A Pleiotropy-Informed Bayesian False Discovery Rate Adapted to a Shared Control Design Finds New Disease Associations From GWAS Summary Statistics

Genome-wide association studies (GWAS) have been successful in identifying single nucleotide polymorphisms (SNPs) associated with many traits and diseases. However, at existing sample sizes, these variants explain only part of the estimated heritability. Leverage of GWAS results from related phenotypes may improve detection without the need for larger datasets. The Bayesian conditional false discovery rate (cFDR) constitutes an upper bound on the expected false discovery rate (FDR) across a set of SNPs whose p values for two diseases are both less than two disease-specific thresholds. Calculation of the cFDR requires only summary statistics and have several advantages over traditional GWAS analysis. However, existing methods require distinct control samples between studies. Here, we extend the technique to allow for some or all controls to be shared, increasing applicability. Several different SNP sets can be defined with the same cFDR value, and we show that the expected FDR across the union of these sets may exceed expected FDR in any single set. We describe a procedure to establish an upper bound for the expected FDR among the union of such sets of SNPs. We apply our technique to pairwise analysis of p values from ten autoimmune diseases with variable sharing of controls, enabling discovery of 59 SNP-disease associations which do not reach GWAS significance after genomic control in individual datasets. Most of the SNPs we highlight have previously been confirmed using replication studies or larger GWAS, a useful validation of our technique; we report eight SNP-disease associations across five diseases not previously declared. Our technique extends and strengthens the previous algorithm, and establishes robust limits on the expected FDR. This approach can improve SNP detection in GWAS, and give insight into shared aetiology between phenotypically related conditions.     

Improved Detection of Common Variants Associated with Schizophrenia and Bipolar Disorder Using Pleiotropy-Informed Conditional False Discovery Rate

Several lines of evidence suggest that genome-wide association studies (GWAS) have the potential to explain more of the “missing heritability” of common complex phenotypes. However, reliable methods to identify a larger proportion of single nucleotide polymorphisms (SNPs) that impact disease risk are currently lacking. Here, we use a genetic pleiotropy-informed conditional false discovery rate (FDR) method on GWAS summary statistics data to identify new loci associated with schizophrenia (SCZ) and bipolar disorders (BD), two highly heritable disorders with significant missing heritability. Epidemiological and clinical evidence suggest similar disease characteristics and overlapping genes between SCZ and BD. Here, we computed conditional Q–Q curves of data from the Psychiatric Genome Consortium (SCZ; n = 9,379 cases and n = 7,736 controls; BD: n = 6,990 cases and n = 4,820 controls) to show enrichment of SNPs associated with SCZ as a function of association with BD and vice versa with a corresponding reduction in FDR. Applying the conditional FDR method, we identified 58 loci associated with SCZ and 35 loci associated with BD below the conditional FDR level of 0.05. Of these, 14 loci were associated with both SCZ and BD (conjunction FDR). Together, these findings show the feasibility of genetic pleiotropy-informed methods to improve gene discovery in SCZ and BD and indicate overlapping genetic mechanisms between these two disorders.     

Estimation of false discovery rates in multiple testing: application to gene microarray data

Testing for significance with gene expression data from DNA microarray experiments involves simultaneous comparisons of hundreds or thousands of genes. If R denotes the number of rejections (declared significant genes) and V denotes the number of false rejections, then V/R, if R > 0, is the proportion of false rejected hypotheses. This paper proposes a model for the distribution of the number of rejections and the conditional distribution of V given R, V / R. Under the independence assumption, the distribution of R is a convolution of two binomials and the distribution of V / R has a noncentral hypergeometric distribution. Under an equicorrelated model, the distributions are more complex and are also derived. Five false discovery rate probability error measures are considered: FDR = E(V/R), pFDR = E(V/R / R > 0) (positive FDR), cFDR = E(V/R / R = r) (conditional FDR), mFDR = E(V)/E(R) (marginal FDR), and eFDR = E(V)/r (empirical FDR). The pFDR, cFDR, and mFDR are shown to be equivalent under the Bayesian framework, in which the number of true null hypotheses is modeled as a random variable. We present a parametric and a bootstrap procedure to estimate the FDRs. Monte Carlo simulations were conducted to evaluate the performance of these two methods. The bootstrap procedure appears to perform reasonably well, even when the alternative hypotheses are correlated (rho = .25). An example from a toxicogenomic microarray experiment is presented for illustration.     

