Direct IBD mapping: identical-by-descent mapping without genotyping

Direct identical-by-descent (IBD) mapping is a technique, that combines genomic mismatch scanning (GMS) and DNA microarray technology, for mapping regions shared IBD between two individuals without locus-by-locus genotyping or sequencing. The lack of reagents has limited its widespread application. In particular, two key reagents have been limiting, 1). mismatch repair proteins MutS, L and H, and 2). genomic microarrays for identifying the genomic locations of the GMS-selected IBD fragments. Here, we describe steps that optimized the procedure and resources that will facilitate the development of direct IBD mapping.     

Comparative analysis of haplotype association mapping algorithms

Background  

Finding the genetic causes of quantitative traits is a complex and difficult task. Classical methods for mapping quantitative trail loci (QTL) in miceuse an F2 cross between two strains with substantially different phenotype and an interval mapping method to compute confidence intervals at each position in the genome. This process requires significant resources for breeding and genotyping, and the data generated are usually only applicable to one phenotype of interest. Recently, we reported the application of a haplotype association mapping method which utilizes dense genotyping data across a diverse panel of inbred mouse strains and a marker association algorithm that is independent of any specific phenotype. As the availability of genotyping data grows in size and density, analysis of these haplotype association mapping methods should be of increasing value to the statistical genetics community.     

Robust multipoint simultaneous identical-by-descent mapping for two linked loci

A challenging issue in genetic mapping of complex human diseases is localizing disease susceptibility genes when the genetic effects are small to moderate. There are greater complexities when multiple loci are linked to a chromosomal region. Liang et al. [Hum Hered 2001;51:64-78] proposed a robust multipoint method that can simultaneously estimate both the position of a trait locus and its effect on disease status by using affected sib pairs (ASPs). Based on the framework of generalized estimating equations (GEEs), the estimate and standard error of the position of a trait locus are robust to different genetic models. To utilize other relative pairs collected in pedigree data, Schaid et al. [Am J Hum Genet 2005;76:128-138] extended Liang's method to various types of affected relative pairs (ARPs) by two approaches: unconstrained and constrained methods. However, the above methods are limited to situations in which only one trait locus exists on the chromosome of interest. The mean functions are no longer correctly specified when there are multiple causative loci linked to a chromosomal region. To overcome this, Biernacka et al. [Genet Epidemiol 2005;28:33-47] considered the multipoint methods for ASPs to allow for two linked disease genes. We further generalize the approach to cover other types of ARPs. To reflect realistic situations for complex human diseases, we set modest sizes of genetic effects in our simulation. Our results suggest that several hundred independent pedigrees are needed, and markers with high information, to provide reliable estimates of trait locus positions and their confidence intervals. Bootstrap resampling can correct the downward bias of the robust variance for location estimates. These methods are applied to a prostate cancer linkage study on chromosome 20 and compared with the results for the one-locus model [Am J Hum Genet 2005;76:128-138]. We have implemented the multipoint IBD mapping for one and two linked loci in our software GEEARP, which allows analyses for five general types of ARPs.     

Robust multipoint identical-by-descent mapping for affected relative pairs

The genetic mapping of complex traits has been challenging and has required new statistical methods that are robust to misspecified models. Liang et al. proposed a robust multipoint method that can be used to simultaneously estimate, on the basis of sib-pair linkage data, both the position of a trait locus on a chromosome and its effect on disease status. The advantage of their method is that it does not require specification of an underlying genetic model, so estimation of the position of a trait locus on a specified chromosome and of its standard error is robust to a wide variety of genetic mechanisms. If multiple loci influence the trait, the method models the marginal effect of a locus on a specified chromosome. The main critical assumption is that there is only one trait locus on the chromosome of interest. We extend this method to different types of affected relative pairs (ARPs) by two approaches. One approach is to estimate the position of a trait locus yet allow unconstrained trait-locus effects across different types of ARPs. This robust approach allows for differences in sharing alleles identical-by-descent across different types of ARPs. Some examples for which an unconstrained model would apply are differences due to secular changes in diagnostic methods that can change the frequency of phenocopies among different types of relative pairs, environmental factors that modify the genetic effect, epistasis, and variation in marker-information content. However, this unconstrained model requires a parameter for each type of relative pair. To reduce the number of parameters, we propose a second approach that models the marginal effect of a susceptibility locus. This constrained model is robust for a trait caused by either a single locus or by multiple loci without epistasis. To evaluate the adequacy of the constrained model, we developed a robust score statistic. These methods are applied to a prostate cancer-linkage study, which emphasizes their potential advantages and limitations.     

