A two-step approach to testing overall effect of gene-environment interaction for multiple phenotypes.

MOTIVATION: While gene-environment (GxE) interactions contribute importantly to many different phenotypes, detecting such interactions requires well-powered studies and has proven difficult. To address this, we combine two approaches to improve GxE power: simultaneously evaluating multiple phenotypes and using a two-step analysis approach. Previous work shows that the power to identify a main genetic effect can be improved by simultaneously analyzing multiple related phenotypes. For a univariate phenotype, two-step methods produce higher power for detecting a GxE interaction compared to single step analysis. Therefore, we propose a two-step approach to test for an overall GxE effect for multiple phenotypes. RESULTS: Using simulations we demonstrate that, when more than one phenotype has GxE effect (i.e. GxE pleiotropy), our approach offers substantial gain in power (18-43%) to detect an aggregate-level GxE effect for a multivariate phenotype compared to an analogous two-step method to identify GxE effect for a univariate phenotype. We applied the proposed approach to simultaneously analyze three lipids, LDL, HDL and Triglyceride with the frequency of alcohol consumption as environmental factor in the UK Biobank. The method identified two loci with an overall GxE effect on the vector of lipids, one of which was missed by the competing approaches. AVAILABILITY AND IMPLEMENTATION: We provide an R package MPGE implementing the proposed approach which is available from CRAN: https://cran.r-project.org/web/packages/MPGE/index.html. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Modest effect of statins on fasting glucose in a longitudinal electronic health record based cohort.

BACKGROUND: Prior studies of the glycemic effect of statins have been inconsistent. Also, most studies have only considered a short duration of statin use; the effect of long-term statin use on fasting glucose (FG) has not been well examined. The aim of this work is to investigate the effect of long-term statin exposure on FG levels. METHODS: Using electronic health record (EHR) data from a large and diverse longitudinal cohort, we defined long-term statin exposure in two ways: the cumulative years of statin use (cumulative supply) and the years' supply-weighted sum of doses (cumulative dose). Simvastatin, lovastatin, atorvastatin and pravastatin were included in the analysis. The relationship between statin exposure and FG was examined using linear regression with mixed effects modeling, comparing statin users before and after initiating statins and statin never-users. RESULTS: We examined 593,130 FG measurements from 87,151 individuals over a median follow up of 20 years. Of these, 42,678 were never-users and 44,473 were statin users with a total of 730,031 statin prescriptions. FG was positively associated with cumulative supply of statin but not comulative dose when both measures were in the same model. While statistically significant, the annual increase in FG attributable to statin exposure was modest at only 0.14 mg/dl, with only slight and non-significant differences among statin types. CONCLUSIONS: Elevation in FG level is associated with statin exposure, but the effect is modest. The results suggest that the risk of a clinically significant increase in FG attributable to long-term statin use is small for most individuals.

Leveraging eQTLs to identify individual-level tissue of interest for a complex trait.

Genetic predisposition for complex traits often acts through multiple tissues at different time points during development. As a simple example, the genetic predisposition for obesity could be manifested either through inherited variants that control metabolism through regulation of genes expressed in the brain, or that control fat storage through dysregulation of genes expressed in adipose tissue, or both. Here we describe a statistical approach that leverages tissue-specific expression quantitative trait loci (eQTLs) corresponding to tissue-specific genes to prioritize a relevant tissue underlying the genetic predisposition of a given individual for a complex trait. Unlike existing approaches that prioritize relevant tissues for the trait in the population, our approach probabilistically quantifies the tissue-wise genetic contribution to the trait for a given individual. We hypothesize that for a subgroup of individuals the genetic contribution to the trait can be mediated primarily through a specific tissue. Through simulations using the UK Biobank, we show that our approach can predict the relevant tissue accurately and can cluster individuals according to their tissue-specific genetic architecture. We analyze body mass index (BMI) and waist to hip ratio adjusted for BMI (WHRadjBMI) in the UK Biobank to identify subgroups of individuals whose genetic predisposition act primarily through brain versus adipose tissue, and adipose versus muscle tissue, respectively. Notably, we find that these individuals have specific phenotypic features beyond BMI and WHRadjBMI that distinguish them from random individuals in the data, suggesting biological effects of tissue-specific genetic contribution for these traits.

An efficient Bayesian meta-analysis approach for studying cross-phenotype genetic associations.