Penalized Multimarker vs. Single-Marker Regression Methods for Genome-Wide Association Studies of Quantitative Traits

The data from genome-wide association studies (GWAS) in humans are still predominantly analyzed using single-marker association methods. As an alternative to single-marker analysis (SMA), all or subsets of markers can be tested simultaneously. This approach requires a form of penalized regression (PR) as the number of SNPs is much larger than the sample size. Here we review PR methods in the context of GWAS, extend them to perform penalty parameter and SNP selection by false discovery rate (FDR) control, and assess their performance in comparison with SMA. PR methods were compared with SMA, using realistically simulated GWAS data with a continuous phenotype and real data. Based on these comparisons our analytic FDR criterion may currently be the best approach to SNP selection using PR for GWAS. We found that PR with FDR control provides substantially more power than SMA with genome-wide type-I error control but somewhat less power than SMA with Benjamini–Hochberg FDR control (SMA-BH). PR with FDR-based penalty parameter selection controlled the FDR somewhat conservatively while SMA-BH may not achieve FDR control in all situations. Differences among PR methods seem quite small when the focus is on SNP selection with FDR control. Incorporating linkage disequilibrium into the penalization by adapting penalties developed for covariates measured on graphs can improve power but also generate more false positives or wider regions for follow-up. We recommend the elastic net with a mixing weight for the Lasso penalty near 0.5 as the best method.     

Efficiency of genome-wide association studies in random cross populations

Genome-wide association studies (GWAS) with plant species have employed inbred lines panels. We evaluated the efficiency of GWAS in non-inbred and inbred populations and assessed factors affecting GWAS. Fifty samples of 800 individuals from populations with linkage disequilibrium were simulated. Individuals were genotyped for 10,000 single nucleotide polymorphisms (SNPs) and phenotyped for traits controlled by ten quantitative trait loci (QTLs) and 90 minor genes, assuming different degrees of dominance and broad sense heritabilities of 40 and 80%. The average SNP density was 0.1 centiMorgan (cM) and the QTL heritabilities ranged from 3.2 to 11.8%. The results for random cross populations evidenced that to increase the QTL detection power, the additive-dominance model must be fitted for traits controlled by dominance effects but must not be fitted for traits showing no dominance. The power of detection was maximized by increasing the sample size to 400 and the false discovery rate (FDR) to 5%. The average power of detection for the low, intermediate, and high heritability QTLs achieved 52.4, 87.0, and 100.0%, respectively. Assuming sample sizes of 400 and 800, the observed FDR was equal to or lower than the specified level of significance. The association mapping was highly precise, since at least 97% of the declared QTLs were detected by the SNP inside it (average bias of 0.4 cM). Besides controlling the FDR, relatedness (and identity by state) efficiently controls the number of significant associations outside the QTL interval (not all false positive associations). The analysis of the inbred random cross population provided essentially the same results as the non-inbred populations.     

Genome-Wide Scan Informed by Age-Related Disease Identifies Loci for Exceptional Human Longevity

We developed a new statistical framework to find genetic variants associated with extreme longevity. The method, informed GWAS (iGWAS), takes advantage of knowledge from large studies of age-related disease in order to narrow the search for SNPs associated with longevity. To gain support for our approach, we first show there is an overlap between loci involved in disease and loci associated with extreme longevity. These results indicate that several disease variants may be depleted in centenarians versus the general population. Next, we used iGWAS to harness information from 14 meta-analyses of disease and trait GWAS to identify longevity loci in two studies of long-lived humans. In a standard GWAS analysis, only one locus in these studies is significant (APOE/TOMM40) when controlling the false discovery rate (FDR) at 10%. With iGWAS, we identify eight genetic loci to associate significantly with exceptional human longevity at FDR < 10%. We followed up the eight lead SNPs in independent cohorts, and found replication evidence of four loci and suggestive evidence for one more with exceptional longevity. The loci that replicated (FDR < 5%) included APOE/TOMM40 (associated with Alzheimer’s disease), CDKN2B/ANRIL (implicated in the regulation of cellular senescence), ABO (tags the O blood group), and SH2B3/ATXN2 (a signaling gene that extends lifespan in Drosophila and a gene involved in neurological disease). Our results implicate new loci in longevity and reveal a genetic overlap between longevity and age-related diseases and traits, including coronary artery disease and Alzheimer’s disease. iGWAS provides a new analytical strategy for uncovering SNPs that influence extreme longevity, and can be applied more broadly to boost power in other studies of complex phenotypes.     