Multivariate eQTL mapping uncovers functional variation on the X-chromosome associated with complex disease traits

Very few studies have investigated the associations between genetic polymorphisms and gene expression on the X-chromosome. This is a major bottleneck when conducting functional follow-up studies of trait-associated variants, as those identified in genome-wide association studies (GWAS). We used a multivariate approach to test the association between individual single nucleotide polymorphisms (SNPs) and exon expression levels measured in 356 Epstein–Barr virus-transformed lymphoblastoid cell lines (LCLs) from the Geuvadis RNA sequencing project to identify SNPs associated with variation in gene expression on the X-chromosome, which we refer to as eSNPs. At an FDR of 5 %, we discovered 548 independent [linkage disequilibrium (LD) r ² < 0.1] eSNPs on the X-chromosome. Of these, 35 were in LD (r ² > 0.8) with previously published disease- or trait-associated variants identified through GWAS. One of the strongest eSNPs identified was rs35975601, which was associated with F8A1 expression (p value = 3 × 10^?20) and was in LD with a type 1 diabetes risk variant. Additionally, we identified a number of genes for which eSNPs were in LD with multiple diseases or traits, including DNASE1L1 which was mapped to bilirubin levels, type 1 diabetes and schizophrenia. Our results also indicate that multivariate exon-level analysis provides a more powerful approach than univariate gene-level analysis, particularly when SNPs influence the expression of different exons with different magnitude and/or direction of effect. The associations identified in our study may provide new insights into the molecular process by which gene expression may contribute to trait variation or disease risk in humans.     

Tree scanning: a method for using haplotype trees in phenotype/genotype association studies

We use evolutionary trees of haplotypes to study phenotypic associations by exhaustively examining all possible biallelic partitions of the tree, a technique we call tree scanning. If the first scan detects significant associations, additional rounds of tree scanning are used to partition the tree into three or more allelic classes. Two worked examples are presented. The first is a reanalysis of associations between haplotypes at the Alcohol Dehydrogenase locus in Drosophila melanogaster that was previously analyzed using a nested clade analysis, a more complicated technique for using haplotype trees to detect phenotypic associations. Tree scanning and the nested clade analysis yield the same inferences when permutation testing is used with both approaches. The second example is an analysis of associations between variation in various lipid traits and genetic variation at the Apolipoprotein E (APOE) gene in three human populations. Tree scanning successfully identified phenotypic associations expected from previous analyses. Tree scanning for the most part detected more associations and provided a better biological interpretative framework than single SNP analyses. We also show how prior information can be incorporated into the tree scan by starting with the traditional three electrophoretic alleles at APOE. Tree scanning detected genetically determined phenotypic heterogeneity within all three electrophoretic allelic classes. Overall, tree scanning is a simple, powerful, and flexible method for using haplotype trees to detect phenotype/genotype associations at candidate loci.     

New hope for haplotype mapping

The systematic analysis of polymorphisms across large parts of the human genome has begun to provide the first information on haplotypes and the problem of linkage disequilibrium across large genomic regions. These data suggest that significant regions of the genome show highly conserved haplotypes, potentially enhancing the ability to detect disease associations.     

Likelihood-based disequilibrium mapping for two-marker haplotype data

We report a theory that gives the sampling distribution of two-marker haplotypes that are linked to a rare disease mutation. The sampling distribution is generated with successive Monte Carlo realizations of the coalescence of the disease mutation having recombination and marker mutation events placed along the lineage. Given a sample of mutation-bearing, two-marker haplotypes, the maximum likelihood estimate of the location of the disease mutation can be calculated from the generated sampling distribution, provided that one knows enough about the population history in order to model it. The two-marker likelihood method is compared to a single-marker likelihood and a composite likelihood. The two-marker maximum likelihood gives smaller confidence intervals for the location of the disease locus than a comparable single-marker maximum likelihood. The composite likelihood can give biased results and the bias increases as the extent of linkage disequilibrium on mutation-bearing chromosomes decreases. Haplotype configurations exist for which the composite likelihood will fail to place the disease locus in the correct marker interval.     