Simultaneous analysis of genetic associations with multiple phenotypes may reveal shared genetic susceptibility across traits (pleiotropy). For a locus exhibiting overall pleiotropy, it is important to identify which specific traits underlie this association. We propose a Bayesian meta-analysis approach (termed CPBayes) that uses summary-level data across multiple phenotypes to simultaneously measure the evidence of aggregate-level pleiotropic association and estimate an optimal subset of traits associated with the risk locus. This method uses a unified Bayesian statistical framework based on a spike and slab prior. CPBayes performs a fully Bayesian analysis by employing the Markov Chain Monte Carlo (MCMC) technique Gibbs sampling. It takes into account heterogeneity in the size and direction of the genetic effects across traits. It can be applied to both cohort data and separate studies of multiple traits having overlapping or non-overlapping subjects. Simulations show that CPBayes can produce higher accuracy in the selection of associated traits underlying a pleiotropic signal than the subset-based meta-analysis ASSET. We used CPBayes to undertake a genome-wide pleiotropic association study of 22 traits in the large Kaiser GERA cohort and detected six independent pleiotropic loci associated with at least two phenotypes. This includes a locus at chromosomal region 1q24.2 which exhibits an association simultaneously with the risk of five different diseases: Dermatophytosis, Hemorrhoids, Iron Deficiency, Osteoporosis and Peripheral Vascular Disease. We provide an R-package 'CPBayes' implementing the proposed method.

Determining Which Phenotypes Underlie a Pleiotropic Signal.

Discovering pleiotropic loci is important to understand the biological basis of seemingly distinct phenotypes. Most methods for assessing pleiotropy only test for the overall association between genetic variants and multiple phenotypes. To determine which specific traits are pleiotropic, we evaluate via simulation and application three different strategies. The first is model selection techniques based on the inverse regression of genotype on phenotypes. The second is a subset-based meta analysis ASSET [Bhattacharjee et al., ], which provides an optimal subset of nonnull traits. And the third is a modified Benjamini-Hochberg (B-H) procedure of controlling the expected false discovery rate [Benjamini and Hochberg, ] in the framework of phenome-wide association study. From our simulations we see that an inverse regression-based approach MultiPhen [O'Reilly et al., ] is more powerful than ASSET for detecting overall pleiotropic association, except for when all the phenotypes are associated and have genetic effects in the same direction. For determining which specific traits are pleiotropic, the modified B-H procedure performs consistently better than the other two methods. The inverse regression-based selection methods perform competitively with the modified B-H procedure only when the phenotypes are weakly correlated. The efficiency of ASSET is observed to lie below and in between the efficiency of the other two methods when the traits are weakly and strongly correlated, respectively. In our application to a large GWAS, we find that the modified B-H procedure also performs well, indicating that this may be an optimal approach for determining the traits underlying a pleiotropic signal.

SLCO1B1 functional variants and statin-induced myopathy in people with recent genealogical ancestors from Africa: a population-based real-world study.

Background: Clinical pharmacogenetic implementation guidelines for statin therapy are derived from evidence of primarily Eurocentric study populations. Functional SLCO1B1 variants that are rare in these study populations have not been investigated as a determinant of statin myotoxicity and are thus missing from guideline inclusion. Objective: Determine the relationship between candidate functional SLCO1B1 variants and statin-induced myopathy in people with recent genealogical ancestors from Africa. Design: Population-based pharmacogenetic study using real-world evidence from electronic health record-linked biobanks. Setting: Various health care settings. Participants: Self-identified white and Black statin users with genome-wide genotyping data available. Measurements: Primarily, the odds of statin-induced myopathy + rhabdomyolysis. Secondarily, total bilirubin levels. Thirdly, cell-based functional assay results. Results: Meta-analyses results demonstrated an increased risk of statin-induced myopathy + rhabdomyolysis with c.481+1G>T (odds ratio [OR] = 3.27, 95% confidence interval [CI] 1.43-7.46, P =.005) and c.1463G>C (OR = 2.45, 95% CI 1.04-5.78, P =.04) for Black participants. For White participants, c.521T>C was also significantly associated with increased risk of statin-induced myopathy + rhabdomyolysis (OR = 1.41, 95% CI 1.20-1.67, P =5.4x10 -5 ). This effect size for c.521T>C was similar in the Black participants, but did not meet the level of statistical significance (OR = 1.47, 95% CI 0.58-3.73, P =0.41). Supporting evidence using total bilirubin as an endogenous biomarker of SLCO1B1 function as well as from cell-based functional studies corroborated these findings. Limitations: Data limited to severe statin myotoxicity events. Conclusion: Our findings implicate Afrocentric SLCO1B1 variants on preemptive pharmacogenetic testing panels, which could have an instant impact on reducing the risk of statin-associated myotoxicity in historically excluded groups. Primary Funding Source: National Institutes of Health, Office of the Director - All of Us (OD-AoURP).

