Single-locus Approaches Research Articles

Gene, pathway and network frameworks to identify epistatic interactions of single nucleotide polymorphisms derived from GWAS data

BackgroundInteractions among genomic loci (also known as epistasis) have been suggested as one of the potential sources of missing heritability in single locus analysis of genome-wide association studies (GWAS). The computational burden of searching for interactions is compounded by the extremely low threshold for identifying significant p-values due to multiple hypothesis testing corrections. Utilizing prior biological knowledge to restrict the set of candidate SNP pairs to be tested can alleviate this problem, but systematic studies that investigate the relative merits of integrating different biological frameworks and GWAS data have not been conducted.ResultsWe developed four biologically based frameworks to identify pairwise interactions among candidate SNP pairs as follows: (1) for each human protein-coding gene, a set of SNPs associated with that gene was constructed providing a gene-based interaction model, (2) for each known biological pathway, a set of SNPs associated with the genes in the pathway was constructed providing a pathway-based interaction model, (3) a set of SNPs associated with genes in a disease-related subnetwork provides a network-based interaction model, and (4) a framework is based on the function of SNPs. The last approach uses expression SNPs (eSNPs or eQTLs), which are SNPs or loci that have defined effects on the abundance of transcripts of other genes. We constructed pairs of eSNPs and SNPs located in the target genes whose expression is regulated by eSNPs. For all four frameworks the SNP sets were exhaustively tested for pairwise interactions within the sets using a traditional logistic regression model after excluding genes that were previously identified to associate with the trait. Using previously published GWAS data for type 2 diabetes (T2D) and the biologically based pair-wise interaction modeling, we identify twelve genes not seen in the previous single locus analysis.ConclusionWe present four approaches to detect interactions associated with complex diseases. The results show our approaches outperform the traditional single locus approaches in detecting genes that previously did not reach significance; the results also provide novel drug targets and biomarkers relevant to the underlying mechanisms of disease.

Increasing power for voxel-wise genome-wide association studies: the random field theory, least square kernel machines and fast permutation procedures.

Imaging traits are thought to have more direct links to genetic variation than diagnostic measures based on cognitive or clinical assessments and provide a powerful substrate to examine the influence of genetics on human brains. Although imaging genetics has attracted growing attention and interest, most brain-wide genome-wide association studies focus on voxel-wise single-locus approaches, without taking advantage of the spatial information in images or combining the effect of multiple genetic variants. In this paper we present a fast implementation of voxel- and cluster-wise inferences based on the random field theory to fully use the spatial information in images. The approach is combined with a multi-locus model based on least square kernel machines to associate the joint effect of several single nucleotide polymorphisms (SNP) with imaging traits. A fast permutation procedure is also proposed which significantly reduces the number of permutations needed relative to the standard empirical method and provides accurate small p-value estimates based on parametric tail approximation. We explored the relation between 448,294 single nucleotide polymorphisms and 18,043 genes in 31,662 voxels of the entire brain across 740 elderly subjects from the Alzheimer's disease neuroimaging initiative (ADNI). Structural MRI scans were analyzed using tensor-based morphometry (TBM) to compute 3D maps of regional brain volume differences compared to an average template image based on healthy elderly subjects. We find method to be more sensitive compared with voxel-wise single-locus approaches. A number of genes were identified as having significant associations with volumetric changes. The most associated gene was GRIN2B, which encodes the N-methyl-d-aspartate (NMDA) glutamate receptor NR2B subunit and affects both the parietal and temporal lobes in human brains. Its role in Alzheimer's disease has been widely acknowledged and studied, suggesting the validity of the approach. The various advantages over existing approaches indicate a great potential offered by this novel framework to detect genetic influences on human brains.

