MIMOSA: a resource consisting of improved methylome prediction models increases power to identify DNA methylation-phenotype associations

Although DNA methylation (DNAm) has been implicated in the pathogenesis of numerous complex diseases, from cancer to cardiovascular disease to autoimmune disease, the exact methylation sites that play key roles in these processes remain elusive. One strategy to identify putative causal CpG sites and enhance disease etiology understanding is to conduct methylome-wide association studies (MWASs), in which predicted DNA methylation that is associated with complex diseases can be identified. However, current MWAS models are primarily trained using the data from single studies, thereby limiting the methylation prediction accuracy and the power of subsequent association studies. Here, we introduce a new resource, MWAS Imputing Methylome Obliging Summary-level mQTLs and Associated LD matrices (MIMOSA), a set of models that substantially improve the prediction accuracy of DNA methylation and subsequent MWAS power through the use of a large summary-level mQTL dataset provided by the Genetics of DNA Methylation Consortium (GoDMC). Through the analyses of GWAS (genome-wide association study) summary statistics for 28 complex traits and diseases, we demonstrate that MIMOSA considerably increases the accuracy of DNA methylation prediction in whole blood, crafts fruitful prediction models for low heritability CpG sites, and determines markedly more CpG site-phenotype associations than preceding methods. Finally, we use MIMOSA to conduct a case study on high cholesterol, pinpointing 146 putatively causal CpG sites.

Impact of hyperuricemia on CKD risk beyond genetic predisposition in a population-based cohort study

The bidirectional effect of hyperuricemia on chronic kidney disease (CKD) underscores the importance of hyperuricemia as a risk factor for CKD. We evaluated the effect of hyperuricemia on the presence and development of CKD after considering genetic background by calculating polygenic risk scores (PRSs). We employed genome-wide association study summary statistics-excluding the United Kingdom Biobank (UKB) datasets among published CKD Gen Consortium papers-to calculate the PRSs for CKD in white background subjects. To validate PRS performance, we divided the UKB into two datasets to validate and test the data. We used logistic regression analysis to evaluate the association between hyperuricemia and CKD, and performed Kaplan-Meier survival analysis exclusively for subjects with available follow-up data. In total, 438,253 clinical data and 4,307,940 single nucleotide polymorphisms from 459,155 samples were included. We observed a significant positive association between PRS and CKD and the presence and development of CKD. Hyperuricemia significantly increased CKD risk (adjusted odds ratio 1.55, 95% confidence interval 1.48-1.61). The impact of hyperuricemia on CKD was maintained irrespective of PRS range. In addition, negative interaction between hyperuricemia and PRS for CKD was found. Survival analysis indicates that the presence of hyperuricemia significantly increased the risk of CKD development. The PRS for CKD thoroughly reflects the risk of CKD development. Hyperuricemia is a significant indicator of CKD risk, even after incorporating the genetic risk score for CKD. Irrespective of genetic risk, patients with a prospective risk of developing CKD require uric acid monitoring and management.

Cardiovascular Disease, Genetic Susceptibility, and Risk of Psychiatric Disorders and Suicide Attempt: A Community-Based Matched Cohort Study Based on the UK Biobank

BACKGROUND

The associations between cardiovascular disease (CVD) and multiple psychiatric disorders and suicide attempt, and whether different genetic susceptibilities affect such links, have not been investigated clearly.

METHODS AND RESULTS

Based on the UK Biobank, we conducted a matched cohort study involving 63 923 patients who were first hospitalized with a CVD diagnosis between 1997 and 2020, and their 127 845 matched unexposed individuals. Cox models were used to examine the subsequent risk of psychiatric disorders and suicide attempt (ie, anxiety, depression, stress-related disorder, substance misuse, psychotic disorder, and suicide behaviors) following CVD. We further performed stratified analyses by polygenic risk score for each studied psychiatric condition to detect the possible effects of genetic susceptibility on the observed associations. We found an increased risk of any psychiatric disorders and suicide attempt among CVD patients, compared with matched unexposed individuals, particularly within 1 year following the CVD (fully adjusted hazard ratio [HR] within 1 year, 1.83 [95% CI, 1.58-2.12]; HR after 1 year, 1.24 [95% CI, 1.16-1.32]). By subtype, the risk elevations existed for any psychiatric disorders and suicide attempt following most categories of CVDs. Analyses stratified by polygenic risk score revealed little impact of genetic predisposition to studied psychiatric conditions on these observed links.

CONCLUSIONS

Patients hospitalized for CVD were at increased subsequent risk of multiple types of psychiatric disorders and suicide attempt, especially in the first year after hospitalization, irrespective of their genetic susceptibilities to studied psychiatric conditions, and these findings underscore the necessity of developing timely psychological interventions for this vulnerable population.

Pharmacogenomics polygenic risk score: Ready or not for prime time?

Pharmacogenomic Polygenic Risk Scores (PRS) have emerged as a tool to address the polygenic nature of pharmacogenetic phenotypes, increasing the potential to predict drug response. Most pharmacogenomic PRS have been extrapolated from disease-associated variants identified by genome wide association studies (GWAS), although some have begun to utilize genetic variants from pharmacogenomic GWAS. As pharmacogenomic PRS hold the promise of enabling precision medicine, including stratified treatment approaches, it is important to assess the opportunities and challenges presented by the current data. This assessment will help determine how pharmacogenomic PRS can be advanced and transitioned into clinical use. In this review, we present a summary of recent evidence, evaluate the current status, and identify several challenges that have impeded the progress of pharmacogenomic PRS. These challenges include the reliance on extrapolations from disease genetics and limitations inherent to pharmacogenomics research such as low sample sizes, phenotyping inconsistencies, among others. We finally propose recommendations to overcome the challenges and facilitate the clinical implementation. These recommendations include standardizing methodologies for phenotyping, enhancing collaborative efforts, developing new statistical methods to capitalize on drug-specific genetic associations for PRS construction. Additional recommendations include enhancing the infrastructure that can integrate genomic data with clinical predictors, along with implementing user-friendly clinical decision tools, and patient education. Ethical and regulatory considerations should address issues related to patient privacy, informed consent and safe use of PRS. Despite these challenges, ongoing research and large-scale collaboration is likely to advance the field and realize the potential of pharmacogenomic PRS.

Polygenic Indices (a.k.a. Polygenic Scores) in Social Science: A Guide for Interpretation and Evaluation

Polygenic indices (PGI)-the new recommended label for polygenic scores (PGS) in social science-are genetic summary scales often used to represent an individual's liability for a disease, trait, or behavior based on the additive effects of measured genetic variants. Enthusiasm for linking genetic data with social outcomes and the inclusion of premade PGIs in social science datasets have facilitated increased uptake of PGIs in social science research-a trend that will likely continue. Yet, most social scientists lack the expertise to interpret and evaluate PGIs in social science research. Here, we provide a primer on PGIs for social scientists focusing on key concepts, unique statistical genetic considerations, and best practices in calculation, estimation, reporting, and interpretation. We summarize our recommended best practices as a checklist to aid social scientists in evaluating and interpreting studies with PGIs. We conclude by discussing the similarities between PGIs and standard social science scales and unique interpretative considerations.