Effect of population stratification on false positive rates of population-based association analyses of quantitative traits.

It is now well established that population stratification can result in spurious association findings in genetic case-control studies. However, very few studies have addressed similar issues for mapping quantitative traits. Because quantitative phenotypes are often precursors of clinical endpoint traits and carry more information on within-genotype trait variability, it has been argued that studying these quantitative traits may be a more powerful strategy to map genes than the binary clinical endpoints. Thus, it is of interest to evaluate the adverse effects of population stratification on the analyses of quantitative traits. The popular statistical tests of association for quantitative traits using population level data are ANOVA, linear regression with an additive allelic effect and Kruskal-Wallis. We have theoretically studied the marginal effects of genetic heterogeneity and phenotypic heterogeneity as well as their joint effects on the false positive rates of these three tests. We have carried out extensive simulations under different genetic models and probability distributions of quantitative traits to assess the rate of false positives in the presence of population stratification. We find that the rate of false positives increases very quickly with simultaneous increase in differences in the standardized phenotypic means and marker allele frequencies in the subpopulations.

Polygenic Risk Score and Statin Relative Risk Reduction for Primary Prevention of Myocardial Infarction in a Real-World Population.

Genetic substudies of randomized controlled trials demonstrate that high coronary heart disease (CHD) polygenic risk score modifies statin CHD relative risk reduction; it is unknown if the association extends to statin users undergoing routine care. We sought to determine how statin effectiveness is modified by CHD polygenic risk score in a real-world cohort of participants without previous myocardial infarction. We determined CHD polygenic risk scores in participants of the Genetic Epidemiology Research on Adult Health and Aging (GERA) cohort. Covariate-adjusted Cox regression models were used to compare the risk of cardiovascular outcomes between statin users and matched nonusers. Statin effectiveness on incident myocardial infarction showed no gradient with increasing 10-year Pooled Cohort Equations atherosclerotic cardiovascular disease (ASCVD) risk across low, borderline, intermediate, and high ASCVD risk score groups. In contrast, statin effectiveness by polygenic risk was largest in the high polygenic risk score group (hazard ratio (HR) 0.41, 95% confidence interval (CI), 0.31-0.53; P = 1.5E-11), intermediate in the intermediate polygenic risk score group (HR 0.56, 95% CI, 0.47-0.66; P = 8.4E-12), and smallest in the low polygenic risk score group (HR 0.67, 95% CI, 0.47-0.97; P = 0.03; P for high vs. low = 0.01). ASCVD risk and statin low-density lipoprotein cholesterol (LDL-C) lowering did not differ across polygenic risk score groups. In patients undergoing routine care, CHD polygenic risk modified statin relative risk reduction of incident myocardial infarction independent of LDL-C lowering. Our findings extend prior work by identifying a subset (i.e., self-identified White individuals with low CHD polygenic risk scores) with attenuated clinical benefit from statins.

Power comparison between population-based case-control studies and family-based transmission-disequilibrium tests: An empirical study.

BACKGROUND: THERE ARE TWO MAJOR CLASSES OF GENETIC ASSOCIATION ANALYSES: population based and family based. Population-based case-control studies have been the method of choice due to the ease of data collection. However, population stratification is one of the major limitations of case-control studies, while family-based studies are protected against stratification. In this study, we carry out extensive simulations under different disease models (both Mendelian as well as complex) to evaluate the relative powers of the two approaches in detecting association. MATERIALS AND METHODS: The power comparisons are based on a case-control design comprising 200 cases and 200 controls versus a Transmission Disequilibrium Test (TDT) or Pedigree Disequilibrium Test (PDT) design with 200 informative trios. We perform the allele-level test for case-control studies, which is based on the difference of allele frequencies at a single nucleotide polymorphism (SNP) between unrelated cases and controls. The TDT and the PDT are based on preferential allelic transmissions at a SNP from heterozygous parents to the affected offspring. We considered five disease modes of inheritance: (i) recessive with complete penetrance (ii) dominant with complete penetrance and (iii), (iv) and (v) complex diseases with varying levels of penetrances and phenocopies. RESULTS: We find that while the TDT/PDT design with 200 informative trios is in general more powerful than a case-control design with 200 cases and 200 controls (except when the heterozygosity at the marker locus is high), it may be necessary to sample a very large number of trios to obtain the requisite number of informative families. CONCLUSION: The current study provides insights into power comparisons between population-based and family-based association studies.

The impact of adjusting for baseline in pharmacogenomic genome-wide association studies of quantitative change.

