Accelerating Genome- and Phenome-Wide Association Studies using GPUs - A case study using data from the Million Veteran Program.

The expansion of biobanks has significantly propelled genomic discoveries yet the sheer scale of data within these repositories poses formidable computational hurdles, particularly in handling extensive matrix operations required by prevailing statistical frameworks. In this work, we introduce computational optimizations to the SAIGE (Scalable and Accurate Implementation of Generalized Mixed Model) algorithm, notably employing a GPU-based distributed computing approach to tackle these challenges. We applied these optimizations to conduct a large-scale genome-wide association study (GWAS) across 2,068 phenotypes derived from electronic health records of 635,969 diverse participants from the Veterans Affairs (VA) Million Veteran Program (MVP). Our strategies enabled scaling up the analysis to over 6,000 nodes on the Department of Energy (DOE) Oak Ridge Leadership Computing Facility (OLCF) Summit High-Performance Computer (HPC), resulting in a 20-fold acceleration compared to the baseline model. We also provide a Docker container with our optimizations that was successfully used on multiple cloud infrastructures on UK Biobank and All of Us datasets where we showed significant time and cost benefits over the baseline SAIGE model.

GWAS-informed data integration and non-coding CRISPRi screen illuminate genetic etiology of bone mineral density.

Over 1,100 independent signals have been identified with genome-wide association studies (GWAS) for bone mineral density (BMD), a key risk factor for mortality-increasing fragility fractures; however, the effector gene(s) for most remain unknown. Informed by a variant-to-gene mapping strategy implicating 89 non-coding elements predicted to regulate osteoblast gene expression at BMD GWAS loci, we executed a single-cell CRISPRi screen in human fetal osteoblast 1.19 cells (hFOBs). The BMD relevance of hFOBs was supported by heritability enrichment from cross-cell type stratified LD-score regression involving 98 cell types grouped into 15 tissues. 24 genes showed perturbation in the screen, with four (ARID5B, CC2D1B, EIF4G2, and NCOA3) exhibiting consistent effects upon siRNA knockdown on three measures of osteoblast maturation and mineralization. Lastly, additional heritability enrichments, genetic correlations, and multi-trait fine-mapping revealed that many BMD GWAS signals are pleiotropic and likely mediate their effects via non-bone tissues that warrant attention in future screens.

Evaluation of Plasma Biomarkers for Causal Association With Peripheral Artery Disease.

BACKGROUND: Hundreds of biomarkers for peripheral artery disease (PAD) have been reported in the literature; however, the observational nature of these studies limits causal inference due to the potential of reverse causality and residual confounding. We sought to evaluate the potential causal impact of putative PAD biomarkers identified in human observational studies through genetic causal inference methods. METHODS: Putative circulating PAD biomarkers were identified from human observational studies through a comprehensive literature search based on terms related to PAD using PubMed, Cochrane, and Embase. Genetic instruments were generated from publicly available genome-wide association studies of circulating biomarkers. Two-sample Mendelian randomization was used to test the association of genetically determined biomarker levels with PAD using summary statistics from a genome-wide association study of 31 307 individuals with and 211 753 individuals without PAD in the Veterans Affairs Million Veteran Program and replicated in data from FinnGen comprised of 11 924 individuals with and 288 638 individuals without PAD. RESULTS: We identified 204 unique circulating biomarkers for PAD from the observational literature, of which 173 were genetically instrumented using genome-wide association study results. After accounting for multiple testing (false discovery rate, <0.05), 10 of 173 (5.8%) biomarkers had significant associations with PAD. These 10 biomarkers represented categories including plasma lipoprotein regulation, lipid homeostasis, and protein-lipid complex remodeling. Observational literature highlighted different pathways including inflammatory response, negative regulation of multicellular organismal processes, and regulation of response to external stimuli. CONCLUSIONS: Integrating human observational studies and genetic causal inference highlights several key pathways in PAD pathophysiology. This work demonstrates that a substantial portion of biomarkers identified in observational studies are not well supported by human genetic evidence and emphasizes the importance of triangulating evidence to understand PAD pathophysiology. Although the identified biomarkers offer insights into atherosclerotic development in the lower limb, their specificity to PAD compared with more widespread atherosclerosis requires further study.