Reprioritizing Genetic Associations in Hit Regions Using LASSO‐Based Resample Model Averaging

Significance testing one SNP at a time has proven useful for identifying genomic regions that harbor variants affecting human disease. But after an initial genome scan has identified a “hit region” of association, single-locus approaches can falter. Local linkage disequilibrium (LD) can make both the number of underlying true signals and their identities ambiguous. Simultaneous modeling of multiple loci should help. However, it is typically applied ad hoc: conditioning on the top SNPs, with limited exploration of the model space and no assessment of how sensitive model choice was to sampling variability. Formal alternatives exist but are seldom used. Bayesian variable selection is coherent but requires specifying a full joint model, including priors on parameters and the model space. Penalized regression methods (e.g., LASSO) appear promising but require calibration, and, once calibrated, lead to a choice of SNPs that can be misleadingly decisive. We present a general method for characterizing uncertainty in model choice that is tailored to reprioritizing SNPs within a hit region under strong LD. Our method, LASSO local automatic regularization resample model averaging (LLARRMA), combines LASSO shrinkage with resample model averaging and multiple imputation, estimating for each SNP the probability that it would be included in a multi-SNP model in alternative realizations of the data. We apply LLARRMA to simulations based on case-control genome-wide association studies data, and find that when there are several causal loci and strong LD, LLARRMA identifies a set of candidates that is enriched for true signals relative to single locus analysis and to the recently proposed method of Stability Selection. Genet. Epidemiol. 36:451–462, 2012. © 2012 Wiley Periodicals, Inc.

A species assemblage approach to comparative phylogeography of birds in southern Australia

We present a novel approach to investigating the divergence history of biomes and their component species using single-locus data prior to investing in multilocus data. We use coalescent-based hierarchical approximate Bayesian computation (HABC) methods (MsBayes) to estimate the number and timing of discrete divergences across a putative barrier and to assign species to their appropriate period of co-divergence. We then apply a coalescent-based full Bayesian model of divergence (IMa) to suites of species shown to have simultaneously diverged. The full Bayesian model results in reduced credibility intervals around divergence times and allows other parameters associated with divergence to be summarized across species assemblages. We apply this approach to 10 bird species that are wholly or patchily discontinuous in semi-arid habitats between Australia's southwest (SW) and southeast (SE) mesic zones. There was substantial support for up to three discrete periods of divergence. HABC indicates that two species wholly restricted to more mesic habitats diverged earliest, between 594,382 and 3,417,699 years ago, three species from semi-arid habitats diverged between 0 and 1,508,049 years ago, and four diverged more recently, between 0 and 396,843 years ago. Eight species were assigned to three periods of co-divergence with confidence. For full Bayesian analyses, we accounted for uncertainty in the two remaining species by analyzing all possible suites of species. Estimates of divergence times from full Bayesian divergence models ranged between 429,105 and 2,006,355; 67,172 and 663,837; and 24,607 and 171,085 for the earliest, middle, and most recent periods of co-divergence, respectively. This single-locus approach uses the power of multitaxa coalescent analyses as an efficient means of generating a foundation for further, targeted research using multilocus and genomic tools applied to an understudied biome.

Comparative study of statistical methods for detecting association with rare variants in exome-resequencing data

Genome-wide association studies for complex traits are based on the common disease/common variant (CDCV) and common disease/rare variant (CDRV) assumptions. Under the CDCV hypothesis, classical genome-wide association studies using single-marker tests are powerful in detecting common susceptibility variants, but under the CDRV hypothesis they are not as powerful. Several methods have been recently proposed to detect association with multiple rare variants collectively in a functional unit such as a gene. In this paper, we compare the relative performance of several of these methods on the Genetic Analysis Workshop 17 data. We evaluate these methods using the unrelated individual and family data sets. Association was tested using 200 replicates for the quantitative trait Q1. Although in these data the power to detect association is often low, our results show that collapsing methods are promising tools. However, we faced the challenge of assessing the proper type I error to validate our power comparisons. We observed that the type I error rate was not well controlled; however, we did not find a general trend specific to each method. Each method can be conservative or nonconservative depending on the studied gene. Our results also suggest that collapsing and the single-locus association approaches may not be affected to the same extent by population stratification. This deserves further investigation.

Missing heritability and missing Fst of candidate genes: why does gene variation differ from trait variation in trees?

