Blindly using Wald's test can miss rare disease-causal variants in case-control association studies

Revisiting the Wald Test in Small Case-Control Studies With a Skewed Covariate

The Wald test is routinely used in case-control studies to test for association between a covariate and disease. However, when the evidence for association is high, the Wald test tends to inflate small P values as a result of the Hauck-Donner effect (HDE). Here, we investigate the HDE in the context of genetic burden, both with and without additional covariates. First, we examine the burden-based P values in the absence of association using whole-exome sequence data from 1000 Genomes Project reference samples (n = 54) and selected preterm infants with neonatal complications (n = 74). Our careful analysis of the burden-based P values shows that the HDE is present and that the cause of the HDE in this setting is likely a natural extension of the well-known cause of the HDE in 2 × 2 contingency tables. Second, in a reanalysis of real data, we find that the permutation test provides increased power over the Wald, Firth, and likelihood ratio tests, which agrees with our intuition since the permutation test is valid for any sample size and since it does not suffer from the HDE. Therefore, we propose a powerful and computationally efficient permutation-based approach for the analysis and reanalysis of small case-control association studies.

The association of clinical phenotypes to known AD/FTD genetic risk loci and their inter-relationship

To elucidate how variants in genetic risk loci previously implicated in Alzheimer’s Disease (AD) and/or frontotemporal dementia (FTD) contribute to expression of disease phenotypes, a phenome-wide association study was performed in two waves. In the first wave, we explored clinical traits associated with thirteen genetic variants previously reported to be linked to disease risk using both the 23andMe and UKB cohorts. We tested 30 additional AD variants in UKB cohort only in the second wave. APOE variants defining ε2/ε3/ε4 alleles and rs646776 were identified to be significantly associated with metabolic/cardiovascular and longevity traits. APOE variants were also significantly associated with neurological traits. ABI3 variant rs28394864 was significantly associated with cardiovascular (e.g. (hypertension, ischemic heart disease, coronary atherosclerosis, angina) and immune-related trait asthma. Both APOE variants and CLU variant were significantly associated with nearsightedness. HLA- DRB1 variant was associated with diseases with immune-related traits. Additionally, variants from 10+ AD genes (BZRAP1-AS1, ADAMTS4, ADAM10, APH1B, SCIMP, ABI3, SPPL2A, ZNF232, GRN, CD2AP, and CD33) were associated with hematological measurements such as white blood cell (leukocyte) count, monocyte count, neutrophill count, platelet count, and/or mean platelet (thrombocyte) volume (an autoimmune disease biomarker). Many of these genes are expressed specifically in microglia. The associations of ABI3 variant with cardiovascular and immune-related traits are one of the novel findings from this study. Taken together, it is evidenced that at least some AD and FTD variants are associated with multiple clinical phenotypes and not just dementia. These findings were discussed in the context of causal relationship versus pleiotropy via Mendelian randomization analysis.

The Evolutionary Forces Shaping Cis- and Trans-Regulation of Gene Expression within a Population of Outcrossing Plants

Understanding the persistence of genetic variation within populations has long been a goal of evolutionary biology. One promising route toward achieving this goal is using population genetic approaches to describe how selection acts on the loci associated with trait variation. Gene expression provides a model trait for addressing the challenge of the maintenance of variation because it can be measured genome-wide without information about how gene expression affects traits. Previous work has shown that loci affecting the expression of nearby genes (local or cis-eQTLs) are under negative selection, but we lack a clear understanding of the selective forces acting on variants that affect the expression of genes in trans. Here, we identify loci that affect gene expression in trans using genomic and transcriptomic data from one population of the obligately outcrossing plant, Capsella grandiflora. The allele frequencies of trans-eQTLs are consistent with stronger negative selection acting on trans-eQTLs than cis-eQTLs, and stronger negative selection acting on trans-eQTLs associated with the expression of multiple genes. However, despite this general pattern, we still observe the presence of a trans-eQTL at intermediate frequency that affects the expression of a large number of genes in the same coexpression module. Overall, our work highlights the different selective pressures shaping variation in cis- and trans-regulation.