Long-Term Body Mass Index Variability and Adverse Cardiovascular Outcomes.

Importance: Body mass index (BMI; calculated as weight in kilograms divided by height in meters squared) is a commonly used estimate of obesity, which is a complex trait affected by genetic and lifestyle factors. Marked weight gain and loss could be associated with adverse biological processes. Objective: To evaluate the association between BMI variability and incident cardiovascular disease (CVD) events in 2 distinct cohorts. Design, Setting, and Participants: This cohort study used data from the Million Veteran Program (MVP) between 2011 and 2018 and participants in the UK Biobank (UKB) enrolled between 2006 and 2010. Participants were followed up for a median of 3.8 (5th-95th percentile, 3.5) years. Participants with baseline CVD or cancer were excluded. Data were analyzed from September 2022 and September 2023. Exposure: BMI variability was calculated by the retrospective SD and coefficient of variation (CV) using multiple clinical BMI measurements up to the baseline. Main Outcomes and Measures: The main outcome was incident composite CVD events (incident nonfatal myocardial infarction, acute ischemic stroke, and cardiovascular death), assessed using Cox proportional hazards modeling after adjustment for CVD risk factors, including age, sex, mean BMI, systolic blood pressure, total cholesterol, high-density lipoprotein cholesterol, smoking status, diabetes status, and statin use. Secondary analysis assessed whether associations were dependent on the polygenic score of BMI. Results: Among 92 363 US veterans in the MVP cohort (81 675 [88%] male; mean [SD] age, 56.7 [14.1] years), there were 9695 Hispanic participants, 22 488 non-Hispanic Black participants, and 60 180 non-Hispanic White participants. A total of 4811 composite CVD events were observed from 2011 to 2018. The CV of BMI was associated with 16% higher risk for composite CVD across all groups (hazard ratio [HR], 1.16; 95% CI, 1.13-1.19). These associations were unchanged among subgroups and after adjustment for the polygenic score of BMI. The UKB cohort included 65 047 individuals (mean [SD] age, 57.30 (7.77) years; 38 065 [59%] female) and had 6934 composite CVD events. Each 1-SD increase in BMI variability in the UKB cohort was associated with 8% increased risk of cardiovascular death (HR, 1.08; 95% CI, 1.04-1.11). Conclusions and Relevance: This cohort study found that among US veterans, higher BMI variability was a significant risk marker associated with adverse cardiovascular events independent of mean BMI across major racial and ethnic groups. Results were consistent in the UKB for the cardiovascular death end point. Further studies should investigate the phenotype of high BMI variability.

Major Depressive Disorder Impacts Peripheral Artery Disease Risk Through Intermediary Risk Factors.

BACKGROUND: Major depressive disorder (MDD) has been identified as a causal risk factor for multiple forms of cardiovascular disease. Although observational evidence has linked MDD to peripheral artery disease (PAD), causal evidence of this relationship is lacking. METHODS AND RESULTS: Inverse variance weighted 2-sample Mendelian randomization was used to test the association the between genetic liability for MDD and genetic liability for PAD. Genetic liability for MDD was associated with increased genetic liability for PAD (odds ratio [OR], 1.17 [95% CI, 1.06-1.29]; P=2.6×10-3). Genetic liability for MDD was also associated with increased genetically determined lifetime smoking (ß=0.11 [95% CI, 0.078-0.14]; P=1.2×10-12), decreased alcohol intake (ß=-0.078 [95% CI, -0.15 to 0]; P=0.043), and increased body mass index (ß=0.10 [95% CI, 0.02-0.19]; P=1.8×10-2), which in turn were associated with genetic liability for PAD (smoking: OR, 2.81 [95% CI, 2.28-3.47], P=9.8×10-22; alcohol: OR, 0.77 [95% CI, 0.66-0.88]; P=1.8×10-4; body mass index: OR, 1.61 [95% CI, 1.52-1.7]; P=1.3×10-57). Controlling for lifetime smoking index, alcohol intake, and body mass index with multivariable Mendelian randomization completely attenuated the association between genetic liability for MDD with genetic liability for PAD. CONCLUSIONS: This work provides evidence for a possible causal association between MDD and PAD that is dependent on intermediate risk factors, adding to the growing body of evidence suggesting that effective management and treatment of cardiovascular diseases may require a composite of physical and mental health interventions.