Molecular footprints of phenotypic variation are usually explored by two different approaches in trees. On the one hand, association genetics is seeking for statistical correlation between SNPs and traits owing to the significant heritability that is usually observed in progeny tests. On the other hand, detection of outlier Fst values of single SNPs has also been implemented to account for the very large differentiation of traits that are observed in provenance tests. The rationale of both approaches is based on the extensive within or between population genetic variability that has been widely recorded in fields tests. While the cataloguing of candidate genes has steadily increased in trees, so has the inventory of their diversity in natural or breeding populations. There is now a rapidly growing body of experimental results showing a very large discrepancy between the expectations of both approaches and the SNP variation that is monitored in candidate genes. Most association studies show that single SNPs explain usually less than 5% of the phenotypic variation of the trait, while Fst values are at best of the same value than neutral markers. In my presentation I will explore the reasons of the decoupling between trait and gene variation, by focusing on the multilocus structure of traits, as compared to the monolocus SNP information. Indeed, association studies and Fst outlier detection are essentially based on single locus approaches, while traits are multidimensional structures. There are at least three properties of multilocus structures that will be investigated: cumulativity, interactions and covariation of gene effects. I will show that cumulativity and interactions may explain the discrepancy in the case of association studies, while covariation of gene effects (at the between population level) explain the missing Fst of genes underlying adaptive traits. These conclusions are supported by experimental results, theoretical background and simulation predictions.

A Network-Based Approach to Prioritize Results from Genome-Wide Association Studies

Genome-wide association studies (GWAS) are a valuable approach to understanding the genetic basis of complex traits. One of the challenges of GWAS is the translation of genetic association results into biological hypotheses suitable for further investigation in the laboratory. To address this challenge, we introduce Network Interface Miner for Multigenic Interactions (NIMMI), a network-based method that combines GWAS data with human protein-protein interaction data (PPI). NIMMI builds biological networks weighted by connectivity, which is estimated by use of a modification of the Google PageRank algorithm. These weights are then combined with genetic association p-values derived from GWAS, producing what we call ‘trait prioritized sub-networks.’ As a proof of principle, NIMMI was tested on three GWAS datasets previously analyzed for height, a classical polygenic trait. Despite differences in sample size and ancestry, NIMMI captured 95% of the known height associated genes within the top 20% of ranked sub-networks, far better than what could be achieved by a single-locus approach. The top 2% of NIMMI height-prioritized sub-networks were significantly enriched for genes involved in transcription, signal transduction, transport, and gene expression, as well as nucleic acid, phosphate, protein, and zinc metabolism. All of these sub-networks were ranked near the top across all three height GWAS datasets we tested. We also tested NIMMI on a categorical phenotype, Crohn’s disease. NIMMI prioritized sub-networks involved in B- and T-cell receptor, chemokine, interleukin, and other pathways consistent with the known autoimmune nature of Crohn’s disease. NIMMI is a simple, user-friendly, open-source software tool that efficiently combines genetic association data with biological networks, translating GWAS findings into biological hypotheses.

Detecting epistatic effects in association studies at a genomic level based on an ensemble approach

Motivation: Most complex diseases involve multiple genes and their interactions. Although genome-wide association studies (GWAS) have shown some success for identifying genetic variants underlying complex diseases, most existing studies are based on limited single-locus approaches, which detect single nucleotide polymorphisms (SNPs) essentially based on their marginal associations with phenotypes.Results: In this article, we propose an ensemble approach based on boosting to study gene–gene interactions. We extend the basic AdaBoost algorithm by incorporating an intuitive importance score based on Gini impurity to select candidate SNPs. Permutation tests are used to control the statistical significance. We have performed extensive simulation studies using three interaction models to evaluate the efficacy of our approach at realistic GWAS sizes, and have compared it with existing epistatic detection algorithms. Our results indicate that our approach is valid, efficient for GWAS and on disease models with epistasis has more power than existing programs.Contact: jingli@case.edu

DASH: A Method for Identical-by-Descent Haplotype Mapping Uncovers Association with Recent Variation

Rare variants affecting phenotype pose a unique challenge for human genetics. Although genome-wide association studies have successfully detected many common causal variants, they are underpowered in identifying disease variants that are too rare or population-specific to be imputed from a general reference panel and thus are poorly represented on commercial SNP arrays. We set out to overcome these challenges and detect association between disease and rare alleles using SNP arrays by relying on long stretches of genomic sharing that are identical by descent. We have developed an algorithm, DASH, which builds upon pairwise identical-by-descent shared segments to infer clusters of individuals likely to be sharing a single haplotype. DASH constructs a graph with nodes representing individuals and links on the basis of such segments spanning a locus and uses an iterative minimum cut algorithm to identify densely connected components. We have applied DASH to simulated data and diverse GWAS data sets by constructing haplotype clusters and testing them for association. In simulations we show this approach to be significantly more powerful than single-marker testing in an isolated population that is from Kosrae, Federated States of Micronesia and has abundant IBD, and we provide orthogonal information for rare, recent variants in the outbred Wellcome Trust Case-Control Consortium (WTCCC) data. In both cohorts, we identified a number of haplotype associations, five such loci in the WTCCC data and ten in the isolated, that were conditionally significant beyond any individual nearby markers. We have replicated one of these loci in an independent European cohort and identified putative structural changes in low-pass whole-genome sequence of the cluster carriers.

