (QTL) based on the strength of linkage disequilibrium (LD) between markers and functional polymorphisms across a set of diverse germplasms. In this study, we used association mapping to detect QTL and genome-wide simple sequence repeat (SSR) markers linked to seed coat colour and oil content in a population of 217 oilseed rape (Brassica napus L.) accessions. We corrected for the population structure of B. napus using 389 genome-wide SSR markers. In total, 25 and 11 SSR markers linked to seed coat colour and oil content were detected, respectively, and these two sets of markers were in different linkage groups. Nine of these markers for seed coat colour spanned the major QTL region for seed coat colour, and been mapped to chromosome A9. Six of these markers showed high levels of association with both seed coat colour and oil content, and markers H081N08.8 and KS20291 were mapped to the major QTL region for seed coat colour on chromosome A9. Another marker, CB10364, was in high LD with all determined seed coat colour and oil content traits, and was mapped to the co-localized QTL region for them on chromosome A8. These data indicate that seed coat colour was found to be an important contributor to seed oil content. Further, we show that association mapping using a heterogeneous set of genotypes is a suitable approach for complementing and enhancing previously obtained QTL information for marker-assisted selection.

The effects of demography and long-term selection on the accuracy of genomic prediction with sequence data.

The use of dense SNPs to predict the genetic value of an individual for a complex trait is often referred to as "genomic selection" in livestock and crops, but is also relevant to human genetics to predict, for example, complex genetic disease risk. The accuracy of prediction depends on the strength of linkage disequilibrium (LD) between SNPs and causal mutations. If sequence data were used instead of dense SNPs, accuracy should increase because causal mutations are present, but demographic history and long-term negative selection also influence accuracy. We therefore evaluated genomic prediction, using simulated sequence in two contrasting populations: one reducing from an ancestrally large effective population size (Ne) to a small one, with high LD common in domestic livestock, while the second had a large constant-sized Ne with low LD similar to that in some human or outbred plant populations. There were two scenarios in each population; causal variants were either neutral or under long-term negative selection. For large Ne, sequence data led to a 22% increase in accuracy relative to ∼600K SNP chip data with a Bayesian analysis and a more modest advantage with a BLUP analysis. This advantage increased when causal variants were influenced by negative selection, and accuracy persisted when 10 generations separated reference and validation populations. However, in the reducing Ne population, there was little advantage for sequence even with negative selection. This study demonstrates the joint influence of demography and selection on accuracy of prediction and improves our understanding of how best to exploit sequence for genomic prediction.

Association of μ-opioid receptor gene (OPRM1) haplotypes with postoperative nausea and vomiting.

Genetic variants, such as single-nucleotide polymorphisms (SNPs), of the μ-opioid receptor gene (OPRM1) might be associated with individual differences in opioid sensitivity, as well as with the incidence and severity of postoperative nausea and vomiting (PONV). The goal of the present study was to determine, in a cohort of Japanese surgical patients, genotypes and haplotypes of several SNPs in the OPRM1 gene, and their association with PONV during the early (first 24 h) postoperative period. We examined the incidence and severity of PONV, during the first 24 h after surgery, in 85 Japanese patients receiving intravenous patient-controlled analgesia fentanyl analgesia for postoperative pain control. Eight tag SNPs of the OPRM1 gene (rs1799971, A/G; rs510769, G/A; rs4870266, G/A; rs3798683, G/A; rs1323042, A/C; rs609623, C/T; rs9397685, A/G; and rs644261, C/G) were selected based on their minor allele frequency (>10%) and linkage disequilibrium strength (<80%), and genotyped for haplotype analysis and determination of associations with PONV. Only one out of eight investigated SNPs, rs9397685, in the intronic part of the OPRM1 gene was associated with differences in the occurrence and severity of PONV. We also found four common haplotypes with a frequency of >10% in the investigated patients, including GGGAACAC (33%), AGGGACAC (19%), GGGAACGC (12%), and AGAGACAC (10%). The severity of PONV in carriers of the GGGAACGC haplotype was significantly lower than in the carriers of the other haplotypes (P < 0.05). One intronic SNP, rs9397685, and haplotypes constructed from eight SNPs within the OPRM1 gene locus might be involved in the severity of PONV associated with general anesthesia and opioid administration. This novel finding, if validated and verified in larger and additional ethnic cohorts, might contribute to better knowledge of the contribution of the OPRM1 gene to PONV.