Effect of non-normality and low count variants on cross-phenotype association tests in GWAS

Many complex human diseases, such as type 2 diabetes, are characterized by multiple underlying traits/phenotypes that have substantially shared genetic architecture. Multivariate analysis of correlated traits has the potential to increase the power of detecting underlying common genetic loci. Several cross-phenotype association methods have been proposed-some require individual-level data on traits and genotypes, while the others require only summary-level data. In this article, we explore whether non-normality of multivariate trait distribution affects the inference from some of the existing multi-trait methods and how that effect is dependent on the allele count of the genetic variant being tested. We find that most of these tests are susceptible to biases that lead to spurious association signals. Even after controlling for confounders that may contribute to non-normality and then applying inverse normal transformation on the residuals of each trait, these tests may have inflated type I errors for variants with low minor allele counts (MACs). A likelihood ratio test of association based on the ordinal regression of individual-level genotype conditional on the traits seems to be the least biased and can maintain type I error when the MAC is reasonably large (e.g., MAC > 30). Application of these methods to publicly available summary statistics of eight amino acid traits on European samples seem to exhibit systematic inflation (especially for variants with low MAC), which is consistent with our findings from simulation experiments.

An African-specific haplotype in MRGPRX4 is associated with menthol cigarette smoking

In the U.S., more than 80% of African-American smokers use mentholated cigarettes, compared to less than 30% of Caucasian smokers. The reasons for these differences are not well understood. To determine if genetic variation contributes to mentholated cigarette smoking, we performed an exome-wide association analysis in a multiethnic population-based sample from Dallas, TX (N = 561). Findings were replicated in an independent cohort of African Americans from Washington, DC (N = 741). We identified a haplotype of MRGPRX4 (composed of rs7102322[G], encoding N245S, and rs61733596[G], T43T), that was associated with a 5-to-8 fold increase in the odds of menthol cigarette smoking. The variants are present solely in persons of African ancestry. Functional studies indicated that the variant G protein-coupled receptor encoded by MRGPRX4 displays reduced agonism in both arrestin-based and G protein-based assays, and alteration of agonism by menthol. These data indicate that genetic variation in MRGPRX4 contributes to inter-individual and inter-ethnic differences in the preference for mentholated cigarettes, and that the existence of genetic factors predisposing vulnerable populations to mentholated cigarette smoking can inform tobacco control and public health policies.

Fast permutation tests and related methods, for association between rare variants and binary outcomes

In large-scale genetic studies, a primary aim is to test for an association between genetic variants and a disease outcome. The variants of interest are often rare and appear with low frequency among subjects. In this situation, statistical tests based on standard asymptotic results do not adequately control the type I error rate, especially if the case : control ratio is unbalanced. In this article, we propose the use of permutation and approximate unconditional tests for testing association with rare variants. We use novel analytical calculations to efficiently approximate the true type I error rate under common study designs, and in numerical studies show that the proposed classes of tests significantly improve upon standard testing methods. We also illustrate our methods in data from a recent case-control study for genetic causes of a severe side effect of a common drug treatment.

Comparison of single-marker and multi-marker tests in rare variant association studies of quantitative traits

In genetic association studies of rare variants, low statistical power and potential violations of established estimator properties are among the main challenges of association tests. Multi-marker tests (MMTs) have been proposed to target these challenges, but any comparison with single-marker tests (SMTs) has to consider that their aim is to identify causal genomic regions instead of variants. Valid power comparisons have been performed for the analysis of binary traits indicating that MMTs have higher power, but there is a lack of conclusive studies for quantitative traits. The aim of our study was therefore to fairly compare SMTs and MMTs in their empirical power to identify the same causal loci associated with a quantitative trait. The results of extensive simulation studies indicate that previous results for binary traits cannot be generalized. First, we show that for the analysis of quantitative traits, conventional estimation methods and test statistics of single-marker approaches have valid properties yielding association tests with valid type I error, even when investigating singletons or doubletons. Furthermore, SMTs lead to more powerful association tests for identifying causal genes than MMTs when the effect sizes of causal variants are large, and less powerful tests when causal variants have small effect sizes. For moderate effect sizes, whether SMTs or MMTs have higher power depends on the sample size and percentage of causal SNVs. For a more complete picture, we also compare the power in studies of quantitative and binary traits, and the power to identify causal genes with the power to identify causal rare variants. In a genetic association analysis of systolic blood pressure in the Genetic Analysis Workshop 19 data, SMTs yielded smaller p-values compared to MMTs for most of the investigated blood pressure genes, and were least influenced by the definition of gene regions.