On the use of mathematically-derived traits in QTL mapping

Mathematically-derived traits from two or more component traits, either by addition, subtraction, multiplication, or division, have been frequently used in genetics and breeding. When used in quantitative trait locus (QTL) mapping, derived traits sometimes show discrepancy with QTL identified for the component traits. We used three QTL distributions and three genetic effects models, and an actual maize mapping population, to investigate the efficiency of using derived traits in QTL mapping, and to understand the genetic and biological basis of derived-only QTL, i.e., QTL identified for a derived trait but not for any component trait. Results indicated that the detection power of the four putative QTL was consistently greater than 90% for component traits in simulated populations, each consisting of 200 recombinant inbred lines. Lower detection power and higher false discovery rate (FDR) were observed when derived traits were used. In an actual maize population, simulations were designed based on the observed QTL distributions and effects. When derived traits were used, QTL detected for both component and derived traits had comparable power, but those detected for component traits but not for derived traits had low detection power. The FDR from subtraction and division in the maize population were higher than the FDR from addition and multiplication. The use of derived traits increased the gene number, caused higher-order gene interactions than observed in component traits, and possibly complicated the linkage relationship between QTL as well. The increased complexity of the genetic architecture with derived traits may be responsible for the reduced detection power and the increased FDR. Derived-only QTL identified in practical genetic populations can be explained either as minor QTL that are not significant in QTL mapping of component traits, or as false positives.     

Improved Detection of Common Variants Associated with Schizophrenia by Leveraging Pleiotropy with Cardiovascular-Disease Risk Factors

Several lines of evidence suggest that genome-wide association studies (GWASs) have the potential to explain more of the “missing heritability” of common complex phenotypes. However, reliable methods for identifying a larger proportion of SNPs are currently lacking. Here, we present a genetic-pleiotropy-informed method for improving gene discovery with the use of GWAS summary-statistics data. We applied this methodology to identify additional loci associated with schizophrenia (SCZ), a highly heritable disorder with significant missing heritability. Epidemiological and clinical studies suggest comorbidity between SCZ and cardiovascular-disease (CVD) risk factors, including systolic blood pressure, triglycerides, low- and high-density lipoprotein, body mass index, waist-to-hip ratio, and type 2 diabetes. Using stratified quantile-quantile plots, we show enrichment of SNPs associated with SCZ as a function of the association with several CVD risk factors and a corresponding reduction in false discovery rate (FDR). We validate this “pleiotropic enrichment” by demonstrating increased replication rate across independent SCZ substudies. Applying the stratified FDR method, we identified 25 loci associated with SCZ at a conditional FDR level of 0.01. Of these, ten loci are associated with both SCZ and CVD risk factors, mainly triglycerides and low- and high-density lipoproteins but also waist-to-hip ratio, systolic blood pressure, and body mass index. Together, these findings suggest the feasibility of using genetic-pleiotropy-informed methods for improving gene discovery in SCZ and identifying potential mechanistic relationships with various CVD risk factors.     