Haplotype differences for copy number variants in the 22q11.23 region among human populations: a pigmentation-based model for selective pressure

Two gene clusters are tightly linked in a narrow region of chromosome 22q11.23: the macrophage migration inhibitory factor (MIF) gene family and the glutathione S-transferase theta class. Within 120 kb in this region, two 30-kb deletions reach high frequencies in human populations. This gives rise to four haplotypic arrangements, which modulate the number of genes in both families. The variable patterns of linkage disequilibrium (LD) between these copy number variants (CNVs) in diverse human populations remain poorly understood. We analyzed 2469 individuals belonging to 27 human populations with different ethnic origins. Then we correlated the genetic variability of 22q11.23 CNVs with environmental variables. We confirmed an increasing strength of LD from Africa to Asia and to Europe. Further, we highlighted strongly significant correlations between the frequency of one of the haplotypes and pigmentation-related variables: skin color (R(2)=0.675, P<0.001), distance from the equator (R(2)=0.454, P<0.001), UVA radiation (R(2)=0.439, P<0.001), and UVB radiation (R(2)=0.313, P=0.002). The fact that all MIF-related genes are retained on this haplotype and the evidences gleaned from experimental systems seem to agree with the role of MIF-related genes in melanogenesis. As such, we propose a model that explains the geographic and ethnic distribution of 22q11.23 CNVs among human populations, assuming that MIF-related gene dosage could be associated with adaptation to low UV radiation.

Association of melanocortin-4 receptor gene polymorphisms with obesity-related parameters in Malaysian Malays

Background: Melanocortin-4 receptor (MC4R) is an important regulator of body weight and energy intake. Genetic polymorphisms of the MC4R gene have been found to be linked to obesity in many recent studies across the globe.Aim: This study aimed to examine the effects of MC4R polymorphisms on obesity parameters, Linkage disequilibrium (LD) pattern and haplotypes in Malaysian Malays.Methods: The study subjects were 652 Malaysian Malays. Genomic DNA was extracted from buccal swabs. Genotyping was performed using Sequenom MassARRAY® iPLEX platform. Anthropometric and blood lipid profiles were measured.Results: MC4R rs571312 SNP was associated with logBMI (p = 0.008) and systolic blood pressure (p = 0.005), while MC4R rs2229616 SNP was associated with total cholesterol (TC) levels (p = 0.016). The MC4R rs7227255 SNP did not show any association with obesity parameters.Conclusions: The strength of LD of the MC4R gene region is low and the haplotypes were not associated with obesity in Malaysian Malays.

A Monte Carlo test of linkage disequilibrium for single nucleotide polymorphisms

BackgroundGenetic association studies, especially genome-wide studies, make use of linkage disequilibrium(LD) information between single nucleotide polymorphisms (SNPs). LD is also used for studying genome structure and has been valuable for evolutionary studies. The strength of LD is commonly measured by r2, a statistic closely related to the Pearson's χ2 statistic. However, the computation and testing of linkage disequilibrium using r2 requires known haplotype counts of the SNP pair, which can be a problem for most population-based studies where the haplotype phase is unknown. Most statistical genetic packages use likelihood-based methods to infer haplotypes. However, the variability of haplotype estimation needs to be accounted for in the test for linkage disequilibrium.FindingsWe develop a Monte Carlo based test for LD based on the null distribution of the r2 statistic. Our test is based on r2 and can be reported together with r2. Simulation studies show that it offers slightly better power than existing methods.ConclusionsOur approach provides an alternative test for LD and has been implemented as a R program for ease of use. It also provides a general framework to account for other haplotype inference methods in LD testing.

On the Uses and Applications of the Most Commonly Used Measures of Linkage Disequilibrium from the Comparative Analysis of Their Statistical Properties

Background/Objective: The analysis of linkage disequilibrium is relevant for the exploration of the structure and evolution of genomes and for the gene mapping of quantitative characters and human diseases. The strength of linkage disequilibrium between diallelic loci is commonly measured by the coefficients D′ and r. Recent studies suggest that r is more useful than D′ as a general measure of the strength of disequilibrium because it provides much more precise (lower sampling variance) and accurate (lower bias) estimates of disequilibrium. We compared for the first time the statistical properties of D′ and r taking into account their differences in range. Methods: The sampling properties of D′ and r were evaluated by simulation under a variety of realistic population conditions and varying sample sizes using standardised statistics that allow for comparisons of the precision, accuracy and efficiency of estimates with different ranges. Results: Simulations revealed that estimates of r do not tend to be significantly more precise, accurate or efficient than those of D′ when compared by means of standardised statistics. Conclusion: The supposed advantage of r over D′ based on direct comparisons of their sampling distributions is more apparent than real. The obtained results are useful to assess the uses and applications of these widely used disequilibrium measures.