Cis-eQTL analysis and functional validation of candidate susceptibility genes for high-grade serous ovarian cancer

Genome-wide association studies have reported 11 regions conferring risk of high-grade serous epithelial ovarian cancer (HGSOC). Expression quantitative trait locus (eQTL) analyses can identify candidate susceptibility genes at risk loci. Here we evaluate cis-eQTL associations at 47 regions associated with HGSOC risk (P≤10(-5)). For three cis-eQTL associations (P<1.4 × 10(-3), FDR<0.05) at 1p36 (CDC42), 1p34 (CDCA8) and 2q31 (HOXD9), we evaluate the functional role of each candidate by perturbing expression of each gene in HGSOC precursor cells. Overexpression of HOXD9 increases anchorage-independent growth, shortens population-doubling time and reduces contact inhibition. Chromosome conformation capture identifies an interaction between rs2857532 and the HOXD9 promoter, suggesting this SNP is a leading causal variant. Transcriptomic profiling after HOXD9 overexpression reveals enrichment of HGSOC risk variants within HOXD9 target genes (P=6 × 10(-10) for risk variants (P<10(-4)) within 10 kb of a HOXD9 target gene in ovarian cells), suggesting a broader role for this network in genetic susceptibility to HGSOC.

Sequence Kernel Association Analysis of Rare Variant Set Based on the Marginal Regression Model for Binary Traits

Recent sequencing efforts have focused on exploring the influence of rare variants on the complex diseases. Gene level based tests by aggregating information across rare variants within a gene have become attractive to enrich the rare variant association signal. Among them, the sequence kernel association test (SKAT) has proved to be a very powerful method for jointly testing multiple rare variants within a gene. In this article, we explore an alternative SKAT. We propose to use the univariate likelihood ratio statistics from the marginal model for individual variants as input into the kernel association test. We show how to compute its significance P-value efficiently based on the asymptotic chi-square mixture distribution. We demonstrate through extensive numerical studies that the proposed method has competitive performance. Its usefulness is further illustrated with application to associations between rare exonic variants and type 2 diabetes (T2D) in the Atherosclerosis Risk in Communities (ARIC) study. We identified an exome-wide significant rare variant set in the gene ZZZ3 worthy of further investigations.

Genome-wide significant risk associations for mucinous ovarian carcinoma

Genome-wide association studies have identified several risk associations for ovarian carcinomas but not for mucinous ovarian carcinomas (MOCs). Our analysis of 1,644 MOC cases and 21,693 controls with imputation identified 3 new risk associations: rs752590 at 2q13 (P = 3.3 × 10(-8)), rs711830 at 2q31.1 (P = 7.5 × 10(-12)) and rs688187 at 19q13.2 (P = 6.8 × 10(-13)). We identified significant expression quantitative trait locus (eQTL) associations for HOXD9 at 2q31.1 in ovarian (P = 4.95 × 10(-4), false discovery rate (FDR) = 0.003) and colorectal (P = 0.01, FDR = 0.09) tumors and for PAX8 at 2q13 in colorectal tumors (P = 0.03, FDR = 0.09). Chromosome conformation capture analysis identified interactions between the HOXD9 promoter and risk-associated SNPs at 2q31.1. Overexpressing HOXD9 in MOC cells augmented the neoplastic phenotype. These findings provide the first evidence for MOC susceptibility variants and insights into the underlying biology of the disease.

The effect of rare variants on inflation of the test statistics in case-control analyses

BACKGROUND

The detection of bias due to cryptic population structure is an important step in the evaluation of findings of genetic association studies. The standard method of measuring this bias in a genetic association study is to compare the observed median association test statistic to the expected median test statistic. This ratio is inflated in the presence of cryptic population structure. However, inflation may also be caused by the properties of the association test itself particularly in the analysis of rare variants. We compared the properties of the three most commonly used association tests: the likelihood ratio test, the Wald test and the score test when testing rare variants for association using simulated data.

RESULTS

We found evidence of inflation in the median test statistics of the likelihood ratio and score tests for tests of variants with less than 20 heterozygotes across the sample, regardless of the total sample size. The test statistics for the Wald test were under-inflated at the median for variants below the same minor allele frequency.

CONCLUSIONS

In a genetic association study, if a substantial proportion of the genetic variants tested have rare minor allele frequencies, the properties of the association test may mask the presence or absence of bias due to population structure. The use of either the likelihood ratio test or the score test is likely to lead to inflation in the median test statistic in the absence of population structure. In contrast, the use of the Wald test is likely to result in under-inflation of the median test statistic which may mask the presence of population structure.

Transethnic replication of association of CTG18.1 repeat expansion of TCF4 gene with Fuchs' corneal dystrophy in Chinese implies common causal variant

PURPOSE

To test the association between the CTG18.1 trinucleotide repeat expansion of TCF4 gene and Fuchs' endothelial corneal dystrophy (FECD) in a Chinese population.