Genetic Risk Stratification of Primary Open-Angle Glaucoma in Japanese Individuals

PURPOSE

To assess the impact of genetic risk estimation for primary open-angle glaucoma (POAG) in Japanese individuals.

DESIGN

Cross-sectional analysis.

PARTICIPANTS

Genetic risk scores (GRSs) were constructed based on a genome-wide association study (GWAS) of POAG in Japanese people. A total of 3625 Japanese individuals, including 1191 patients and 2434 controls (Japanese Tohoku), were used for the model selection. We also evaluated the discriminative accuracy of constructed GRSs in a dataset comprising 1034 patients and 1147 controls (the Japan Glaucoma Society Omics Group [JGS-OG] and the Genomic Research Committee of the Japanese Ophthalmological Society [GRC-JOS]) and 1900 participants from a population-based study (Hisayama Study).

METHODS

We evaluated 2 types of GRSs: polygenic risk scores using the pruning and thresholding procedure and a GRS using variants associated with POAG in the GWAS of the International Glaucoma Genetics Consortium (IGGC). We selected the model with the highest areas under the receiver operating characteristic curve (AUC). In the population-based study, we evaluated the correlations between GRS and ocular measurements.

MAIN OUTCOME MEASURE

Proportion of patients with POAG after stratification according to the GRS.

RESULTS

We found that a GRS using 98 variants, which showed genome-wide significance in the IGGC, showed the best discriminative accuracy (AUC, 0.65). In the Japanese Tohoku, the proportion of patients with POAG in the top 10% individuals was significantly higher than that in the lowest 10% (odds ratio [OR], 6.15; 95% confidence interval [CI], 4.35-8.71). In the JGS-OG and GRC-JOS, we confirmed similar impact of POAG GRS (AUC, 0.64; OR [top vs. bottom decile], 5.81; 95% CI, 3.79-9.01). In the population-based study, POAG prevalence was significantly higher in the top 20% individuals of the GRS compared with the bottom 20% (9.2% vs. 5.0%). However, the discriminative accuracy was low (AUC, 0.56). The POAG GRS was correlated positively with intraocular pressure (r = 0.08: P = 4.0 × 10-4) and vertical cup-to-disc ratio (r = 0.11; P = 4.0 × 10-6).

CONCLUSIONS

The GRS showed moderate discriminative accuracy for POAG in the Japanese population. However, risk stratification in the general population showed relatively weak discriminative performance.

FINANCIAL DISCLOSURE(S)

Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Utility of polygenic risk scores to aid in the diagnosis of rheumatic diseases

Rheumatic diseases (RDs) are characterized by autoimmunity and autoinflammation and are recognized as complex due to the interplay of multiple genetic, environmental, and lifestyle factors in their pathogenesis. The rapid advancement of genome-wide association studies (GWASs) has enabled the identification of numerous single nucleotide polymorphisms (SNPs) associated with RD susceptibility. Based on these SNPs, polygenic risk scores (PRSs) have emerged as promising tools for quantifying genetic risk in this disease group. This chapter reviews the current status of PRSs in assessing the risk of RDs and discusses their potential to improve the accuracy of the diagnosis of these complex diseases through their ability to discriminate among different RDs. PRSs demonstrate a high discriminatory capacity for various RDs and show potential clinical utility. As GWASs continue to evolve, PRSs are expected to enable more precise risk stratification by integrating genetic, environmental, and lifestyle factors, thereby refining individual risk predictions and advancing disease management strategies.

KnockoffHybrid: A knockoff framework for hybrid analysis of trio and population designs in genome-wide association studies

Both trio and population designs are popular study designs for identifying risk genetic variants in genome-wide association studies (GWASs). The trio design, as a family-based design, is robust to confounding due to population structure, whereas the population design is often more powerful due to larger sample sizes. Here, we propose KnockoffHybrid, a knockoff-based statistical method for hybrid analysis of both the trio and population designs. KnockoffHybrid provides a unified framework that brings together the advantages of both designs and produces powerful hybrid analysis while controlling the false discovery rate (FDR) in the presence of linkage disequilibrium and population structure. Furthermore, KnockoffHybrid has the flexibility to leverage different types of summary statistics for hybrid analyses, including expression quantitative trait loci (eQTL) and GWAS summary statistics. We demonstrate in simulations that KnockoffHybrid offers power gains over non-hybrid methods for the trio and population designs with the same number of cases while controlling the FDR with complex correlation among variants and population structure among subjects. In hybrid analyses of three trio cohorts for autism spectrum disorders (ASDs) from the Autism Speaks MSSNG, Autism Sequencing Consortium, and Autism Genome Project with GWAS summary statistics from the iPSYCH project and eQTL summary statistics from the MetaBrain project, KnockoffHybrid outperforms conventional methods by replicating several known risk genes for ASDs and identifying additional associations with variants in other genes, including the PRAME family genes involved in axon guidance and which may act as common targets for human speech/language evolution and related disorders.

Evaluation of polygenic scoring methods in five biobanks shows larger variation between biobanks than methods and finds benefits of ensemble learning

Methods of estimating polygenic scores (PGSs) from genome-wide association studies are increasingly utilized. However, independent method evaluation is lacking, and method comparisons are often limited. Here, we evaluate polygenic scores derived via seven methods in five biobank studies (totaling about 1.2 million participants) across 16 diseases and quantitative traits, building on a reference-standardized framework. We conducted meta-analyses to quantify the effects of method choice, hyperparameter tuning, method ensembling, and the target biobank on PGS performance. We found that no single method consistently outperformed all others. PGS effect sizes were more variable between biobanks than between methods within biobanks when methods were well tuned. Differences between methods were largest for the two investigated autoimmune diseases, seropositive rheumatoid arthritis and type 1 diabetes. For most methods, cross-validation was more reliable for tuning hyperparameters than automatic tuning (without the use of target data). For a given target phenotype, elastic net models combining PGS across methods (ensemble PGS) tuned in the UK Biobank provided consistent, high, and cross-biobank transferable performance, increasing PGS effect sizes (β coefficients) by a median of 5.0% relative to LDpred2 and MegaPRS (the two best-performing single methods when tuned with cross-validation). Our interactively browsable online-results and open-source workflow prspipe provide a rich resource and reference for the analysis of polygenic scoring methods across biobanks.