Common variants at the GCK, GCKR, G6PC2–ABCB11 and MTNR1B loci are associated with fasting glucose in two Asian populations

To test fasting glucose association at four loci recently identified or verified by genome-wide association (GWA) studies of European populations, we performed a replication study in two Asian populations. We genotyped five common variants previously reported in Europeans: rs1799884 (GCK), rs780094 (GCKR), rs560887 (G6PC2-ABCB11) and both rs1387153 and rs10830963 (MTNR1B) in the general Japanese (n = 4,813) and Sri Lankan (n = 2,319) populations. To identify novel variants, we further examined genetic associations near each locus by using GWA scan data on 776 non-diabetic Japanese samples. Fasting glucose association was replicated for the five single nucleotide polymorphisms (SNPs) at p < 0.05 (one-tailed test) in South Asians (Sri Lankan) as well as in East Asians (Japanese). In fine-mapping by GWA scan data, we identified in the G6PC2-ABCB11 region a novel SNP, rs3755157, with significant association in Japanese (p = 2.6 x 10(-8)) and Sri Lankan (p = 0.001) populations. The strength of association was more prominent at rs3755157 than that of the original SNP rs560887, with allelic heterogeneity detected between the SNPs. On analysing the cumulative effect of associated SNPs, we found the per-allele gradients (beta = 0.055 and 0.069 mmol/l in Japanese and Sri Lankans, respectively) to be almost equivalent to those reported in Europeans. Fasting glucose association at four tested loci was proven to be replicable across ethnic groups. Despite this overall consistency, ethnic diversity in the pattern and strength of linkage disequilibrium certainly exists and can help to appreciably reduce potential causal variants after GWA studies.

Further Assessment of the Complement Component 2 and Factor B Region Associated with Age-Related Macular Degeneration

Polymorphic variation in genes involved in regulation of the complement system has been implicated as a major cause of genetic risk, in addition to the LOC387715/HTRA1 locus and other environmental influences. Previous studies have identified polymorphisms in the complement component 2 (CC2) and factor B (CFB) genes, as potential functional variants associated with AMD, in particular CFB R32Q and CC2 rs547154, both of which share strong linkage disequilibrium (LD). Data derived from the HapMap Project were used to select 18 haplotype-tagging SNPs across the extended CC2/CFB region for genotyping, to measure the strength of LD in 318 patients with neovascular AMD and 243 age-matched control subjects to identify additional potential functional variants in addition to those originally reported. Strong LD was measured across this region as far as the superkiller viralicidic activity 2-like gene (SKIV2L). Nine SNPs were identified to be significantly associated with the genetic effect observed at this locus. Of these, a nonsynonymous coding variant SKIV2L R151Q (rs438999; OR, 0.48; 95% confidence interval [CI], 0.31-0.74; P<0.001), was in strong LD with CFB R32Q, rs641153 (r(2)=0.95) and may exert a functional effect. When assessed within a logistic regression model measuring the effects of genetic variation at the CFH and LOC387715/HTRA1 loci and smoking, the effect remained significant (OR, 0.38; 95% CI, 0.22-0.65; P<0.001). Additional variation identified within this region may also confer a weaker but independent effect and implicate additional genes within the pathogenesis of AMD. Because of the high level of LD within the extended CC2/CFB region, variation within SKIV2L may exert a functional effect in AMD.

Extent of Linkage Disequilibrium in Holstein Cattle in North America

The success of fine-scale mapping and genomic selection depends mainly on the strength of linkage disequilibrium (LD) between markers and causal mutations. With Lewontin's measure of LD (known as D′), high levels of LD that extend over several million base pairs have been reported in livestock. However, this measure of LD can be strongly biased upward by small samples and by low allele frequencies. The aim of this study was to characterize the level and extent of LD in Holstein cattle in North America (Canada and the United States for purposes of this study) by using the squared correlation of the alleles at 2 loci (r2). The Affymetrix MegAllele GeneChip Bovine Mapping 10K single nucleotide polymorphism (SNP) array was used to genotype 821 bulls, from which 497 were used in the analysis of the extent of LD. A total of 5,564 SNP were used after filtering out SNP with more than 5% of Mendelian inconsistencies, with more than 20% missing genotypes, or with a minor allele frequency of less than 10%. Analysis of syntenic pairs revealed that useful LD (measured as r2>0.3) occurred at distances shorter than 100kb. Linkage disequilibrium decayed very rapidly, within a few hundred kilobase pairs. In addition, no substantial LD between unlinked loci was found. Using a sliding window analysis, we observed an irregular pattern of LD across the genome. These findings suggest that to capture useful LD, which is required for whole-genome fine mapping and genomic selection, a denser SNP map would be needed.