Identification of a 3.7-Mb region for a marbling QTL on bovine chromosome 4 by identical-by-descent and association analysis

QTL mapping for growth and carcass traits was performed using a paternal half-sib family composed of 325 Japanese Black cattle offspring. Nine QTL were detected at the 1% chromosome-wise significance level at a false discovery rate of less than 0.1. These included two QTL for marbling on BTA 4 and 18, two QTL for carcass weight on BTA 14 and 24, two QTL for longissimus muscle area on BTA 1 and 4, two QTL for subcutaneous fat thickness on BTA 1 and 15 and one QTL for rib thickness on BTA 6. Although the marbling QTL on BTA 4 has been replicated with significant linkages in two Japanese Black cattle sires, the three Q (more marbling) haplotypes, each inherited maternally, were apparently different. To compare the three Q haplotypes in more detail, high-density microsatellite markers for the overlapping regions were developed within the 95% CIs (65 markers in 44–78 cM). A detailed haplotype comparison indicated that a small region (<3.7 Mb) around 46 cM was shared between the Qs of the two sires, whose dams were related. An association of this region with marbling was shown by a regression analysis using the local population, in which the two sires were produced and this was confirmed by an association study using a population collected throughout Japan. These results strongly suggest that the marbling QTL on BTA 4 is located in the 3.7-Mb region at around 46 cM.     

Leukotriene-related gene polymorphisms in ASA-intolerant asthma: an association with a haplotype of 5-lipoxygenase

A recent study has demonstrated the possible involvement of a leukotriene C4 synthase (LTC4S) gene polymorphism in ASA-intolerant asthma (AIA) in a Polish population, whereas no significant association was noted in other populations. To investigate the role of genetic polymorphism in AIA development, we screened single nucleotide polymorphisms (SNPs) of the key enzymes involved in arachidonate metabolism, and the cysteinyl leukotriene receptor 1 (CYSLTR1) in a large Korean population with AIA: 93 AIA and 181 ASA-tolerant asthma (ATA) patients, and 123 normal controls. The single-base extension method was used to genotype SNPs in 5-lipoxygenase (ALOX5, –1708GA, 21CT, 270GA, 1728GA), ALOX5-activating protein (ALOX5AP, 218AG), prostaglandin-endoperoxide synthase 2 (PTGS2, COX2, –162CG, 10TG, R228H, V511A), LTC4S (–444AC), and CYSLTR1 (927TC). Haplotype analyses were undertaken for the SNPs in ALOX5. No significant differences in allele and genotype frequencies of single SNPs were observed between the patient groups (P>0.05). However, the frequency of the ALOX5-ht1[G-C-G-A] haplotype in the AIA group was significantly higher than its frequency in the ATA group with a probability (P) of 0.01, odds ratio (OR) of 5.0, and 95% confidence interval (95%CI) of 1.54–17.9, and in the normal controls (P=0.03, OR=4.5, 95%CI=1.1–18.4), by using a dominant model. These results suggest a lack of association between the ALOX5AP, PTGS2, LTC4S, and CYSLTR1 gene polymorphisms and the AIA phenotype in the Korean population. However, the possible involvement of ALOX5-ht1[G-C-G-A] in AIA development is suggested.J.-H. Choi and H.-S. Park contributed equally to this work as first authors     

Efficient mining of haplotype patterns for linkage disequilibrium mapping

Effective identification of disease-causing gene locations can have significant impact on patient management decisions that will ultimately increase survival rates and improve the overall quality of health care. Linkage disequilibrium mapping is the process of finding disease gene locations through comparisons of haplotype frequencies between disease chromosomes and normal chromosomes. This work presents a new method for linkage disequilibrium mapping. The main advantage of the proposed algorithm, called LinkageTracker, is its consistency in producing good predictive accuracy under different conditions, including extreme conditions where the occurrence of disease samples with the mutation of interest is very low and there is presence of error or noise. We compared our method with some leading methods in linkage disequilibrium mapping such as HapMiner, Blade, GeneRecon, and Haplotype Pattern Mining (HPM). Experimental results show that for a substantial class of problems, our method has good predictive accuracy while taking reasonably short processing time. Furthermore, LinkageTracker does not require any population ancestry information about the disease and the genealogy of the haplotypes. Therefore, it is useful for linkage disequilibrium mapping when the users do not have such information about their datasets.     