In pharmacogenomic studies of quantitative change, any association between genetic variants and the pretreatment (baseline) measurement can bias the estimate of effect between those variants and drug response. A putative solution is to adjust for baseline. We conducted a series of genome-wide association studies (GWASs) for low-density lipoprotein cholesterol (LDL-C) response to statin therapy in 34,874 participants of the Genetic Epidemiology Research on Adult Health and Aging (GERA) cohort as a case study to investigate the impact of baseline adjustment on results generated from pharmacogenomic studies of quantitative change. Across phenotypes of statin-induced LDL-C change, baseline adjustment identified variants from six loci meeting genome-wide significance (SORT/CELSR2/PSRC1, LPA, SLCO1B1, APOE, APOB, and SMARCA4/LDLR). In contrast, baseline-unadjusted analyses yielded variants from three loci meeting the criteria for genome-wide significance (LPA, APOE, and SLCO1B1). A genome-wide heterogeneity test of baseline versus statin on-treatment LDL-C levels was performed as the definitive test for the true effect of genetic variants on statin-induced LDL-C change. These findings were generally consistent with the models not adjusting for baseline signifying that genome-wide significant hits generated only from baseline-adjusted analyses (SORT/CELSR2/PSRC1, APOB, SMARCA4/LDLR) were likely biased. We then comprehensively reviewed published GWASs of drug-induced quantitative change and discovered that more than half (59%) inappropriately adjusted for baseline. Altogether, we demonstrate that (1) baseline adjustment introduces bias in pharmacogenomic studies of quantitative change and (2) this erroneous methodology is highly prevalent. We conclude that it is critical to avoid this common statistical approach in future pharmacogenomic studies of quantitative change.

A large electronic-health-record-based genome-wide study of serum lipids.

A genome-wide association study (GWAS) of 94,674 ancestrally diverse Kaiser Permanente members using 478,866 longitudinal electronic health record (EHR)-derived measurements for untreated serum lipid levels empowered multiple new findings: 121 new SNP associations (46 primary, 15 conditional, and 60 in meta-analysis with Global Lipids Genetic Consortium data); an increase of 33-42% in variance explained with multiple measurements; sex differences in genetic impact (greater impact in females for LDL, HDL, and total cholesterol and the opposite for triglycerides); differences in variance explained among non-Hispanic whites, Latinos, African Americans, and East Asians; genetic dominance and epistatic interaction, with strong evidence for both at the ABO and FUT2 genes for LDL; and tissue-specific enrichment of GWAS-associated SNPs among liver, adipose, and pancreas eQTLs. Using EHR pharmacy data, both LDL and triglyceride genetic risk scores (477 SNPs) were strongly predictive of age at initiation of lipid-lowering treatment. These findings highlight the value of longitudinal EHRs for identifying new genetic features of cholesterol and lipoprotein metabolism with implications for lipid treatment and risk of coronary heart disease.

Statistical equivalent of the classical TDT for quantitative traits and multivariate phenotypes.

Clinical end-point traits are usually governed by quantitative precursors. Hence, there is active research interest in developing statistical methods for association mapping of quantitative traits. Unlike population-based tests for association, family-based tests for transmission disequilibrium are protected against population stratification. In this study, we propose a logistic regression model to test the association for quantitative traits based on a trio design. We show that the method can be viewed as a direct extension of the classical transmission diequilibrium test for binary traits to quantitative traits. We evaluate the performance of our method usingextensive simulations and compare it with an existing method, family-based association test. We found that the two methods yield comparable powers if all families are considered. However, unlike FBAT, which yields an inflated rate of false positives when noninformative trios with all three individuals' heterozygous are removed, our method maintains the correct size without compromising too much on power. We show that our method can be easily modified to incorporate multivariate phenotypes. Here, we applied this method to analyse a quantitative endophenotype associated with alcoholism.

A novel transmission-based test of association for multivariate phenotypes: an application to systolic and diastolic blood pressure levels.

Unlike case-control studies, family-based tests for association are protected against population stratification. Complex genetic traits are often governed by quantitative precursors and it has been argued that it may be a more powerful strategy to analyze these quantitative precursors instead of the clinical end point trait. Although methods have been developed for family-based association tests for single quantitative traits, it is of interest to develop such methods for multivariate phenotypes. We propose a novel transmission-based approach based on a trio design using a simple logistic regression to test for association with a multivariate phenotype. We use our proposed method to analyze data on systolic and diastolic blood pressure levels provided in Genetic Analysis Workshop 18. However, we find that the bivariate analysis of the two phenotypes did not provide more promising results compared to univariate analyses, suggesting a possibility of a different set of major genetic variants modulating the two phenotypes.