Effects of polygenic risk score and sodium and potassium intake on hypertension in Asians: A nationwide prospective cohort study

Genetic factors, lifestyle, and diet have been shown to play important roles in the development of hypertension. Increased salt intake is an important risk factor for hypertension. However, research on the involvement of genetic factors in the relationship between salt intake and hypertension in Asians is lacking. We aimed to investigate the risk of hypertension in relation to sodium and potassium intake and the effects of genetic factors on their interactions. We used Korean Genome and Epidemiology Study data and calculated the polygenic risk score (PRS) for the effect of systolic and diastolic blood pressure (SBP and DBP). We also conducted multivariable logistic modeling to evaluate associations among incident hypertension, PRSSBP, PRSDBP, and sodium and potassium intake. In total, 41,351 subjects were included in the test set. The top 10% PRSSBP group was the youngest of the three groups (bottom 10%, middle, top 10%), had the highest proportion of women, and had the highest body mass index, baseline BP, red meat intake, and alcohol consumption. The multivariable logistic regression model revealed the risk of hypertension was significantly associated with higher PRSSBP, higher sodium intake, and lower potassium intake. There was significant interaction between sodium intake and PRSSBP for incident hypertension especially in sodium intake ≥2.0 g/day and PRSSBP top 10% group (OR 1.27 (1.07-1.51), P = 0.007). Among patients at a high risk of incident hypertension due to sodium intake, lifestyle modifications and sodium restriction were especially important to prevent hypertension.

Polygenic hazard score models for the prediction of Alzheimer's free survival using the lasso for Cox's proportional hazards model

The prediction of the susceptibility of an individual to a certain disease is an important and timely research area. An established technique is to estimate the risk of an individual with the help of an integrated risk model, that is, a polygenic risk score with added epidemiological covariates. However, integrated risk models do not capture any time dependence, and may provide a point estimate of the relative risk with respect to a reference population. The aim of this work is twofold. First, we explore and advocate the idea of predicting the time-dependent hazard and survival (defined as disease-free time) of an individual for the onset of a disease. This provides a practitioner with a much more differentiated view of absolute survival as a function of time. Second, to compute the time-dependent risk of an individual, we use published methodology to fit a Cox's proportional hazard model to data from a genetic SNP study of time to Alzheimer's disease (AD) onset, using the lasso to incorporate further epidemiological variables such as sex, APOE (apolipoprotein E, a genetic risk factor for AD) status, 10 leading principal components, and selected genomic loci. We apply the lasso for Cox's proportional hazards to a data set of 6792 AD patients (composed of 4102 cases and 2690 controls) and 87 covariates. We demonstrate that fitting a lasso model for Cox's proportional hazards allows one to obtain more accurate survival curves than with state-of-the-art (likelihood-based) methods. Moreover, the methodology allows one to obtain personalized survival curves for a patient, thus giving a much more differentiated view of the expected progression of a disease than the view offered by integrated risk models. The runtime to compute personalized survival curves is under a minute for the entire data set of AD patients, thus enabling it to handle datasets with 60,000-100,000 subjects in less than 1 h.

Microglial function interacts with the environment to affect sex-specific depression risk

There is a two-fold higher incidence of depression in females compared to men with recent studies suggesting a role for microglia in conferring this sex-dependent depression risk. In this study we investigated the nature of this relation. Using GWAS enrichment, gene-set enrichment analysis and Mendelian randomization, we found minimal evidence for a direct relation between genes functionally related to microglia and sex-dependent genetic risk for depression. We then used expression quantitative trait loci and single nucleus RNA-sequencing resources to generate polygenic scores (PGS) representative of individual variation in microglial function in the adult (UK Biobank; N = 54753-72682) and fetal (ALSPAC; N = 1452) periods. The adult microglial PGS moderated the association between BMI (UK Biobank; beta = 0.001, 95 %CI 0.0009 to 0.003, P = 7.74E-6) and financial insecurity (UK Biobank; beta = 0.001, 95 %CI 0.005 to 0.015, P = 2E-4) with depressive symptoms in females. The fetal microglia PGS moderated the association between maternal prenatal depressive symptoms and offspring depressive symptoms at 24 years in females (ALSPAC; beta = 0.04, 95 %CI 0.004 to 0.07, P = 0.03). We found no evidence for an interaction between the microglial PGS and depression risk factors in males. Our results illustrate a role for microglial function in the conferral of sex-dependent depression risk following exposure to a depression risk factor.

Understanding the genetic complexity of puberty timing across the allele frequency spectrum

Pubertal timing varies considerably and is associated with later health outcomes. We performed multi-ancestry genetic analyses on ~800,000 women, identifying 1,080 signals for age at menarche. Collectively, these explained 11% of trait variance in an independent sample. Women at the top and bottom 1% of polygenic risk exhibited ~11 and ~14-fold higher risks of delayed and precocious puberty, respectively. We identified several genes harboring rare loss-of-function variants in ~200,000 women, including variants in ZNF483, which abolished the impact of polygenic risk. Variant-to-gene mapping approaches and mouse gonadotropin-releasing hormone neuron RNA sequencing implicated 665 genes, including an uncharacterized G-protein-coupled receptor, GPR83, which amplified the signaling of MC3R, a key nutritional sensor. Shared signals with menopause timing at genes involved in DNA damage response suggest that the ovarian reserve might signal centrally to trigger puberty. We also highlight body size-dependent and independent mechanisms that potentially link reproductive timing to later life disease.

Use of Diagnostic Codes for Primary Open-Angle Glaucoma Polygenic Risk Score Construction in Electronic Health Record-Linked Biobanks

PURPOSE

Polygenic risk scores (PRSs) likely predict risk and prognosis of glaucoma. We compared the PRS performance for primary open-angle glaucoma (POAG), defined using International Classification of Diseases (ICD) codes vs manual medical record review.

DESIGN

Retrospective cohort study.

METHODS

We identified POAG cases in the Mount Sinai BioMe and Mass General Brigham (MGB) biobanks using ICD codes. We confirmed POAG based on optical coherence tomograms and visual fields. In a separate 5% sample, the absence of POAG was confirmed with intraocular pressure and cup-disc ratio criteria. We used genotype data and either self-reported glaucoma diagnoses or ICD-10 codes for glaucoma diagnoses from the UK Biobank and the lassosum method to compute a genome-wide POAG PRS. We compared the area under the curve (AUC) for POAG prediction based on ICD codes vs medical records.

RESULTS

We reviewed 804 of 996 BioMe and 367 of 1006 MGB ICD-identified cases. In BioMe and MGB, respectively, positive predictive value was 53% and 55%; negative predictive value was 96% and 97%; sensitivity was 97% and 97%; and specificity was 44% and 53%. Adjusted PRS AUCs for POAG using ICD codes vs manual record review in BioMe were not statistically different (P ≥.21) by ancestry: 0.77 vs 0.75 for African, 0.80 vs 0.80 for Hispanic, and 0.81 vs 0.81 for European. Results were similar in MGB (P ≥.18): 0.72 vs 0.80 for African, 0.83 vs 0.86 for Hispanic, and 0.74 vs 0.73 for European.

CONCLUSIONS

A POAG PRS performed similarly using either manual review or ICD codes in 2 electronic health record-linked biobanks; manual assessment of glaucoma status might not be necessary for some PRS studies. However, caution should be exercised when using ICD codes for glaucoma diagnosis given their low specificity (44%-53%) for manually confirmed cases of glaucoma.