Tools for efficient epistasis detection in genome-wide association study

BackgroundGenome-wide association study (GWAS) aims to find genetic factors underlying complex phenotypic traits, for which epistasis or gene-gene interaction detection is often preferred over single-locus approach. However, the computational burden has been a major hurdle to apply epistasis test in the genome-wide scale due to a large number of single nucleotide polymorphism (SNP) pairs to be tested.ResultsWe have developed a set of three efficient programs, FastANOVA, COE and TEAM, that support epistasis test in a variety of problem settings in GWAS. These programs utilize permutation test to properly control error rate such as family-wise error rate (FWER) and false discovery rate (FDR). They guarantee to find the optimal solutions, and significantly speed up the process of epistasis detection in GWAS.ConclusionsA web server with user interface and source codes are available at the website http://www.csbio.unc.edu/epistasis/. The source codes are also available at SourceForge http://sourceforge.net/projects/epistasis/.

Analysis of North American Rheumatoid Arthritis Consortium data using a penalized logistic regression approach

We applied a penalized regression approach to single-nucleotide polymorphisms in regions on chromosomes 1, 6, and 9 of the North American Rheumatoid Arthritis Consortium data. Results were compared with a standard single-locus association test. Overall, the penalized regression approach did not appear to offer any advantage with respect to either detection or localization of disease-associated polymorphisms, compared with the single-locus approach.

Fine-scale phylogenetic discordance across the house mouse genome.

Population genetic theory predicts discordance in the true phylogeny of different genomic regions when studying recently diverged species. Despite this expectation, genome-wide discordance in young species groups has rarely been statistically quantified. The house mouse subspecies group provides a model system for examining phylogenetic discordance. House mouse subspecies are recently derived, suggesting that even if there has been a simple tree-like population history, gene trees could disagree with the population history due to incomplete lineage sorting. Subspecies of house mice also hybridize in nature, raising the possibility that recent introgression might lead to additional phylogenetic discordance. Single-locus approaches have revealed support for conflicting topologies, resulting in a subspecies tree often summarized as a polytomy. To analyze phylogenetic histories on a genomic scale, we applied a recently developed method, Bayesian concordance analysis, to dense SNP data from three closely related subspecies of house mice: Mus musculus musculus, M. m. castaneus, and M. m. domesticus. We documented substantial variation in phylogenetic history across the genome. Although each of the three possible topologies was strongly supported by a large number of loci, there was statistical evidence for a primary phylogenetic history in which M. m. musculus and M. m. castaneus are sister subspecies. These results underscore the importance of measuring phylogenetic discordance in other recently diverged groups using methods such as Bayesian concordance analysis, which are designed for this purpose.

Single locus typing of MHC class I and class II B loci in a population of red jungle fowl

In species with duplicated major histocompatibility complex (MHC) genes, estimates of genetic variation often rely on multilocus measures of diversity. It is possible that such measures might not always detect more detailed patterns of selection at individual loci. Here, we describe a method that allows us to investigate classical MHC diversity in red jungle fowl (Gallus gallus), the wild ancestor of the domestic chicken, using a single locus approach. This is possible due to the well-characterised gene organisation of the 'minimal essential' MHC (BF/BL region) of the domestic chicken, which comprises two differentially expressed duplicated class I (BF) and two class II B (BLB) genes. Using a combination of reference strand-mediated conformation analysis, cloning and sequencing, we identify nine BF and ten BLB alleles in a captive population of jungle fowl. We show that six BF and five BLB alleles are from the more highly expressed locus of each gene, BF2 and BLB2, respectively. An excess of non-synonymous substitutions across the jungle fowl BF/BL region suggests that diversifying selection has acted on this population. Importantly, single locus screening reveals that the strength of selection is greatest on the highly expressed BF2 locus. This is the first time that a population of red jungle fowl has been typed at the MHC region, laying the basis for further research into the underlying processes acting to maintain MHC diversity in this and other species.