Profiling the genome and proteome of metabolic dysfunction-associated steatotic liver disease identifies potential therapeutic targets.

BACKGROUND & AIMS: Metabolic dysfunction-associated steatotic liver disease (MASLD) affects over 25% of the population and currently has no effective treatments. Plasma proteins with causal evidence may represent promising drug targets. We aimed to identify plasma proteins in the causal pathway of MASLD and explore their interaction with obesity. METHODS: We analysed 2,941 plasma proteins in 43,978 European participants from UK Biobank. We performed genome-wide association study (GWAS) for all MASLD-associated proteins and created the largest MASLD GWAS (109,885 cases/1,014,923 controls). We performed Mendelian Randomization (MR) and integrated proteins and their encoding genes in MASLD ranges to identify candidate causal proteins. We then validated them through independent replication, exome sequencing, liver imaging, bulk and single-cell gene expression, liver biopsies, pathway, and phenome-wide data. We explored the role of obesity by MR and multivariable MR across proteins, body mass index, and MASLD. RESULTS: We found 929 proteins associated with MASLD, reported five novel genetic loci associated with MASLD, and identified 17 candidate MASLD protein targets. We identified four novel targets for MASLD (CD33, GRHPR, HMOX2, and SCG3), provided protein evidence supporting roles of AHCY, FCGR2B, ORM1, and RBKS in MASLD, and validated nine previously known targets. We found that CD33, FCGR2B, ORM1, RBKS, and SCG3 mediated the association of obesity and MASLD, and HMOX2, ORM1, and RBKS had effect on MASLD independent of obesity. CONCLUSIONS: This study identified new protein targets in the causal pathway of MASLD, providing new insights into the multi-omics architecture and pathophysiology of MASLD. These findings advise further therapeutic interventions for MASLD.

Projecting genetic associations through gene expression patterns highlights disease etiology and drug mechanisms.

Genes act in concert with each other in specific contexts to perform their functions. Determining how these genes influence complex traits requires a mechanistic understanding of expression regulation across different conditions. It has been shown that this insight is critical for developing new therapies. Transcriptome-wide association studies have helped uncover the role of individual genes in disease-relevant mechanisms. However, modern models of the architecture of complex traits predict that gene-gene interactions play a crucial role in disease origin and progression. Here we introduce PhenoPLIER, a computational approach that maps gene-trait associations and pharmacological perturbation data into a common latent representation for a joint analysis. This representation is based on modules of genes with similar expression patterns across the same conditions. We observe that diseases are significantly associated with gene modules expressed in relevant cell types, and our approach is accurate in predicting known drug-disease pairs and inferring mechanisms of action. Furthermore, using a CRISPR screen to analyze lipid regulation, we find that functionally important players lack associations but are prioritized in trait-associated modules by PhenoPLIER. By incorporating groups of co-expressed genes, PhenoPLIER can contextualize genetic associations and reveal potential targets missed by single-gene strategies.

Deciphering the genetic architecture of atrial fibrillation offers insights into disease prediction, pathophysiology and downstream sequelae.

Aims: The study aimed to discover novel genetic loci for atrial fibrillation (AF), explore the shared genetic etiologies between AF and other cardiovascular and cardiometabolic traits, and uncover AF pathogenesis using Mendelian randomization analysis. Methods and results: We conducted a genome-wide association study meta-analysis including 109,787 AF cases and 1,165,920 controls of European ancestry and identified 215 loci, among which 91 were novel. We performed Genomic Structural Equation Modeling analysis between AF and four cardiovascular comorbidities (coronary artery disease, ischemic stroke, heart failure, and vneous thromboembolism) and found 189 loci shared across these diseases as well as a universal genetic locus shared by atherosclerotic outcomes (i.e., rs1537373 near CDKN2B). Three genetic loci (rs10740129 near JMJD1C, rs2370982 near NRXN3, and rs9931494 near FTO) were associated with AF and cardiometabolic traits. A polygenic risk score derived from this genome-wide meta-analysis was associated with AF risk (odds ratio 2.36, 95% confidence interval 2.31-2.41 per standard deviation increase) in the UK biobank. This score, combined with age, sex, and basic clinical features, predicted AF risk (AUC 0.784, 95% CI 0.781-0.787) in Europeans. Phenome-wide association analysis of the polygenic risk score identified many AF-related comorbidities of the circulatory, endocrine, and respiratory systems. Phenome-wide and multi-omic Mendelian randomization analyses identified associations of blood lipids and pressure, diabetes, insomnia, obesity, short sleep, and smoking, 27 blood proteins, one gut microbe (genus.Catenibacterium), and 11 blood metabolites with risk to AF. Conclusions: This genome-wide association study and trans-omic Mendelian randomization analysis provides insights into disease risk prediction, pathophysiology and downstream sequelae.