A polygenic score method boosted by non-additive models

Dominance heritability in complex traits has received increasing recognition. However, most polygenic score (PGS) approaches do not incorporate non-additive effects. Here, we present GenoBoost, a flexible PGS modeling framework capable of considering both additive and non-additive effects, specifically focusing on genetic dominance. Building on statistical boosting theory, we derive provably optimal GenoBoost scores and provide its efficient implementation for analyzing large-scale cohorts. We benchmark it against seven commonly used PGS methods and demonstrate its competitive predictive performance. GenoBoost is ranked the best for four traits and second-best for three traits among twelve tested disease outcomes in UK Biobank. We reveal that GenoBoost improves prediction for autoimmune diseases by incorporating non-additive effects localized in the MHC locus and, more broadly, works best in less polygenic traits. We further demonstrate that GenoBoost can infer the mode of genetic inheritance without requiring prior knowledge. For example, GenoBoost finds non-zero genetic dominance effects for 602 of 900 selected genetic variants, resulting in 2.5% improvements in predicting psoriasis cases. Lastly, we show that GenoBoost can prioritize genetic loci with genetic dominance not previously reported in the GWAS catalog. Our results highlight the increased accuracy and biological insights from incorporating non-additive effects in PGS models.

Stacked neural network for predicting polygenic risk score

In recent years, the utility of polygenic risk scores (PRS) in forecasting disease susceptibility from genome-wide association studies (GWAS) results has been widely recognised. Yet, these models face limitations due to overfitting and the potential overestimation of effect sizes in correlated variants. To surmount these obstacles, we devised the Stacked Neural Network Polygenic Risk Score (SNPRS). This novel approach synthesises outputs from multiple neural network models, each calibrated using genetic variants chosen based on diverse p-value thresholds. By doing so, SNPRS captures a broader array of genetic variants, enabling a more nuanced interpretation of the combined effects of these variants. We assessed the efficacy of SNPRS using the UK Biobank data, focusing on the genetic risks associated with breast and prostate cancers, as well as quantitative traits like height and BMI. We also extended our analysis to the Korea Genome and Epidemiology Study (KoGES) dataset. Impressively, our results indicate that SNPRS surpasses traditional PRS models and an isolated deep neural network in terms of accuracy, highlighting its promise in refining the efficacy and relevance of PRS in genetic studies.

Integrating single cell expression quantitative trait loci summary statistics to understand complex trait risk genes

Transcriptome-wide association study (TWAS) is a popular approach to dissect the functional consequence of disease associated non-coding variants. Most existing TWAS use bulk tissues and may not have the resolution to reveal cell-type specific target genes. Single-cell expression quantitative trait loci (sc-eQTL) datasets are emerging. The largest bulk- and sc-eQTL datasets are most conveniently available as summary statistics, but have not been broadly utilized in TWAS. Here, we present a new method EXPRESSO (EXpression PREdiction with Summary Statistics Only), to analyze sc-eQTL summary statistics, which also integrates 3D genomic data and epigenomic annotation to prioritize causal variants. EXPRESSO substantially improves existing methods. We apply EXPRESSO to analyze multi-ancestry GWAS datasets for 14 autoimmune diseases. EXPRESSO uniquely identifies 958 novel gene x trait associations, which is 26% more than the second-best method. Among them, 492 are unique to cell type level analysis and missed by TWAS using whole blood. We also develop a cell type aware drug repurposing pipeline, which leverages EXPRESSO results to identify drug compounds that can reverse disease gene expressions in relevant cell types. Our results point to multiple drugs with therapeutic potentials, including metformin for type 1 diabetes, and vitamin K for ulcerative colitis.

Association of pre-existing depression and anxiety with Omicron variant infection

Pre-existing psychiatric disorders were linked to an increased susceptibility to COVID-19 during the initial outbreak of the pandemic, while evidence during Omicron prevalence is lacking. Leveraging data from two prospective cohorts in China, we identified incident Omicron infections between January 2023 and April 2023. Participants with a self-reported history or self-rated symptoms of depression or anxiety before the Omicron pandemic were considered the exposed group, whereas the others were considered unexposed. We employed multivariate logistic regression models to examine the association of pre-existing depression or anxiety with the risk of any or severe Omicron infection indexed by medical interventions or severe symptoms. Further, we stratified the analyses by polygenic risk scores (PRSs) for COVID-19 and repeated the analyses using the UK Biobank data. We included 10,802 individuals from the Chinese cohorts (mean age = 51.1 years, 45.6% male), among whom 7841 (72.6%) were identified as cases of Omicron infection. No association was found between any pre-existing depression or anxiety and the overall risk of Omicron infection (odds ratio [OR] =1.04, 95% confidence interval [CI] 0.95-1.14). However, positive associations were noted for severe Omicron infection, either as infections requiring medical interventions (1.26, 1.02-1.54) or with severe symptoms (≥3: 1.73, 1.51-1.97). We obtained comparable estimates when stratified by COVID-19 PRS level. Additionally, using clustering method, we identified eight distinct symptom patterns and found associations between pre-existing depression or anxiety and the patterns characterized by multiple or complex severe symptoms including cough and taste and smell decline (ORs = 1.42-2.35). The results of the UK Biobank analyses corroborated findings of the Chinese cohorts. In conclusion, pre-existing depression and anxiety was not associated with the risk of Omicron infection overall but an elevated risk of severe Omicron infection, supporting the continued efforts on monitoring and possible early intervention in this high-risk population during Omicron prevalence.

Brain and Blood Transcriptome-Wide Association Studies Identify Five Novel Genes Associated with Alzheimer's Disease

INTRODUCTION

Transcriptome-wide Association Studies (TWAS) extend genome-wide association studies (GWAS) by integrating genetically-regulated gene expression models. We performed the most powerful AD-TWAS to date, using summary statistics from cis -eQTL meta-analyses and the largest clinically-adjudicated Alzheimer's Disease (AD) GWAS.

METHODS

We implemented the OTTERS TWAS pipeline, leveraging cis -eQTL data from cortical brain tissue (MetaBrain; N=2,683) and blood (eQTLGen; N=31,684) to predict gene expression, then applied these models to AD-GWAS data (Cases=21,982; Controls=44,944).

RESULTS

We identified and validated five novel gene associations in cortical brain tissue ( PRKAG1 , C3orf62 , LYSMD4 , ZNF439 , SLC11A2 ) and six genes proximal to known AD-related GWAS loci (Blood: MYBPC3 ; Brain: MTCH2 , CYB561 , MADD , PSMA5 , ANXA11 ). Further, using causal eQTL fine-mapping, we generated sparse models that retained the strength of the AD-TWAS association for MTCH2 , MADD , ZNF439 , CYB561 , and MYBPC3 .

DISCUSSION

Our comprehensive AD-TWAS discovered new gene associations and provided insights into the functional relevance of previously associated variants.

Polygenic Risk Score Predicts Earlier-Onset Adult Systemic Lupus Erythematosus and First-Year Renal Diseases in a Taiwanese Cohort

OBJECTIVES

This study aimed to develop a predictive model using polygenic risk score (PRS) to forecast renal outcomes for adult systemic lupus erythematosus (SLE) in a Taiwanese population.