Lung eQTLs to Help Reveal the Molecular Underpinnings of Asthma

Genome-wide association studies (GWAS) have identified loci reproducibly associated with pulmonary diseases; however, the molecular mechanism underlying these associations are largely unknown. The objectives of this study were to discover genetic variants affecting gene expression in human lung tissue, to refine susceptibility loci for asthma identified in GWAS studies, and to use the genetics of gene expression and network analyses to find key molecular drivers of asthma. We performed a genome-wide search for expression quantitative trait loci (eQTL) in 1,111 human lung samples. The lung eQTL dataset was then used to inform asthma genetic studies reported in the literature. The top ranked lung eQTLs were integrated with the GWAS on asthma reported by the GABRIEL consortium to generate a Bayesian gene expression network for discovery of novel molecular pathways underpinning asthma. We detected 17,178 cis- and 593 trans- lung eQTLs, which can be used to explore the functional consequences of loci associated with lung diseases and traits. Some strong eQTLs are also asthma susceptibility loci. For example, rs3859192 on chr17q21 is robustly associated with the mRNA levels of GSDMA (P = 3.55×10⁻¹⁵¹). The genetic-gene expression network identified the SOCS3 pathway as one of the key drivers of asthma. The eQTLs and gene networks identified in this study are powerful tools for elucidating the causal mechanisms underlying pulmonary disease. This data resource offers much-needed support to pinpoint the causal genes and characterize the molecular function of gene variants associated with lung diseases.     

Improving power of genome-wide association studies with weighted false discovery rate control and prioritized subset analysis

The issue of large-scale testing has caught much attention with the advent of high-throughput technologies. In genomic studies, researchers are often confronted with a large number of tests. To make simultaneous inference for the many tests, the false discovery rate (FDR) control provides a practical balance between the number of true positives and the number of false positives. However, when few hypotheses are truly non-null, controlling the FDR may not provide additional advantages over controlling the family-wise error rate (e.g., the Bonferroni correction). To facilitate discoveries from a study, weighting tests according to prior information is a promising strategy. A 'weighted FDR control' (WEI) and a 'prioritized subset analysis' (PSA) have caught much attention. In this work, we compare the two weighting schemes with systematic simulation studies and demonstrate their use with a genome-wide association study (GWAS) on type 1 diabetes provided by the Wellcome Trust Case Control Consortium. The PSA and the WEI both can increase power when the prior is informative. With accurate and precise prioritization, the PSA can especially create substantial power improvements over the commonly-used whole-genome single-step FDR adjustment (i.e., the traditional un-weighted FDR control). When the prior is uninformative (true disease susceptibility regions are not prioritized), the power loss of the PSA and the WEI is almost negligible. However, a caution is that the overall FDR of the PSA can be slightly inflated if the prioritization is not accurate and precise. Our study highlights the merits of using information from mounting genetic studies, and provides insights to choose an appropriate weighting scheme to FDR control on GWAS.     

Systematic identification of interaction effects between genome- and environment-wide associations in type 2 diabetes mellitus

Diseases such as type 2 diabetes (T2D) result from environmental and genetic factors, and risk varies considerably in the population. T2D-related genetic loci discovered to date explain only a small portion of the T2D heritability. Some heritability may be due to gene–environment interactions. However, documenting these interactions has been difficult due to low availability of concurrent genetic and environmental measures, selection bias, and challenges in controlling for multiple hypothesis testing. Through genome-wide association studies (GWAS), investigators have identified over 90 single nucleotide polymorphisms (SNPs) associated to T2D. Using a method analogous to GWAS [environment-wide association study (EWAS)], we found five environmental factors associated with the disease. By focusing on risk factors that emerge from GWAS and EWAS, it is possible to overcome difficulties in uncovering gene–environment interactions. Using data from the National Health and Nutrition Examination Survey (NHANES), we screened 18 SNPs and 5 serum-based environmental factors for interaction in association to T2D. We controlled for multiple hypotheses using false discovery rate (FDR) and Bonferroni correction and found four interactions with FDR <20 %. The interaction between rs13266634 (SLC30A8) and trans-β-carotene withstood Bonferroni correction (corrected p = 0.006, FDR <1.5 %). The per-risk-allele effect sizes in subjects with low levels of trans-β-carotene were 40 % greater than the marginal effect size [odds ratio (OR) 1.8, 95 % CI 1.3–2.6]. We hypothesize that impaired function driven by rs13266634 increases T2D risk when combined with serum levels of nutrients. Unbiased consideration of environmental and genetic factors may help identify larger and more relevant effect sizes for disease associations.     