Common- and rare-variant genetic architecture of heart failure across the allele frequency spectrum.

Heart failure (HF) is a complex trait, influenced by environmental and genetic factors, that affects over 30 million individuals worldwide. Historically, the genetics of HF have been studied in Mendelian forms of disease, where rare genetic variants have been linked to familial cardiomyopathies. More recently, genome-wide association studies (GWAS) have successfully identified common genetic variants associated with risk of HF. However, the relative importance of genetic variants across the allele-frequency spectrum remains incompletely characterized. Here, we report the results of common- and rare-variant association studies of all-cause heart failure, applying recently developed methods to quantify the heritability of HF attributable to different classes of genetic variation. We combine GWAS data across multiple populations including 207,346 individuals with HF and 2,151,210 without, identifying 176 risk loci at genome-wide significance (p < 5×10-8). Signals at newly identified common-variant loci include coding variants in Mendelian cardiomyopathy genes (MYBPC3, BAG3), as well as regulators of lipoprotein (LPL) and glucose metabolism (GIPR, GLP1R), and are enriched in cardiac, muscle, nerve, and vascular tissues, as well as myocyte and adipocyte cell types. Gene burden studies across three biobanks (PMBB, UKB, AOU) including 27,208 individuals with HF and 349,126 without uncover exome-wide significant (p < 3.15×10-6) associations for HF and rare predicted loss-of-function (pLoF) variants in TTN, MYBPC3, FLNC, and BAG3. Total burden heritability of rare coding variants (2.2%, 95% CI 0.99-3.5%) is highly concentrated in a small set of Mendelian cardiomyopathy genes, and is lower than heritability attributable to common variants (4.3%, 95% CI 3.9-4.7%) which is more diffusely spread throughout the genome. Finally, we demonstrate that common-variant background, in the form of a polygenic risk score (PRS), significantly modifies the risk of HF among carriers of pathogenic truncating variants in the Mendelian cardiomyopathy gene TTN. These findings suggest a significant polygenic component to HF exists that is not captured by current clinical genetic testing.

Regularized sequence-context mutational trees capture variation in mutation rates across the human genome.

Germline mutation is the mechanism by which genetic variation in a population is created. Inferences derived from mutation rate models are fundamental to many population genetics methods. Previous models have demonstrated that nucleotides flanking polymorphic sites-the local sequence context-explain variation in the probability that a site is polymorphic. However, limitations to these models exist as the size of the local sequence context window expands. These include a lack of robustness to data sparsity at typical sample sizes, lack of regularization to generate parsimonious models and lack of quantified uncertainty in estimated rates to facilitate comparison between models. To address these limitations, we developed Baymer, a regularized Bayesian hierarchical tree model that captures the heterogeneous effect of sequence contexts on polymorphism probabilities. Baymer implements an adaptive Metropolis-within-Gibbs Markov Chain Monte Carlo sampling scheme to estimate the posterior distributions of sequence-context based probabilities that a site is polymorphic. We show that Baymer accurately infers polymorphism probabilities and well-calibrated posterior distributions, robustly handles data sparsity, appropriately regularizes to return parsimonious models, and scales computationally at least up to 9-mer context windows. We demonstrate application of Baymer in three ways-first, identifying differences in polymorphism probabilities between continental populations in the 1000 Genomes Phase 3 dataset, second, in a sparse data setting to examine the use of polymorphism models as a proxy for de novo mutation probabilities as a function of variant age, sequence context window size, and demographic history, and third, comparing model concordance between different great ape species. We find a shared context-dependent mutation rate architecture underlying our models, enabling a transfer-learning inspired strategy for modeling germline mutations. In summary, Baymer is an accurate polymorphism probability estimation algorithm that automatically adapts to data sparsity at different sequence context levels, thereby making efficient use of the available data.