Renin-Angiotensin System Gene Polymorphisms and Atrial Fibrillation: A Regression Approach for the Detection of Gene-Gene Interactions in a Large Hospitalized Population

Objectives: To test the association between renin-angiotensin system gene variants and atrial fibrillation (AF) using a regression approach. Methods: A total of 1,236 consecutive patients (227 with AF and 1,009 with normal sinus rhythm as controls) were recruited. Angiotensin-converting enzyme (ACE) gene I/D polymorphism; T174M, M235T, G–6A, A–20C, G–152A and G–217A polymorphisms of the angiotensinogen (AGT) gene, and A1166C polymorphism of the angiotensin II type I receptor (AT1R) gene were genotyped. We used a regression approach based on a generalized linear model to evaluate haplotype effects and to detect gene-gene interactions by incorporating interaction terms in the model. Results: In single-locus analyses, no locus was associated with AF. After adjustment for AF risk factors, we found significant differences in the global AGT gene haplotype profile (the global score statistic = 30.364, p = 0.001) and individual haplotype frequencies between AF patients and controls. Furthermore, significant 2-way gene-gene interactions between ACE I/D polymorphism and AGT gene haplotypes and between AT1R A1166C polymorphism and AGT gene haplotypes, and 3-way interaction between ACE I/D, AT1R A1166C and AGT gene haplotypes were detected. Conclusions: These results are compatible with the concept of multilocus and multigene effects in determining the risk of complex diseases such as AF, which would be missed with conventional single-locus approaches.

A novel strategy for detecting multiple loci in Genome-Wide Association Studies of complex diseases

Large-scale Genome-Wide Association Studies (GWAS) for complex diseases are increasingly common, due to recent advances in genotyping technology. Gene-gene interactions play an important role in the etiology of complex diseases and have to be addressed in GWAS. In this paper, an efficient strategy based on two-stage analysis is proposed. It combines a single-locus approach with a Goodness-Of-Fit (GOF) test in stage one, and selects a promising subset of SNPs to be modelled using a full interaction model in stage two. Extensive simulations using different disease models with different levels of epistasis demonstrate that it achieves higher power than existing approaches.

Two-Stage Two-Locus Models in Genome-Wide Association

Studies in model organisms suggest that epistasis may play an important role in the etiology of complex diseases and traits in humans. With the era of large-scale genome-wide association studies fast approaching, it is important to quantify whether it will be possible to detect interacting loci using realistic sample sizes in humans and to what extent undetected epistasis will adversely affect power to detect association when single-locus approaches are employed. We therefore investigated the power to detect association for an extensive range of two-locus quantitative trait models that incorporated varying degrees of epistasis. We compared the power to detect association using a single-locus model that ignored interaction effects, a full two-locus model that allowed for interactions, and, most important, two two-stage strategies whereby a subset of loci initially identified using single-locus tests were analyzed using the full two-locus model. Despite the penalty introduced by multiple testing, fitting the full two-locus model performed better than single-locus tests for many of the situations considered, particularly when compared with attempts to detect both individual loci. Using a two-stage strategy reduced the computational burden associated with performing an exhaustive two-locus search across the genome but was not as powerful as the exhaustive search when loci interacted. Two-stage approaches also increased the risk of missing interacting loci that contributed little effect at the margins. Based on our extensive simulations, our results suggest that an exhaustive search involving all pairwise combinations of markers across the genome might provide a useful complement to single-locus scans in identifying interacting loci that contribute to moderate proportions of the phenotypic variance.