Comparison of strategies to detect epistasis from eQTL data

Genome-wide association studies have been instrumental in identifying genetic variants associated with complex traits such as human disease or gene expression phenotypes. It has been proposed that extending existing analysis methods by considering interactions between pairs of loci may uncover additional genetic effects. However, the large number of possible two-marker tests presents significant computational and statistical challenges. Although several strategies to detect epistasis effects have been proposed and tested for specific phenotypes, so far there has been no systematic attempt to compare their performance using real data. We made use of thousands of gene expression traits from linkage and eQTL studies, to compare the performance of different strategies. We found that using information from marginal associations between markers and phenotypes to detect epistatic effects yielded a lower false discovery rate (FDR) than a strategy solely using biological annotation in yeast, whereas results from human data were inconclusive. For future studies whose aim is to discover epistatic effects, we recommend incorporating information about marginal associations between SNPs and phenotypes instead of relying solely on biological annotation. Improved methods to discover epistatic effects will result in a more complete understanding of complex genetic effects.     

A Novel Test for Gene-Ancestry Interactions in Genome-Wide Association Data

Genome-wide association study (GWAS) data on a disease are increasingly available from multiple related populations. In this scenario, meta-analyses can improve power to detect homogeneous genetic associations, but if there exist ancestry-specific effects, via interactions on genetic background or with a causal effect that co-varies with genetic background, then these will typically be obscured. To address this issue, we have developed a robust statistical method for detecting susceptibility gene-ancestry interactions in multi-cohort GWAS based on closely-related populations. We use the leading principal components of the empirical genotype matrix to cluster individuals into “ancestry groups” and then look for evidence of heterogeneous genetic associations with disease or other trait across these clusters. Robustness is improved when there are multiple cohorts, as the signal from true gene-ancestry interactions can then be distinguished from gene-collection artefacts by comparing the observed interaction effect sizes in collection groups relative to ancestry groups. When applied to colorectal cancer, we identified a missense polymorphism in iron-absorption gene CYBRD1 that associated with disease in individuals of English, but not Scottish, ancestry. The association replicated in two additional, independently-collected data sets. Our method can be used to detect associations between genetic variants and disease that have been obscured by population genetic heterogeneity. It can be readily extended to the identification of genetic interactions on other covariates such as measured environmental exposures. We envisage our methodology being of particular interest to researchers with existing GWAS data, as ancestry groups can be easily defined and thus tested for interactions.     

Improving false discovery rate estimation

MOTIVATION: Recent attempts to account for multiple testing in the analysis of microarray data have focused on controlling the false discovery rate (FDR). However, rigorous control of the FDR at a preselected level is often impractical. Consequently, it has been suggested to use the q-value as an estimate of the proportion of false discoveries among a set of significant findings. However, such an interpretation of the q-value may be unwarranted considering that the q-value is based on an unstable estimator of the positive FDR (pFDR). Another method proposes estimating the FDR by modeling p-values as arising from a beta-uniform mixture (BUM) distribution. Unfortunately, the BUM approach is reliable only in settings where the assumed model accurately represents the actual distribution of p-values. METHODS: A method called the spacings LOESS histogram (SPLOSH) is proposed for estimating the conditional FDR (cFDR), the expected proportion of false positives conditioned on having k 'significant' findings. SPLOSH is designed to be more stable than the q-value and applicable in a wider variety of settings than BUM. RESULTS: In a simulation study and data analysis example, SPLOSH exhibits the desired characteristics relative to the q-value and BUM. AVAILABILITY: The Web site www.stjuderesearch.org/statistics/splosh.html has links to freely available S-plus code to implement the proposed procedure.     

An Efficient Stepwise Statistical Test to Identify Multiple Linked Human Genetic Variants Associated with Specific Phenotypic Traits

Recent advances in genotyping methodologies have allowed genome-wide association studies (GWAS) to accurately identify genetic variants that associate with common or pathological complex traits. Although most GWAS have focused on associations with single genetic variants, joint identification of multiple genetic variants, and how they interact, is essential for understanding the genetic architecture of complex phenotypic traits. Here, we propose an efficient stepwise method based on the Cochran-Mantel-Haenszel test (for stratified categorical data) to identify causal joint multiple genetic variants in GWAS. This method combines the CMH statistic with a stepwise procedure to detect multiple genetic variants associated with specific categorical traits, using a series of associated I × J contingency tables and a null hypothesis of no phenotype association. Through a new stratification scheme based on the sum of minor allele count criteria, we make the method more feasible for GWAS data having sample sizes of several thousands. We also examine the properties of the proposed stepwise method via simulation studies, and show that the stepwise CMH test performs better than other existing methods (e.g., logistic regression and detection of associations by Markov blanket) for identifying multiple genetic variants. Finally, we apply the proposed approach to two genomic sequencing datasets to detect linked genetic variants associated with bipolar disorder and obesity, respectively.     