Evaluation of the Performance of the RECODe Equation with the Addition of Polygenic Risk Scores for Adverse Cardiovascular Outcomes in Individuals with Type II Diabetes.

Aims/Hypothesis: Individuals with T2D are at an increased risk of developing cardiovascular complications; early identification of individuals can lead to an alteration of the natural history of the disease. Current approaches to risk prediction tailored to individuals with T2D are exemplified by the RECODe algorithms which predict CVD outcomes among individuals with T2D. Recent efforts to improve CVD risk prediction among the general population have included the incorporation of polygenic risk scores (PRS). This paper aims to investigate the utility of the addition of a coronary artery disease (CAD), stroke and heart failure risk score to the current RECODe model for disease stratification. Methods: We derived PRS using summary statistics for ischemic stroke (IS) from the coronary artery disease (CAD) and heart failure (HF) and tested prediction accuracy in the Penn Medicine Biobank (PMBB). A Cox proportional hazards model was used for time-to-event analyses within our cohort, and we compared model discrimination for the RECODe model with and without a PRS using AUC. Results: The RECODe model alone demonstrated an AUC [95% CI] of 0.67 [0.62-0.72] for ASCVD; the addition of the three PRS to the model demonstrated an AUC [95% CI] of 0.66 [0.63-0.70]. A z-test to compare the AUCs of the two models did not demonstrate a detectable difference between the two models (p=0.97). Conclusions/Interpretation: In the present study, we demonstrate that although PRS associate with CVD outcomes independent of traditional risk factors among individuals with T2D, the addition of PRS to contemporary clinical risk models does not specifically improve the predictive performance as compared to the baseline model.

Mendelian randomization analyses clarify the effects of height on cardiovascular diseases.

An inverse correlation between stature and risk of coronary artery disease (CAD) has been observed in several epidemiologic studies, and recent Mendelian randomization (MR) experiments have suggested causal association. However, the extent to which the effect estimated by MR can be explained by established cardiovascular risk factors is unclear, with a recent report suggesting that lung function traits could fully explain the height-CAD effect. To clarify this relationship, we utilized a well-powered set of genetic instruments for human stature, comprising >1,800 genetic variants for height and CAD. In univariable analysis, we confirmed that a one standard deviation decrease in height (~6.5 cm) was associated with a 12.0% increase in the risk of CAD, consistent with previous reports. In multivariable analysis accounting for effects from up to 12 established risk factors, we observed a >3-fold attenuation in the causal effect of height on CAD susceptibility (3.7%, p = 0.02). However, multivariable analyses demonstrated independent effects of height on other cardiovascular traits beyond CAD, consistent with epidemiologic associations and univariable MR experiments. In contrast with published reports, we observed minimal effects of lung function traits on CAD risk in our analyses, indicating that these traits are unlikely to explain the residual association between height and CAD risk. In sum, these results suggest the impact of height on CAD risk beyond previously established cardiovascular risk factors is minimal and not explained by lung function measures.

Evaluating the Contribution of Cell Type-Specific Alternative Splicing to Variation in Lipid Levels.