METHODS

Patients with SLE (n=2782) and matched non-SLE controls (n=11 128) were genotyped using Genome-Wide TWB 2.0 single-nucleotide polymorphism (SNP) array. PRS models (C+T, LDpred2, Lassosum, PRSice-2, PRS-continuous shrinkage (CS)) were constructed for predicting SLE susceptibility. Logistic regression was assessed for C+T-based PRS association with renal involvement in patients with SLE.

RESULTS

In the training set, C+T-based SLE-PRS, only incorporating 27 SNPs, outperformed other models with area under the curve (AUC) values of 0.629, surpassing Lassosum (AUC=0.621), PRSice-2 (AUC=0.615), LDpred2 (AUC=0.609) and PRS-CS (AUC=0.602). Additionally, C+T-based SLE-PRS demonstrated consistent predictive capacity in the testing set (AUC=0.620). Individuals in the highest quartile exhibited earlier SLE onset (39.06 vs 44.22 years, p<0.01), higher Systemic Lupus Erythematosus Disease Activity Index scores (3.00 vs 2.37, p=0.04), elevated risks of renal involvement within the first year of SLE diagnosis, including WHO class III-IV lupus nephritis (OR 2.36, 95% CI 1.47 to 3.80, p<0.01), estimated glomerular filtration rate <60 mL/min/1.73m2 (OR 1.49, 95% CI 1.18 to 1.89, p<0.01) and urine protein-to-creatinine ratio >150 mg/day (OR 2.07, 95% CI 1.49 to 2.89, p<0.01), along with increased seropositivity risks, compared with those in the lowest quartile. Furthermore, among patients with SLE with onset before 50 years, the highest PRS quartile was significantly associated with more serious renal diseases within the first year of SLE diagnosis.

CONCLUSIONS

PRS of SLE is associated with earlier onset, renal involvement within the first year of SLE diagnosis and seropositivity in Taiwanese patients. Integrating PRS with clinical decision-making may enhance lupus nephritis screening and early treatment to improve renal outcomes in patients with SLE.

Enhancing prediction accuracy of coronary artery disease through machine learning-driven genomic variant selection

Machine learning (ML) methods are increasingly becoming crucial in genome-wide association studies for identifying key genetic variants or SNPs that statistical methods might overlook. Statistical methods predominantly identify SNPs with notable effect sizes by conducting association tests on individual genetic variants, one at a time, to determine their relationship with the target phenotype. These genetic variants are then used to create polygenic risk scores (PRSs), estimating an individual's genetic risk for complex diseases like cancer or cardiovascular disorders. Unlike traditional methods, ML algorithms can identify groups of low-risk genetic variants that improve prediction accuracy when combined in a mathematical model. However, the application of ML strategies requires addressing the feature selection challenge to prevent overfitting. Moreover, ensuring the ML model depends on a concise set of genomic variants enhances its clinical applicability, where testing is feasible for only a limited number of SNPs. In this study, we introduce a robust pipeline that applies ML algorithms in combination with feature selection (ML-FS algorithms), aimed at identifying the most significant genomic variants associated with the coronary artery disease (CAD) phenotype. The proposed computational approach was tested on individuals from the UK Biobank, differentiating between CAD and non-CAD individuals within this extensive cohort, and benchmarked against standard PRS-based methodologies like LDpred2 and Lassosum. Our strategy incorporates cross-validation to ensure a more robust evaluation of genomic variant-based prediction models. This method is commonly applied in machine learning strategies but has often been neglected in previous studies assessing the predictive performance of polygenic risk scores. Our results demonstrate that the ML-FS algorithm can identify panels with as few as 50 genetic markers that can achieve approximately 80% accuracy when used in combination with known risk factors. The modest increase in accuracy over PRS performances is noteworthy, especially considering that PRS models incorporate a substantially larger number of genetic variants. This extensive variant selection can pose practical challenges in clinical settings. Additionally, the proposed approach revealed novel CAD-genetic variant associations.

Association between dyslipidemia and the risk of incident chronic kidney disease affected by genetic susceptibility: Polygenic risk score analysis

BACKGROUND

The effect of dyslipidemia on kidney disease outcomes has been inconclusive, and it requires further clarification. Therefore, we aimed to investigate the effects of genetic factors on the association between dyslipidemia and the risk of chronic kidney disease (CKD) using polygenic risk score (PRS).

METHODS

We analyzed data from 373,523 participants from the UK Biobank aged 40-69 years with no history of CKD. Baseline data included plasma levels of total cholesterol, low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), and triglyceride, as well as genome-wide genotype data for PRS. Our primary outcome, incident CKD, was defined as a composite of estimated glomerular filtration rate < 60 ml/min/1.73 m2 and CKD diagnosis according to International Classification of Disease-10 codes. The effects of the association between lipid levels and PRS on incident CKD were assessed using the Cox proportional hazards model. To investigate the effect of this association, we introduced multiplicative interaction terms into a multivariate analysis model and performed subgroup analysis stratified by PRS tertiles.

RESULTS

In total, 4,424 participants developed CKD. In the multivariable analysis, PRS was significantly predictive of the risk of incident CKD as both a continuous variable and a categorized variable. In addition, lower total cholesterol, LDL-C, HDL-C, and higher triglyceride levels were significantly associated with the risk of incident CKD. There were interactions between triglycerides and intermediate and high PRS, and the interactions were inversely associated with the risk of incident CKD.

CONCLUSIONS

This study showed that PRS presented significant predictive power for incident CKD and individuals in the low-PRS group had a higher risk of triglyceride-related incident CKD.

An ensemble penalized regression method for multi-ancestry polygenic risk prediction

Great efforts are being made to develop advanced polygenic risk scores (PRS) to improve the prediction of complex traits and diseases. However, most existing PRS are primarily trained on European ancestry populations, limiting their transferability to non-European populations. In this article, we propose a novel method for generating multi-ancestry Polygenic Risk scOres based on enSemble of PEnalized Regression models (PROSPER). PROSPER integrates genome-wide association studies (GWAS) summary statistics from diverse populations to develop ancestry-specific PRS with improved predictive power for minority populations. The method uses a combination ofL 1 (lasso) andL 2 (ridge) penalty functions, a parsimonious specification of the penalty parameters across populations, and an ensemble step to combine PRS generated across different penalty parameters. We evaluate the performance of PROSPER and other existing methods on large-scale simulated and real datasets, including those from 23andMe Inc., the Global Lipids Genetics Consortium, and All of Us. Results show that PROSPER can substantially improve multi-ancestry polygenic prediction compared to alternative methods across a wide variety of genetic architectures. In real data analyses, for example, PROSPER increased out-of-sample prediction R2 for continuous traits by an average of 70% compared to a state-of-the-art Bayesian method (PRS-CSx) in the African ancestry population. Further, PROSPER is computationally highly scalable for the analysis of large SNP contents and many diverse populations.

Polygenic risk score-based phenome-wide association for glaucoma and its impact on disease susceptibility in two large biobanks

BACKGROUND

Glaucoma is a leading cause of worldwide irreversible blindness. Considerable uncertainty remains regarding the association between a variety of phenotypes and the genetic risk of glaucoma, as well as the impact they exert on the glaucoma development.