Genome-wide CNV analysis revealed variants associated with growth traits in African indigenous goats

Using the CaprineSNP50 data generated by the AGIN consortium, we detected common CNVs in 126 samples from four African indigenous goat breeds. A total of 30 CNVs ranging from 30,237 bp to 4,910,757 bp were identified. These CNVs were then associated with six growth traits by a linear regression analysis. Three significant associations were identified between two CNVs and two body traits after false discovery rate (FDR) correction (P < .05). One of them (CNV27) was significantly associated with both chest width and width of pin bones. It overlaps the SNX29 gene, the Gene Ontology (GO) annotations of which indicate CNV27 could be a potential functional candidate for meat production, health and reproduction traits. To our knowledge, this study is the first CNV-based association test of growth traits using SNP chip data in African meat goats.     

A Bayesian Method to Incorporate Hundreds of Functional Characteristics with Association Evidence to Improve Variant Prioritization

The increasing quantity and quality of functional genomic information motivate the assessment and integration of these data with association data, including data originating from genome-wide association studies (GWAS). We used previously described GWAS signals (“hits”) to train a regularized logistic model in order to predict SNP causality on the basis of a large multivariate functional dataset. We show how this model can be used to derive Bayes factors for integrating functional and association data into a combined Bayesian analysis. Functional characteristics were obtained from the Encyclopedia of DNA Elements (ENCODE), from published expression quantitative trait loci (eQTL), and from other sources of genome-wide characteristics. We trained the model using all GWAS signals combined, and also using phenotype specific signals for autoimmune, brain-related, cancer, and cardiovascular disorders. The non-phenotype specific and the autoimmune GWAS signals gave the most reliable results. We found SNPs with higher probabilities of causality from functional characteristics showed an enrichment of more significant p-values compared to all GWAS SNPs in three large GWAS studies of complex traits. We investigated the ability of our Bayesian method to improve the identification of true causal signals in a psoriasis GWAS dataset and found that combining functional data with association data improves the ability to prioritise novel hits. We used the predictions from the penalized logistic regression model to calculate Bayes factors relating to functional characteristics and supply these online alongside resources to integrate these data with association data.     

Mining the Unknown: A Systems Approach to Metabolite Identification Combining Genetic and Metabolic Information

Recent genome-wide association studies (GWAS) with metabolomics data linked genetic variation in the human genome to differences in individual metabolite levels. A strong relevance of this metabolic individuality for biomedical and pharmaceutical research has been reported. However, a considerable amount of the molecules currently quantified by modern metabolomics techniques are chemically unidentified. The identification of these “unknown metabolites” is still a demanding and intricate task, limiting their usability as functional markers of metabolic processes. As a consequence, previous GWAS largely ignored unknown metabolites as metabolic traits for the analysis. Here we present a systems-level approach that combines genome-wide association analysis and Gaussian graphical modeling with metabolomics to predict the identity of the unknown metabolites. We apply our method to original data of 517 metabolic traits, of which 225 are unknowns, and genotyping information on 655,658 genetic variants, measured in 1,768 human blood samples. We report previously undescribed genotype–metabotype associations for six distinct gene loci (SLC22A2, COMT, CYP3A5, CYP2C18, GBA3, UGT3A1) and one locus not related to any known gene (rs12413935). Overlaying the inferred genetic associations, metabolic networks, and knowledge-based pathway information, we derive testable hypotheses on the biochemical identities of 106 unknown metabolites. As a proof of principle, we experimentally confirm nine concrete predictions. We demonstrate the benefit of our method for the functional interpretation of previous metabolomics biomarker studies on liver detoxification, hypertension, and insulin resistance. Our approach is generic in nature and can be directly transferred to metabolomics data from different experimental platforms.