Linkage disequilibrium analysis in the genetically isolated Norfolk Island population

Norfolk Island is a human genetic isolate, possessing unique population characteristics that could be utilized for complex disease gene localization. Our intention was to evaluate the extent and strength of linkage disequilibrium (LD) in the Norfolk isolate by investigating markers within Xq13.3 and the NOS2A gene encoding the inducible nitric oxide synthase. A total of six microsatellite markers spanning approximately 11 Mb were assessed on chromosome Xq13.3 in a group of 56 men from Norfolk Island. Additionally, three single nucleotide polymorphisms (SNPs) localizing to the NOS2A gene were analyzed in a subset of the complex Norfolk pedigree. With the exception of two of the marker pairs, one of which is the most distantly spaced marker, all the Xq13.3 marker pairs were found to be in significant LD indicating that LD extends up to 9.5-11.5 Mb in the Norfolk Island population. Also, all SNPs studied showed significant LD in both Norfolk Islanders and Australian Caucasians, with two of the marker pairs in complete LD in the Norfolk population only. The Norfolk Island study population possesses a unique set of characteristics including founder effect, geographical isolation, exhaustive genealogical information and phenotypic data of use to cardiovascular disease risk traits. With LD extending up to 9.5-11 Mb, the Norfolk isolate should be a powerful resource for the localization of complex disease genes.

Detecting population structure using STRUCTURE software: effect of background linkage disequilibrium

STRUCTURE is the most widely used clustering software to detect population genetic structure. The last version of this software (STRUCTURE 2.1) has been enhanced recently to take into account the occurrence of linkage disequilibrium (LD) caused by admixture between populations. This last version, however, still does not consider the effects of strong background LD caused by genetic drift, and which may cause spurious results. STRUCTURE authors have, therefore, suggested a rough threshold value of the distance (1.0 cM) between two loci below which the pair of loci should not be used. Because of the sensitiveness of LD to demographic events, the distance between loci is not always a good indicator of the strength of LD. In this study, we examine the link between genomic distance and the strength of the correlation between loci (r(LD)) in a free-ranging population of mouflon (Ovis aries), and we present an empirical test of effect of r(LD) on the clustering results provided by the linkage model in STRUCTURE. We showed that a high r(LD) value increases the probability of detecting spurious clustering. We propose to use r(LD) as an index to base a decision on whether or not to use a pair of loci in a clustering analysis.

HLA-DRB1 Allele Frequency and TNF Polymorphism in Patients with Myelodysplastic Syndromes (MDS).

MDS, a heterogeneous group of hematopoietic stem cell disorders, are associated with dysregulated cytokine profiles in marrow and blood plasma, including upregulation of tumor necrosis factor alpha (TNF-alpha). Other data suggest that patients with MDS who carry the human leukocyte antigen (HLA)-DRB1*1501 allele are more likely to respond to immunosuppressive therapy. The genes for HLA-DR and TNF are in close linkage on human chromosome 6. Preliminary data had shown a higher than expected frequency of single nucleotide polymorphism (SNP) at position -238 of the TNF gene. We wanted to determine, therefore, the strength of linkage disequilibrium between a given TNF polymorphism and HLA-DR alleles. We analyzed HLA-DR allele frequencies in 1,040 patients with MDS, in comparison to 2,474 patients with chronic myeloid leukemia (CML); 1,422 and 160 of these patients, respectively, had TNF polymorphism data (AA vs. AC vs. CC alleles) for SNP 308 and 863 available. Significant differences in frequency were observed for HLA-DRB1*0401 (9.1% vs. 6.9%; p=0.03), *0701 (3.8% vs. 1.3%; p<0.0001), *1302 (2.7% vs. 4.3%; p=0.03), and *1501 (13.6% vs. 9.9%; p=0.0009). When determining the association of a given TNF SNP with the presence of two, one, or no copy of a given HLA-DR allele, TNF SNP 863 showed a significant association with HLA-DRB1*0101 (p=0.003), *0301 (p<0.0001), *0401 (p=0.0002), and *1101 (p=0.02); HLA-DRB1*0301 was also associated with TNF SNP 308 (p<0.0001). There was no significant correlation of either one of these two TNF SNPs with diagnosis. However, when only MDS patients were considered, there was a suggestion of a correlation of SNP 308 and HLA-DR*1501 (p=0.11); among patients with the TNF genotype AA or AC only 3 of 24 (12%) carried at least one DRB1*1501 allele, compared to 20 of 56 (36%) among those with the CC genotype. While TNF SNPs remain to be typed in a large proportion of MDS patients, results suggest a higher frequency of some HLA-DR alleles, in particular *0701 and *1501, in patients with MDS than seen in CML, and a correlation of TNF SNPs, in particular, TNF 308 CC with HLA-DRB1*1501. Thus, these data are consistent with an association of DRB1 *1501 and TNF alleles, and it will be of interest to identify the pattern of TNF polymorphism in HLA-DRB1*1501 negative patients, Additional studies are needed to further determine potential haplotypes that are associated with MDS and possible correlations with disease course and treatment responses.