BACKGROUND: Genome-wide association studies have identified hundreds of loci associated with lipid levels. However, the genetic mechanisms underlying most of these loci are not well-understood. Recent work indicates that changes in the abundance of alternatively spliced transcripts contribute to complex trait variation. Consequently, identifying genetic loci that associate with alternative splicing in disease-relevant cell types and determining the degree to which these loci are informative for lipid biology is of broad interest. METHODS: We analyze gene splicing in 83 sample-matched induced pluripotent stem cell (iPSC) and hepatocyte-like cell lines (n=166), as well as in an independent collection of primary liver tissues (n=96) to perform discovery of splicing quantitative trait loci (sQTLs). RESULTS: We observe that transcript splicing is highly cell type specific, and the genes that are differentially spliced between iPSCs and hepatocyte-like cells are enriched for metabolism pathway annotations. We identify 1384 hepatocyte-like cell sQTLs and 1455 iPSC sQTLs at a false discovery rate of <5% and find that sQTLs are often shared across cell types. To evaluate the contribution of sQTLs to variation in lipid levels, we conduct colocalization analysis using lipid genome-wide association data. We identify 19 lipid-associated loci that colocalize either with an hepatocyte-like cell expression quantitative trait locus or sQTL. Only 2 loci colocalize with both a sQTL and expression quantitative trait locus, indicating that sQTLs contribute information about genome-wide association studies loci that cannot be obtained by analysis of steady-state gene expression alone. CONCLUSIONS: These results provide an important foundation for future efforts that use iPSC and iPSC-derived cells to evaluate genetic mechanisms influencing both cardiovascular disease risk and complex traits in general.

Genetically proxied glucose-lowering drug target perturbation and risk of cancer: a Mendelian randomisation analysis.

AIMS/HYPOTHESIS: Epidemiological studies have generated conflicting findings on the relationship between glucose-lowering medication use and cancer risk. Naturally occurring variation in genes encoding glucose-lowering drug targets can be used to investigate the effect of their pharmacological perturbation on cancer risk. METHODS: We developed genetic instruments for three glucose-lowering drug targets (peroxisome proliferator activated receptor γ [PPARG]; sulfonylurea receptor 1 [ATP binding cassette subfamily C member 8 (ABCC8)]; glucagon-like peptide 1 receptor [GLP1R]) using summary genetic association data from a genome-wide association study of type 2 diabetes in 148,726 cases and 965,732 controls in the Million Veteran Program. Genetic instruments were constructed using cis-acting genome-wide significant (p<5×10-8) SNPs permitted to be in weak linkage disequilibrium (r2<0.20). Summary genetic association estimates for these SNPs were obtained from genome-wide association study (GWAS) consortia for the following cancers: breast (122,977 cases, 105,974 controls); colorectal (58,221 cases, 67,694 controls); prostate (79,148 cases, 61,106 controls); and overall (i.e. site-combined) cancer (27,483 cases, 372,016 controls). Inverse-variance weighted random-effects models adjusting for linkage disequilibrium were employed to estimate causal associations between genetically proxied drug target perturbation and cancer risk. Co-localisation analysis was employed to examine robustness of findings to violations of Mendelian randomisation (MR) assumptions. A Bonferroni correction was employed as a heuristic to define associations from MR analyses as 'strong' and 'weak' evidence. RESULTS: In MR analysis, genetically proxied PPARG perturbation was weakly associated with higher risk of prostate cancer (for PPARG perturbation equivalent to a 1 unit decrease in inverse rank normal transformed HbA1c: OR 1.75 [95% CI 1.07, 2.85], p=0.02). In histological subtype-stratified analyses, genetically proxied PPARG perturbation was weakly associated with lower risk of oestrogen receptor-positive breast cancer (OR 0.57 [95% CI 0.38, 0.85], p=6.45×10-3). In co-localisation analysis, however, there was little evidence of shared causal variants for type 2 diabetes liability and cancer endpoints in the PPARG locus, although these analyses were likely underpowered. There was little evidence to support associations between genetically proxied PPARG perturbation and colorectal or overall cancer risk or between genetically proxied ABCC8 or GLP1R perturbation with risk across cancer endpoints. CONCLUSIONS/INTERPRETATION: Our drug target MR analyses did not find consistent evidence to support an association of genetically proxied PPARG, ABCC8 or GLP1R perturbation with breast, colorectal, prostate or overall cancer risk. Further evaluation of these drug targets using alternative molecular epidemiological approaches may help to further corroborate the findings presented in this analysis. DATA AVAILABILITY: Summary genetic association data for select cancer endpoints were obtained from the public domain: breast cancer ( https://bcac.ccge.medschl.cam.ac.uk/bcacdata/ ); and overall prostate cancer ( http://practical.icr.ac.uk/blog/ ). Summary genetic association data for colorectal cancer can be accessed by contacting GECCO (kafdem at fredhutch.org). Summary genetic association data on advanced prostate cancer can be accessed by contacting PRACTICAL (practical at icr.ac.uk). Summary genetic association data on type 2 diabetes from Vujkovic et al (Nat Genet, 2020) can be accessed through dbGAP under accession number phs001672.v3.p1 (pha004945.1 refers to the European-specific summary statistics). UK Biobank data can be accessed by registering with UK Biobank and completing the registration form in the Access Management System (AMS) ( https://www.ukbiobank.ac.uk/enable-your-research/apply-for-access ).