METHODS

We investigated the associations of genetic liability for primary open angle glaucoma (POAG) with a wide range of potential risk factors and to assess its impact on the risk of incident glaucoma. The phenome-wide association study (PheWAS) approach was applied to determine the association of POAG polygenic risk score (PRS) with a wide range of phenotypes in 377, 852 participants from the UK Biobank study and 43,623 participants from the Penn Medicine Biobank study, all of European ancestry. Participants were stratified into four risk tiers: low, intermediate, high, and very high-risk. Cox proportional hazard models assessed the relationship of POAG PRS and ocular factors with new glaucoma events.

RESULTS

In both discovery and replication set in the PheWAS, a higher genetic predisposition to POAG was specifically correlated with ocular disease phenotypes. The POAG PRS exhibited correlations with low corneal hysteresis, refractive error, and ocular hypertension, demonstrating a strong association with the onset of glaucoma. Individuals carrying a high genetic burden exhibited a 9.20-fold, 11.88-fold, and 28.85-fold increase in glaucoma incidence when associated with low corneal hysteresis, high myopia, and elevated intraocular pressure, respectively.

CONCLUSION

Genetic susceptibility to POAG primarily influences ocular conditions, with limited systemic associations. Notably, the baseline polygenic risk for POAG robustly associates with new glaucoma events, revealing a large combined effect of genetic and ocular risk factors on glaucoma incidents.

Evaluating and improving health equity and fairness of polygenic scores

Polygenic scores (PGSs) are quantitative metrics for predicting phenotypic values, such as human height or disease status. Some PGS methods require only summary statistics of a relevant genome-wide association study (GWAS) for their score. One such method is Lassosum, which inherits the model selection advantages of Lasso to select a meaningful subset of the GWAS single-nucleotide polymorphisms as predictors from their association statistics. However, even efficient scores like Lassosum, when derived from European-based GWASs, are poor predictors of phenotype for subjects of non-European ancestry; that is, they have limited portability to other ancestries. To increase the portability of Lassosum, when GWAS information and estimates of linkage disequilibrium are available for both ancestries, we propose Joint-Lassosum (JLS). In the simulation settings we explore, JLS provides more accurate PGSs compared to other methods, especially when measured in terms of fairness. In analyses of UK Biobank data, JLS was computationally more efficient but slightly less accurate than a Bayesian comparator, SDPRX. Like all PGS methods, JLS requires selection of predictors, which are determined by data-driven tuning parameters. We describe a new approach to selecting tuning parameters and note its relevance for model selection for any PGS. We also draw connections to the literature on algorithmic fairness and discuss how JLS can help mitigate fairness-related harms that might result from the use of PGSs in clinical settings. While no PGS method is likely to be universally portable, due to the diversity of human populations and unequal information content of GWASs for different ancestries, JLS is an effective approach for enhancing portability and reducing predictive bias.

An Ensemble Penalized Regression Method for Multi-ancestry Polygenic Risk Prediction

Great efforts are being made to develop advanced polygenic risk scores (PRS) to improve the prediction of complex traits and diseases. However, most existing PRS are primarily trained on European ancestry populations, limiting their transferability to non-European populations. In this article, we propose a novel method for generating multi-ancestry Polygenic Risk scOres based on enSemble of PEnalized Regression models (PROSPER). PROSPER integrates genome-wide association studies (GWAS) summary statistics from diverse populations to develop ancestry-specific PRS with improved predictive power for minority populations. The method uses a combination of ℒ 1 (lasso) and ℒ 2 (ridge) penalty functions, a parsimonious specification of the penalty parameters across populations, and an ensemble step to combine PRS generated across different penalty parameters. We evaluate the performance of PROSPER and other existing methods on large-scale simulated and real datasets, including those from 23andMe Inc., the Global Lipids Genetics Consortium, and All of Us. Results show that PROSPER can substantially improve multi-ancestry polygenic prediction compared to alternative methods across a wide variety of genetic architectures. In real data analyses, for example, PROSPER increased out-of-sample prediction R2 for continuous traits by an average of 70% compared to a state-of-the-art Bayesian method (PRS-CSx) in the African ancestry population. Further, PROSPER is computationally highly scalable for the analysis of large SNP contents and many diverse populations.

MUSSEL: Enhanced Bayesian polygenic risk prediction leveraging information across multiple ancestry groups

Polygenic risk scores (PRSs) are now showing promising predictive performance on a wide variety of complex traits and diseases, but there exists a substantial performance gap across populations. We propose MUSSEL, a method for ancestry-specific polygenic prediction that borrows information in summary statistics from genome-wide association studies (GWASs) across multiple ancestry groups via Bayesian hierarchical modeling and ensemble learning. In our simulation studies and data analyses across four distinct studies, totaling 5.7 million participants with a substantial ancestral diversity, MUSSEL shows promising performance compared to alternatives. For example, MUSSEL has an average gain in prediction R2 across 11 continuous traits of 40.2% and 49.3% compared to PRS-CSx and CT-SLEB, respectively, in the African ancestry population. The best-performing method, however, varies by GWAS sample size, target ancestry, trait architecture, and linkage disequilibrium reference samples; thus, ultimately a combination of methods may be needed to generate the most robust PRSs across diverse populations.

Protein-truncating variants in BSN are associated with severe adult-onset obesity, type 2 diabetes and fatty liver disease

Obesity is a major risk factor for many common diseases and has a substantial heritable component. To identify new genetic determinants, we performed exome-sequence analyses for adult body mass index (BMI) in up to 587,027 individuals. We identified rare loss-of-function variants in two genes (BSN and APBA1) with effects substantially larger than those of well-established obesity genes such as MC4R. In contrast to most other obesity-related genes, rare variants in BSN and APBA1 were not associated with normal variation in childhood adiposity. Furthermore, BSN protein-truncating variants (PTVs) magnified the influence of common genetic variants associated with BMI, with a common variant polygenic score exhibiting an effect twice as large in BSN PTV carriers than in noncarriers. Finally, we explored the plasma proteomic signatures of BSN PTV carriers as well as the functional consequences of BSN deletion in human induced pluripotent stem cell-derived hypothalamic neurons. Collectively, our findings implicate degenerative processes in synaptic function in the etiology of adult-onset obesity.