Introgression variability among Iberian brown trout Evolutionary Significant Units: the influence of local management and environmental features

Summary1. A comprehensive analysis was carried out on the effects of stocking on the genetic structure of Iberian brown trout evolutionary lineages. Introgression and genetic diversity were estimated from allozyme results of 307 populations based on own data (180) and available literature (127). Stocking records, angling regulations and environmental features related to hatchery trout performance were also analysed to determine the underlying mechanisms of the introgression effects.2. Fifty per cent of analysed populations showed introgression by genes of hatchery origin. The mean introgression estimated by the single locus approach was 0.134. An increment of both heterozygosity and polymorphism was observed when introgression increased in stream‐dwelling populations, which could finally produce a homogenisation of the genetic structure of populations and a decrease of the species’ genetic diversity.3. Introgression rate varied among Iberian evolutionary lineages (Evolutionary Significant Units), and was correlated with the stocking effort, except for the North Atlantic basins. The lack of adaptations for migratory behaviour in hatchery trout could explain the low impact of stocking in North Atlantic rivers where anadromous populations occur.4. Angling regulation did not seem to influence the survival of hatchery trout. Introgression tends to be higher in heavily stocked localities with fertile waters and stable discharge, which may favour the performance of hatchery trout.5. Trout management must be based on increasing population size by means of habitat improvement and sustainability of naturally reproducing wild stocks through appropriate angling regulations.

Haplotype association analysis of human disease traits using genotype data of unrelated individuals

Haplotype inference has become an important part of human genetic data analysis due to its functional and statistical advantages over the single-locus approach in linkage disequilibrium mapping. Different statistical methods have been proposed for detecting haplotype - disease associations using unphased multi-locus genotype data, ranging from the early approach by the simple gene-counting method to the recent work using the generalized linear model. However, these methods are either confined to case - control design or unable to yield unbiased point and interval estimates of haplotype effects. Based on the popular logistic regression model, we present a new approach for haplotype association analysis of human disease traits. Using haplotype-based parameterization, our model infers the effects of specific haplotypes (point estimation) and constructs confidence interval for the risks of haplotypes (interval estimation). Based on the estimated parameters, the model calculates haplotype frequency conditional on the trait value for both discrete and continuous traits. Moreover, our model provides an overall significance level for the association between the disease trait and a group or all of the haplotypes. Featured by the direct maximization in haplotype estimation, our method also facilitates a computer simulation approach for correcting the significance level of individual haplotype to adjust for multiple testing. We show, by applying the model to an empirical data set, that our method based on the well-known logistic regression model is a useful tool for haplotype association analysis of human disease traits.

Haplotype Effects on Human Survival: Logistic Regression Models Applied to Unphased Genotype Data

SummaryHaplotype based linkage disequilibrium (LD) mapping exhibits higher power than the single locus approach because it makes use of the LD information contained in the flanking markers. New statistical methods have been proposed to help to infer haplotype effects on human diseases using multi‐locus genotype data collected from unrelated individuals. In this paper, we introduce a statistical procedure for measuring haplotype effects on human survival using the popular logistic regression model with haplotype based parameterizations. By modeling haplotype frequency as a function of age, our model infers haplotype effects by estimating and testing the slope parameters under different genetic mechanisms (multiplicative, dominant, or recessive). In addition, by estimating the sex‐specific slope parameters, our model allows the detection of sex‐specific haplotype effects or haplotype‐sex interactions. As an example, we apply our model to an empirical dataset on a stress related gene, interleukin‐6, to look for haplotypes that affect individual survival and for haplotype‐sex interactions. We show that our logistic regression based haplotype model can be a helpful tool for researchers interested in the genetics of human aging and longevity.

Haplotype Effects on Human Survival: Logistic Regression Models Applied to Unphased Genotype Data

Haplotype based linkage disequilibrium (LD) mapping exhibits higher power than the single locus approach because it makes use of the LD information contained in the flanking markers. New statistical methods have been proposed to help to infer haplotype effects on human diseases using multi-locus genotype data collected from unrelated individuals. In this paper, we introduce a statistical procedure for measuring haplotype effects on human survival using the popular logistic regression model with haplotype based parameterizations. By modeling haplotype frequency as a function of age, our model infers haplotype effects by estimating and testing the slope parameters under different genetic mechanisms (multiplicative, dominant, or recessive). In addition, by estimating the sex-specific slope parameters, our model allows the detection of sex-specific haplotype effects or haplotype-sex interactions. As an example, we apply our model to an empirical dataset on a stress related gene, interleukin-6, to look for haplotypes that affect individual survival and for haplotype-sex interactions. We show that our logistic regression based haplotype model can be a helpful tool for researchers interested in the genetics of human aging and longevity.