Mendelian Randomization Analysis Reveals a Complex Genetic Interplay among Atopic Dermatitis, Asthma, and Gastroesophageal Reflux Disease.

Rationale: Gastroesophageal reflux disease (GERD) is commonly associated with atopic disorders, but cause-effect relationships remain unclear. Objectives: We applied Mendelian randomization analysis to explore whether GERD is causally related to atopic disorders of the lung (asthma) and/or skin (atopic dermatitis [AD]). Methods: We conducted two-sample bidirectional Mendelian randomization to infer the magnitude and direction of causality between asthma and GERD, using summary statistics from the largest genome-wide association studies conducted on asthma (Ncases = 56,167) and GERD (Ncases = 71,522). In addition, we generated instrumental variables for AD from the latest population-level genome-wide association study meta-analysis (Ncases = 22,474) and assessed their fidelity and confidence of predicting the likely causal pathway(s) leading to asthma and/or GERD. Measurements and Main Results: Applying three different methods, each method revealed similar magnitude of causal estimates that were directionally consistent across the sensitivity analyses. Using an inverse variance-weighted method, the largest effect size was detected for asthma predisposition to AD (odds ratio [OR], 1.46; 95% confidence interval [CI], 1.34-1.59), followed by AD to asthma (OR, 1.34; 95% CI, 1.24-1.45). A significant association was detected for genetically determined asthma on risk of GERD (OR, 1.06; 95% CI, 1.03-1.09) but not genetically determined AD on GERD. In contrast, GERD equally increased risks of asthma (OR, 1.21; 95% CI, 1.09-1.35) and AD (OR, 1.21; 95% CI, 1.07-1.37). Conclusions: This study uncovers previously unrecognized causal pathways that have clinical implications in European-ancestry populations: 1) asthma is a causal risk for AD, and 2) the predisposition to AD, including asthma, can arise from specific pathogenic mechanisms manifested by GERD.

Genomics and phenomics of body mass index reveals a complex disease network.

Elevated body mass index (BMI) is heritable and associated with many health conditions that impact morbidity and mortality. The study of the genetic association of BMI across a broad range of common disease conditions offers the opportunity to extend current knowledge regarding the breadth and depth of adiposity-related diseases. We identify 906 (364 novel) and 41 (6 novel) genome-wide significant loci for BMI among participants of European (N~1.1 million) and African (N~100,000) ancestry, respectively. Using a BMI genetic risk score including 2446 variants, 316 diagnoses are associated in the Million Veteran Program, with 96.5% showing increased risk. A co-morbidity network analysis reveals seven disease communities containing multiple interconnected diseases associated with BMI as well as extensive connections across communities. Mendelian randomization analysis confirms numerous phenotypes across a breadth of organ systems, including conditions of the circulatory (heart failure, ischemic heart disease, atrial fibrillation), genitourinary (chronic renal failure), respiratory (respiratory failure, asthma), musculoskeletal and dermatologic systems that are deeply interconnected within and across the disease communities. This work shows that the complex genetic architecture of BMI associates with a broad range of major health conditions, supporting the need for comprehensive approaches to prevent and treat obesity.

Genome-wide association and multi-trait analyses characterize the common genetic architecture of heart failure.