Proteome-wide association studies using summary proteomic data identified 23 risk genes of Alzheimer's disease

Characterizing the genetic mechanisms underlying Alzheimer's disease (AD) dementia is crucial for developing new therapeutics. Proteome-wide association study (PWAS) integrating proteomics data with genome-wide association study (GWAS) summary data was shown as a powerful tool for detecting risk genes. The identified PWAS risk genes can be interpretated as having genetic effects mediated through the genetically regulated protein abundances. Existing PWAS analyses of AD often rely on the availability of individual-level proteomics and genetics data of a reference cohort. Leveraging summary-level protein quantitative trait loci (pQTL) reference data of multiple relevant tissues is expected to improve PWAS findings for studying AD. Here, we applied our recently developed OTTERS tool to conduct PWAS of AD dementia, by leveraging summary-level pQTL data of brain, cerebrospinal fluid (CSF), and plasma tissues, and multiple statistical methods. For each target protein, imputation models of the protein abundance with genetic predictors were trained from summary-level pQTL data, estimating a set of pQTL weights for considered genetic predictors. PWAS p-values were obtained by integrating GWAS summary data of AD dementia with estimated pQTL weights. PWAS p-values from multiple statistical methods were combined by the aggregated Cauchy association test to yield one omnibus PWAS p-value for the target protein. We identified significant PWAS risk genes through omnibus PWAS p-values and analyzed their protein-protein interactions using STRING. Their potential causal effects were assessed by the probabilistic Mendelian randomization (PMR-Egger). As a result, we identified a total of 23 significant PWAS risk genes for AD dementia in brain, CSF, and plasma tissues, including 7 novel findings. We showed that 15 of these risk genes were interconnected within a protein-protein interaction network involving the well-known AD risk gene of APOE and 5 novel findings, and enriched in immune functions and lipids pathways including positive regulation of immune system process, positive regulation of macrophage proliferation, humoral immune response, and high-density lipoprotein particle clearance. Existing biological evidence was found to relate our novel findings with AD. We validated the mediated causal effects of 14 risk genes (60.8%). In conclusion, we identified both known and novel PWAS risk genes, providing novel insights into the genetic mechanisms in brain, CSF, and plasma tissues, and targeted therapeutics development of AD dementia. Our study also demonstrated the effectiveness of integrating public available summary-level pQTL data with GWAS summary data for mapping risk genes of complex human diseases.

Association between genetic risk and adherence to healthy lifestyle for developing age-related hearing loss

BACKGROUND

Previous studies have shown that lifestyle/environmental factors could accelerate the development of age-related hearing loss (ARHL). However, there has not yet been a study investigating the joint association among genetics, lifestyle/environmental factors, and adherence to healthy lifestyle for risk of ARHL. We aimed to assess the association between ARHL genetic variants, lifestyle/environmental factors, and adherence to healthy lifestyle as pertains to risk of ARHL.

METHODS

This case-control study included 376,464 European individuals aged 40 to 69 years, enrolled between 2006 and 2010 in the UK Biobank (UKBB). As a replication set, we also included a total of 26,523 individuals considered of European ancestry and 9834 individuals considered of African-American ancestry through the Penn Medicine Biobank (PMBB). The polygenic risk score (PRS) for ARHL was derived from a sensorineural hearing loss genome-wide association study from the FinnGen Consortium and categorized as low, intermediate, high, and very high. We selected lifestyle/environmental factors that have been previously studied in association with hearing loss. A composite healthy lifestyle score was determined using seven selected lifestyle behaviors and one environmental factor.

RESULTS

Of the 376,464 participants, 87,066 (23.1%) cases belonged to the ARHL group, and 289,398 (76.9%) individuals comprised the control group in the UKBB. A very high PRS for ARHL had a 49% higher risk of ARHL than those with low PRS (adjusted OR, 1.49; 95% CI, 1.36-1.62; P < .001), which was replicated in the PMBB cohort. A very poor lifestyle was also associated with risk of ARHL (adjusted OR, 3.03; 95% CI, 2.75-3.35; P < .001). These risk factors showed joint effects with the risk of ARHL. Conversely, adherence to healthy lifestyle in relation to hearing mostly attenuated the risk of ARHL even in individuals with very high PRS (adjusted OR, 0.21; 95% CI, 0.09-0.52; P < .001).

CONCLUSIONS

Our findings of this study demonstrated a significant joint association between genetic and lifestyle factors regarding ARHL. In addition, our analysis suggested that lifestyle adherence in individuals with high genetic risk could reduce the risk of ARHL.

A review of disease risk prediction methods and applications in the omics era

Risk prediction and disease prevention are the innovative care challenges of the 21st century. Apart from freeing the individual from the pain of disease, it will lead to low medical costs for society. Until very recently, risk assessments have ushered in a new era with the emergence of omics technologies, including genomics, transcriptomics, epigenomics, proteomics, and so on, which potentially advance the ability of biomarkers to aid prediction models. While risk prediction has achieved great success, there are still some challenges and limitations. We reviewed the general process of omics-based disease risk model construction and the applications in four typical diseases. Meanwhile, we highlighted the problems in current studies and explored the potential opportunities and challenges for future clinical practice.

Risk prediction and interaction analysis using polygenic risk score of type 2 diabetes in a Korean population

Joint modelling of genetic and environmental risk factors can provide important information to predict the risk of type 2 diabetes (T2D). Therefore, to predict the genetic risk of T2D, we constructed a polygenic risk score (PRS) using genotype data of one Korean cohort, KARE (745 cases and 2549 controls), and the genome-wide association study summary statistics of Biobank Japan. We evaluated the performance of PRS in an independent Korean cohort, HEXA (5684 cases and 35,703 controls). Individuals with T2D had a significantly higher mean PRS than controls (0.492 vs. - 0.078, p ≈ 0 ). PRS predicted the risk of T2D with an AUC of 0.658 (95% CI 0.651-0.666). We also evaluated interaction between PRS and waist circumference (WC) in the HEXA cohort. PRS exhibited a significant sub-multiplicative interaction with WC (ORinteraction 0.991, 95% CI 0.987-0.995, pinteraction = 4.93 × 10-6) in T2D. The effect of WC on T2D decreased as PRS increased. The sex-specific analyses produced similar interaction results, revealing a decreased WC effect on T2D as the PRS increased. In conclusion, the risk of WC for T2D may differ depending on PRS and those with a high PRS might develop T2D with a lower WC threshold. Our findings are expected to improve risk prediction for T2D and facilitate the identification of individuals at an increased risk of T2D.

Association of retinal optical coherence tomography metrics and polygenic risk scores with cognitive function and future cognitive decline

PURPOSE

To evaluate the potential of retinal optical coherence tomography (OCT) measurements and polygenic risk scores (PRS) to identify people at risk of cognitive impairment.

METHODS

Using OCT images from 50 342 UK Biobank participants, we examined associations between retinal layer thickness and genetic risk for neurodegenerative disease and combined these metrics with PRS to predict baseline cognitive function and future cognitive deterioration. Multivariate Cox proportional hazard models were used to predict cognitive performance. P values for retinal thickness analyses are false-discovery-rate-adjusted.

RESULTS

Higher Alzheimer's disease PRS was associated with a thicker inner nuclear layer (INL), chorio-scleral interface (CSI) and inner plexiform layer (IPL) (all p<0.05). Higher Parkinson's disease PRS was associated with thinner outer plexiform layer (p<0.001). Worse baseline cognitive performance was associated with thinner retinal nerve fibre layer (RNFL) (aOR=1.038, 95% CI (1.029 to 1.047), p<0.001) and photoreceptor (PR) segment (aOR=1.035, 95% CI (1.019 to 1.051), p<0.001), ganglion cell complex (aOR=1.007, 95% CI (1.002 to 1.013), p=0.004) and thicker ganglion cell layer (aOR=0.981, 95% CI (0.967 to 0.995), p=0.009), IPL (aOR=0.976, 95% CI (0.961 to 0.992), p=0.003), INL (aOR=0.923, 95% CI (0.905 to 0.941), p<0.001) and CSI (aOR=0.998, 95% CI (0.997 to 0.999), p<0.001). Worse future cognitive performance was associated with thicker IPL (aOR=0.945, 95% CI (0.915 to 0.999), p=0.045) and CSI (aOR=0.996, 95% CI (0.993 to 0.999) 95% CI, p=0.014). Prediction of cognitive decline was significantly improved with the addition of PRS and retinal measurements.