METHODS

The trinucleotide repeat polymorphism CTG18.1 was genotyped using short tandem repeat and triplet repeat primed polymerase chain reaction assays in 57 Chinese subjects with FECD and 121 controls. Statistical association of the expanded CTG18.1 allele and 18 single nucleotide polymorphisms (SNPs) across TCF4 with FECD was evaluated. To investigate the linkage disequilibrium structure of the TCF4 region, haplotype analysis was performed on our study subjects and compared with genotyping data of 97 Han Chinese and 85 Caucasians in the 1000 Genomes Project.

RESULTS

The expanded CTG18.1 allele was associated with FECD (P = 4.7 × 10(-14)), with the odds ratio of each copy of the expanded allele estimated to be 66.5 (95% confidence interval: 12.6-350.1). Five TCF4 SNPs showed association with FECD at a nominal level (P < 5.0 × 10(-2)); however, conditional on the expanded CTG18.1 polymorphism, none of the SNPs showed association with FECD. The only haplotype associated with the disease was the one with the expansion at the CTG18.1 locus.

CONCLUSIONS

Transethnic replication of the association between the CTG18.1 repeat expansion in the TCF4 gene and FECD suggests it is a common, causal variant shared in Eurasian populations conferring significant risk for the development of FECD. Our data suggest that the expanded CTG18.1 allele is the main, if not sole, causal variant at this gene locus in the Chinese population.

Genome-wide association analysis of radiation resistance in Drosophila melanogaster

Background

Ionizing radiation is genotoxic to cells. Healthy tissue toxicity in patients and radiation resistance in tumors present common clinical challenges in delivering effective radiation therapies. Radiation response is a complex, polygenic trait with unknown genetic determinants. The Drosophila Genetic Reference Panel (DGRP) provides a model to investigate the genetics of natural variation for sensitivity to radiation.

Methods and Findings

Radiation response was quantified in 154 inbred DGRP lines, among which 92 radiosensitive lines and 62 radioresistant lines were classified as controls and cases, respectively. A case-control genome-wide association screen for radioresistance was performed. There are 32 single nucleotide polymorphisms (SNPs) associated with radio resistance at a nominal p<10⁻⁵; all had modest effect sizes and were common variants with the minor allele frequency >5%. All the genes implicated by those SNP hits were novel, many without a known role in radiation resistance and some with unknown function. Variants in known DNA damage and repair genes associated with radiation response were below the significance threshold of p<10⁻⁵ and were not present among the significant hits. No SNP met the genome-wide significance threshold (p = 1.49×10⁻⁷), indicating a necessity for a larger sample size.

Conclusions

Several genes not previously associated with variation in radiation resistance were identified. These genes, especially the ones with human homologs, form the basis for exploring new pathways involved in radiation resistance in novel functional studies. An improved DGRP model with a sample size of at least 265 lines and ideally up to 793 lines is recommended for future studies of complex traits.

Enhancing the power to detect low-frequency variants in genome-wide screens

In genetic association studies a conventional test statistic is proportional to the correlation coefficient between the trait and the variant, with the result that it lacks power to detect association for low-frequency variants. Considering the link between the conventional association test statistics and the linkage disequilibrium measure r(2), we propose a test statistic analogous to the standardized linkage disequilibrium D' to increase the power of detecting association for low-frequency variants. By both simulation and real data analysis we show that the proposed D' test is more powerful than the conventional methods for detecting association for low-frequency variants in a genome-wide setting. The optimal coding strategy for the D' test and its asymptotic properties are also investigated. In summary, we advocate using the D' test in a dominant model as a complementary approach to enhancing the power of detecting association for low-frequency variants with moderate to large effect sizes in case-control genome-wide association studies.

Association and familial segregation of CTG18.1 trinucleotide repeat expansion of TCF4 gene in Fuchs' endothelial corneal dystrophy

PURPOSE

We tested the association between two intronic polymorphisms (CTG18.1 and rs613872) in TCF4 and Fuchs' endothelial corneal dystrophy (FECD), and analyzed their segregation patterns in families.

METHODS

We recruited 120 unrelated Caucasian subjects with FECD and 100 controls. Available family members of probands were recruited. Genotyping of the single nucleotide polymorphism (SNP) rs613872 was performed using Sanger sequencing or real-time allelic discrimination assay. The trinucleotide repeat polymorphism, CTG18.1, was genotyped using a combination of short tandem repeat assay and triplet repeat primed PCR assay. The cytosine-thymine-guanine (CTG) repeat length of ≥40 was classified as an expanded CTG18.1 allele. Association of the two loci with FECD was evaluated. Segregation in 29 families was examined.