Fetal Hemoglobin (HbF) in Sickle Cell Anemia: Genome-Wide Association Studies Using Pooled DNA Samples Can Reveal Genetic Associations with HbF Concentration.

The complex regulatory environment controlling HbF levels provides ample opportunity for modulating expression of this key modifier of sickle cell anemia (HbSS). Identifing genetic modulators by single gene or biochemical pathway-based candidate gene association studies is limited by the a priori hypothesis that they are responsible for modulation of HbF expression and thus, fail to capture the totality of possible modulators. In contrast, genome-wide association (GWA) studies using hundreds of thousands of SNPs across the entire genome can capture an unbiased assessment of gene association or linkage disequilibrium (LD) with a phenotype. A disadvantage of GWA is the cost when analyzing thousands of subjects. We developed a GWA screening strategy using pooled DNA samples followed by analysis using the Sentrix® HumanHap300 genotyping beadchip that includes over 317,000 tag SNPs, many derived from the Phase I of the HapMap project. These SNPs are within 10kb of a gene, or in regions that are evolutionary conserved, and incorporate the majority of the variation in regions of the genome with high LD by capturing approximately 80% of the loci genotyped in HapMap Phase I and II. DNA concentration for pooling was measured by RNAase-P assays and each pool was run in duplicate. Pools of approximately 60 DNA samples each were analyzed; pools included age, sex and β-globin gene haplotype-matched HbSS patients with the 1st and 4th quartiles of HbF levels. Raw data were processed to compute initial estimates of the allele frequencies. Estimates were normalized and SNPs with a minor allele frequency >0.15 were analyzed for allelic association using a Bayesian approach. Our analysis integrated the measure of association based on the allele frequencies, estimated from the pooled DNA samples, with the strength of LD between consecutive SNPs, information on minor allele frequencies derived from the HapMap project and an estimate of the variability between pools based on preliminary experimental data. By leveraging on this exogenous information we could reduce false positive associations without imposing unnecessarily stringent corrections for multiple comparisons that may lead to substantial loss of power. We identified more than 100 novel genes potentially involved with regulation of HbF and with roles in signal transduction, cell adhesion, and regulation of transcription. These genes are distributed unevenly among the whole genome, and chromosome 8 appears to be the most "enriched" for genes associated with regulation of HbF. Consistently with previous findings, our analysis confirms an involvement of genes in 6q22, 8q12 and 11p15.5 in the regulation of HbF. For further validation, we compared allele frequencies based on pooled DNA samples with those from individual genotyping of several SNPs. The high degree of agreement suggests that GWA based on pooled DNA samples can provide a cost-effective approach to the identification of the full array of genes responsible for HbF expression and can identify new candidates that must be further examined in follow-up genetic and biological studies.

Linkage Disequilibria and Haplotype Structure of Four SNPs of the Interleukin 1 Gene Cluster in Seven Asian Indian Populations

Variation at four single nucleotide polymorphism (SNP) sites of the interleukin 1 (IL1) gene cluster was investigated among 280 unrelated individuals, representing 7 caste groups from the state of Karnataka, India, and one European American community of Boston, Massachusetts. Allele and haplotype frequencies, strength of linkage disequilibrium, and signatures of recombination varied considerably among populations. Variable community sizes and traditions of consanguinity may account for the observed variation.