Identification of genetic variation and haplotype structure of the canine ABCA4 gene for retinal disease association studies

Over 200 mutations in the retina specific member of the ATP-binding cassette transporter superfamily (ABCA4) have been associated with a diverse group of human retinal diseases. The disease mechanisms, and genotype–phenotype associations, nonetheless, remain elusive in many cases. As orthologous genes are commonly mutated in canine models of human blinding disorders, canine ABCA4 appears to be an ideal candidate gene to identify and study sequence changes in dogs affected by various forms of inherited retinal degeneration. However, the size of the gene and lack of haplotype assignment significantly limit targeted association and/or linkage approaches. This study assessed the naturally observed sequence diversity of ABCA4 in the dog, identifying 80% of novel variations. While none of the observed polymorphisms have been associated with blinding disorders to date, breed and potentially disease specific haplotypes have been identified. Moreover, a tag SNP map of 17 (15) markers has been established that accurately predicts common ABCA4 haplotypes (frequency > 5%) explaining >85% (>80%) of the observed genetic diversity and will considerably advance future studies. Our sequence analysis of the complete canine ABCA4 coding region will clearly provide a baseline and tools for future association studies and comparative genomics to further delineate the role of ABCA4 in canine blinding disorders.     

Genetic variation and association mapping of silica concentration in rice hulls using a germplasm collection

An association analysis on the genetic variability for silica concentration in rice hulls was performed using a "Mini-Core" set of 174 accessions representative of the germplasm diversity found in the USDA world collection of rice. Hull silica concentration was determined in replicated trials conducted in two southern states in the USA and was analyzed for its association with 164 genome-wide DNA markers. Among the accessions, the average silica concentration ranged from 120 to 251?mg?g(-1). Ample variation was seen within each of the five sub-populations of rice, as well as the 14 geographic regions that the accessions originated from. There was also an effect due to location and accession?×?location (G?×?E) interaction demonstrating the importance of assessing silica concentration across multiple environments. Twelve markers on ten chromosomes were significantly associated with hull silica concentration. Six markers (RM5644, RM5371, RM1335, RM283, RM263, and RM178) corroborated quantitative trait locus for silica concentration identified in other mapping studies. Our results provide germplasm and genetic markers that will assist breeding efforts to develop cultivars that have either high or low hull silica concentration. High silica hulls are good raw material for silica based industrial compounds, while low silica hulls are more biodegradable.     

Nucleotide variation,haplotype structure,and association with end-stage renal disease of the human interleukin-1 gene cluster

A dense gene-based SNP map was constructed across a 360-kb region containing the interleukin-1 gene cluster (IL1A, IL1B, and IL1RN), focusing on IL1RN. In total, 95 polymorphisms were confirmed or identified primarily by direct sequencing. Polymorphisms were precisely mapped to completed BAC and genomic sequences spanning this region. The polymorphisms were typed in 443 case-control subjects from Caucasian and African American groups. Consecutive pair-wise marker linkage disequilibrium was not strictly correlated with distance and ranged from D'=0.0079 to 1.000 and D'=0.0521 to 1.0000 in Caucasians and African Americans, respectively. Single markers and haplotypes in IL1 cluster genes were evaluated for association with end-stage renal disease (ESRD). Eleven SNPs show some evidence of association with ESRD, with the strongest associations in two IL1A variants, one SNP, rs1516792-3, in intron 5 (p=0.0015) and a 4-bp insertion/deletion within the 3'UTR, rs16347-2 (p=0.0024), among African Americans with non-T2DM-associated ESRD.     

Chemical genetic interrogation of natural variation uncovers a molecule that is glycoactivated

Natural variation in human drug metabolism and target genes can cause pharmacogenetic or interindividual variation in drug sensitivity. We reasoned that natural pharmacogenetic variation in model organisms could be systematically exploited to facilitate the characterization of new small molecules. To test this, we subjected multiple Arabidopsis thaliana accessions to chemical genetic screens and discovered 12 accession-selective hit molecules. As a model for understanding this variation, we characterized natural resistance to hypostatin, a new inhibitor of cell expansion. Map-based cloning identified HYR1, a UDP glycosyltransferase (UGT), as causative for hypostatin resistance. Multiple lines of evidence demonstrate that HYR1 glucosylates hypostatin in vivo to form a bioactive glucoside. Additionally, we delineated a HYR1 substrate motif and used it to identify another molecule modulated by glucosylation. Our results demonstrate that natural variation can be exploited to inform the biology of new small molecules, and that UGT sequence variation affects xenobiotic sensitivity across biological kingdoms.