Heart failure is a leading cause of cardiovascular morbidity and mortality. However, the contribution of common genetic variation to heart failure risk has not been fully elucidated, particularly in comparison to other common cardiometabolic traits. We report a multi-ancestry genome-wide association study meta-analysis of all-cause heart failure including up to 115,150 cases and 1,550,331 controls of diverse genetic ancestry, identifying 47 risk loci. We also perform multivariate genome-wide association studies that integrate heart failure with related cardiac magnetic resonance imaging endophenotypes, identifying 61 risk loci. Gene-prioritization analyses including colocalization and transcriptome-wide association studies identify known and previously unreported candidate cardiomyopathy genes and cellular processes, which we validate in gene-expression profiling of failing and healthy human hearts. Colocalization, gene expression profiling, and Mendelian randomization provide convergent evidence for the roles of BCKDHA and circulating branch-chain amino acids in heart failure and cardiac structure. Finally, proteome-wide Mendelian randomization identifies 9 circulating proteins associated with heart failure or quantitative imaging traits. These analyses highlight similarities and differences among heart failure and associated cardiovascular imaging endophenotypes, implicate common genetic variation in the pathogenesis of heart failure, and identify circulating proteins that may represent cardiomyopathy treatment targets.

Estimating the effect size of a hidden causal factor between SNPs and a continuous trait: a mediation model approach.

BACKGROUND: Observational studies and Mendelian randomization experiments have been used to identify many causal factors for complex traits in humans. Given a set of causal factors, it is important to understand the extent to which these causal factors explain some, all, or none of the genetic heritability, as measured by single-nucleotide polymorphisms (SNPs) that are associated with the trait. Using the mediation model framework with SNPs as the exposure, a trait of interest as the outcome, and the known causal factors as the mediators, we hypothesize that any unexplained association between the SNPs and the outcome trait is mediated by an additional unobserved, hidden causal factor. RESULTS: We propose a method to infer the effect size of this hidden mediating causal factor on the outcome trait by utilizing the estimated associations between a continuous outcome trait, the known causal factors, and the SNPs. The proposed method consists of three steps and, in the end, implements Markov chain Monte Carlo to obtain a posterior distribution for the effect size of the hidden mediator. We evaluate our proposed method via extensive simulations and show that when model assumptions hold, our method estimates the effect size of the hidden mediator well and controls type I error rate if the hidden mediator does not exist. In addition, we apply the method to the UK Biobank data and estimate parameters for a potential hidden mediator for waist-hip ratio beyond body mass index (BMI), and find that the hidden mediator has a large effect size relatively to the effect size of the known mediator BMI. CONCLUSIONS: We develop a framework to infer the effect of potential, hidden mediators influencing complex traits. This framework can be used to place boundaries on unexplained risk factors contributing to complex traits.

Polygenic Risk Scores in Alzheimer's Disease Genetics: Methodology, Applications, Inclusion, and Diversity.

The success of genome-wide association studies (GWAS) completed in the last 15 years has reinforced a key fact: polygenic architecture makes a substantial contribution to variation of susceptibility to complex disease, including Alzheimer's disease. One straight-forward way to capture this architecture and predict which individuals in a population are most at risk is to calculate a polygenic risk score (PRS). This score aggregates the risk conferred across multiple genetic variants, ultimately representing an individual's predicted genetic susceptibility for a disease. PRS have received increasing attention after having been successfully used in complex traits. This has brought with it renewed attention on new methods which improve the accuracy of risk prediction. While these applications are initially informative, their utility is far from equitable: the majority of PRS models use samples heavily if not entirely of individuals of European descent. This basic approach opens concerns of health equity if applied inaccurately to other population groups, or health disparity if we fail to use them at all. In this review we will examine the methods of calculating PRS and some of their previous uses in disease prediction. We also advocate for, with supporting scientific evidence, inclusion of data from diverse populations in these existing and future studies of population risk via PRS.

ColocQuiaL: a QTL-GWAS colocalization pipeline.

SUMMARY: Identifying genomic features responsible for genome-wide association study (GWAS) signals has proven to be a difficult challenge; many researchers have turned to colocalization analysis of GWAS signals with expression quantitative trait loci (eQTL) and splicing quantitative trait loci (sQTL) to connect GWAS signals to candidate causal genes. The ColocQuiaL pipeline provides a framework to perform these colocalization analyses at scale across the genome and returns summary files and locus visualization plots to allow for detailed review of the results. As an example, we used ColocQuiaL to perform colocalization between a recent type 2 diabetes GWAS and Genotype-Tissue Expression (GTEx) v8 single-tissue eQTL and sQTL data. AVAILABILITY AND IMPLEMENTATION: ColocQuiaL is primarily written in R and is freely available on GitHub: https://github.com/bvoightlab/ColocQuiaL.