RESULTS

The two polymorphisms are in linkage disequilibrium (r(2) = 0.65 in cases and 0.31 in controls). Significant associations were found between FECD and rs613872 (P = 3.1 × 10(-17)), expanded CTG18.1 allele (P = 6.5 × 10(-25)), and their haplotypes (P = 5.9 × 10(-19)). The odds ratio (OR) of each copy of the rs613872 G allele for FECD was estimated to be 9.5 (95% confidence interval [CI], 5.1-17.5). The OR of each copy of the CTG18.1 expanded allele was estimated to be 32.3 (95% CI, 13.4-77.6). The expanded CTG 18.1 allele cosegregated with the trait in 52% (15/29) of families with complete penetrance and 10% (3/29) with incomplete penetrance.

CONCLUSIONS

We report, to our knowledge, the first independent replication of the expanded CTG 18.1 allele conferring significant risk for FECD (>30-fold increase). The expanded allele cosegregates with the trait with complete penetrance in a majority of families, but we also document cases of incomplete penetrance.

Lung fibrosis-associated surfactant protein A1 and C variants induce latent transforming growth factor β1 secretion in lung epithelial cells

Missense mutations of surfactant proteins are recognized as important causes of inherited lung fibrosis. Here, we study rare and common surfactant protein (SP)-A1 and SP-C variants, either discovered in our familial pulmonary fibrosis cohort or described by others. We show that expression of two SP-A1 (R219W and R242*) and three SP-C (I73T, M71V, and L188Q) variant proteins lead to the secretion of the profibrotic latent transforming growth factor (TGF)-β1 in lung epithelial cell lines. The secreted TGF-β1 is capable of autocrine and paracrine signaling and is dependent upon expression of the latent TGF-β1 binding proteins. The dependence upon unfolded protein response (UPR) mediators for TGF-β1 induction differs for each variant. TGF-β1 secretion induced by the expression of the common SP-A1 R219W variant is nearly completely blocked by silencing the UPR transducers IRE-1α and ATF6. In contrast, the secretion of TGF-β1 induced by two rare SP-C mutant proteins (I73T and M71V), is largely unaffected by UPR silencing or by the addition of the small molecular chaperone 4-phenylbutyric acid, implicating a UPR-independent mechanism for these variants. Blocking TGF-β1 secretion reverses cell death of RLE-6TN cells expressing these SP-A1 and SP-C variants suggesting that anti-TGF-β therapeutics may be beneficial to this molecularly defined subgroup of pulmonary fibrosis patients.

Recommended joint and meta-analysis strategies for case-control association testing of single low-count variants

In genome-wide association studies of binary traits, investigators typically use logistic regression to test common variants for disease association within studies, and combine association results across studies using meta-analysis. For common variants, logistic regression tests are well calibrated, and meta-analysis of study-specific association results is only slightly less powerful than joint analysis of the combined individual-level data. In recent sequencing and dense chip based association studies, investigators increasingly test low-frequency variants for disease association. In this paper, we seek to (1) identify the association test with maximal power among tests with well controlled type I error rate and (2) compare the relative power of joint and meta-analysis tests. We use analytic calculation and simulation to compare the empirical type I error rate and power of four logistic regression based tests: Wald, score, likelihood ratio, and Firth bias-corrected. We demonstrate for low-count variants (roughly minor allele count [MAC] < 400) that: (1) for joint analysis, the Firth test has the best combination of type I error and power; (2) for meta-analysis of balanced studies (equal numbers of cases and controls), the score test is best, but is less powerful than Firth test based joint analysis; and (3) for meta-analysis of sufficiently unbalanced studies, all four tests can be anti-conservative, particularly the score test. We also establish MAC as the key parameter determining test calibration for joint and meta-analysis.

GWAS meta-analysis and replication identifies three new susceptibility loci for ovarian cancer

Genome-wide association studies (GWAS) have identified four susceptibility loci for epithelial ovarian cancer (EOC), with another two suggestive loci reaching near genome-wide significance. We pooled data from a GWAS conducted in North America with another GWAS from the UK. We selected the top 24,551 SNPs for inclusion on the iCOGS custom genotyping array. We performed follow-up genotyping in 18,174 individuals with EOC (cases) and 26,134 controls from 43 studies from the Ovarian Cancer Association Consortium. We validated the two loci at 3q25 and 17q21 that were previously found to have associations close to genome-wide significance and identified three loci newly associated with risk: two loci associated with all EOC subtypes at 8q21 (rs11782652, P = 5.5 × 10(-9)) and 10p12 (rs1243180, P = 1.8 × 10(-8)) and another locus specific to the serous subtype at 17q12 (rs757210, P = 8.1 × 10(-10)). An integrated molecular analysis of genes and regulatory regions at these loci provided evidence for functional mechanisms underlying susceptibility and implicated CHMP4C in the pathogenesis of ovarian cancer.