CONCLUSIONS AND RELEVANCE

Retinal OCT measurements are significantly associated with genetic risk of neurodegenerative disease and may serve as biomarkers predictive of future cognitive impairment.

SUMMIT-FA: a new resource for improved transcriptome imputation using functional annotations

Transcriptome-wide association studies (TWAS) integrate gene expression prediction models and genome-wide association studies (GWAS) to identify gene-trait associations. The power of TWAS is determined by the sample size of GWAS and the accuracy of the expression prediction model. Here, we present a new method, the Summary-level Unified Method for Modeling Integrated Transcriptome using Functional Annotations (SUMMIT-FA), which improves gene expression prediction accuracy by leveraging functional annotation resources and a large expression quantitative trait loci (eQTL) summary-level dataset. We build gene expression prediction models in whole blood using SUMMIT-FA with the comprehensive functional database MACIE and eQTL summary-level data from the eQTLGen consortium. We apply these models to GWAS for 24 complex traits and show that SUMMIT-FA identifies significantly more gene-trait associations and improves predictive power for identifying "silver standard" genes compared to several benchmark methods. We further conduct a simulation study to demonstrate the effectiveness of SUMMIT-FA.

The clinical relevance of a polygenic risk score for type 2 diabetes mellitus in the Korean population

The clinical utility of a type 2 diabetes mellitus (T2DM) polygenic risk score (PRS) in the East Asian population remains underexplored. We aimed to examine the potential prognostic value of a T2DM PRS and assess its viability as a clinical instrument. We first established a T2DM PRS for 5490 Korean individuals using East Asian Biobank data (269,487 samples). Subsequently, we assessed the predictive capability of this T2DM PRS in a prospective longitudinal study with baseline data and data from seven additional follow-ups. Our analysis showed that the T2DM PRS could predict the transition of glucose tolerance stages from normal glucose tolerance to prediabetes and from prediabetes to T2DM. Moreover, T2DM patients in the top-decile PRS group were more likely to be treated with insulin (hazard ratio = 1.69, p value = 2.31E-02) than were those in the remaining PRS groups. T2DM PRS values were significantly high in the severe diabetes subgroup, characterized by insulin resistance and β -cell dysfunction (p value = 0.0012). The prediction models with the T2DM PRS had significantly greater Harrel's C-indices than did corresponding models without it. By utilizing prospective longitudinal study data and extensive clinical risk factor information, our analysis provides valuable insights into the multifaceted clinical utility of the T2DM PRS.

A SIMPLE AND FLEXIBLE TEST OF SAMPLE EXCHANGEABILITY WITH APPLICATIONS TO STATISTICAL GENOMICS

In scientific studies involving analyses of multivariate data, basic but important questions often arise for the researcher: Is the sample exchangeable, meaning that the joint distribution of the sample is invariant to the ordering of the units? Are the features independent of one another, or perhaps the features can be grouped so that the groups are mutually independent? In statistical genomics, these considerations are fundamental to downstream tasks such as demographic inference and the construction of polygenic risk scores. We propose a non-parametric approach, which we call the V test, to address these two questions, namely, a test of sample exchangeability given dependency structure of features, and a test of feature independence given sample exchangeability. Our test is conceptually simple, yet fast and flexible. It controls the Type I error across realistic scenarios, and handles data of arbitrary dimensions by leveraging large-sample asymptotics. Through extensive simulations and a comparison against unsupervised tests of stratification based on random matrix theory, we find that our test compares favorably in various scenarios of interest. We apply the test to data from the 1000 Genomes Project, demonstrating how it can be employed to assess exchangeability of the genetic sample, or find optimal linkage disequilibrium (LD) splits for downstream analysis. For exchangeability assessment, we find that removing rare variants can substantially increase the p -value of the test statistic. For optimal LD splitting, the V test reports different optimal splits than previous approaches not relying on hypothesis testing. Software for our methods is available in R (CRAN: flintyR) and Python (PyPI: flintyPy).

Age-dependent genes in adipose stem and precursor cells affect regulation of fat cell differentiation and link aging to obesity via cellular and genetic interactions

BACKGROUND

Age and obesity are dominant risk factors for several common cardiometabolic disorders, and both are known to impair adipose tissue function. However, the underlying cellular and genetic factors linking aging and obesity on adipose tissue function have remained elusive. Adipose stem and precursor cells (ASPCs) are an understudied, yet crucial adipose cell type due to their deterministic adipocyte differentiation potential, which impacts the capacity to store fat in a metabolically healthy manner.

METHODS

We integrated subcutaneous adipose tissue (SAT) bulk (n=435) and large single-nucleus RNA sequencing (n=105) data with the UK Biobank (UKB) (n=391,701) data to study age-obesity interactions originating from ASPCs by performing cell-type decomposition, differential expression testing, cell-cell communication analyses, and construction of polygenic risk scores for body mass index (BMI).

RESULTS

We found that the SAT ASPC proportions significantly decrease with age in an obesity-dependent way consistently in two independent cohorts, both showing that the age dependency of ASPC proportions is abolished by obesity. We further identified 76 genes (72 SAT ASPC marker genes and 4 transcription factors regulating ASPC marker genes) that are differentially expressed by age in SAT and functionally enriched for developmental processes and adipocyte differentiation (i.e., adipogenesis). The 76 age-perturbed ASPC genes include multiple negative regulators of adipogenesis, such as RORA, SMAD3, TWIST2, and ZNF521, form tight clusters of longitudinally co-expressed genes during human adipogenesis, and show age-based differences in cellular interactions between ASPCs and adipose cell types. Finally, our genetic data demonstrate that cis-regional variants of these genes interact with age as predictors of BMI in an obesity-dependent way in the large UKB, while no such gene-age interaction on BMI is observed with non-age-dependent ASPC marker genes, thus independently confirming our cellular ASPC results at the biobank level.

CONCLUSIONS

Overall, we discover that obesity prematurely induces a decrease in ASPC proportions and identify 76 developmentally important ASPC genes that implicate altered negative regulation of fat cell differentiation as a mechanism for aging and directly link aging to obesity via significant cellular and genetic interactions.

Inferring a directed acyclic graph of phenotypes from GWAS summary statistics

Estimating phenotype networks is a growing field in computational biology. It deepens the understanding of disease etiology and is useful in many applications. In this study, we present a method that constructs a phenotype network by assuming a Gaussian linear structure model embedding a directed acyclic graph (DAG). We utilize genetic variants as instrumental variables and show how our method only requires access to summary statistics from a genome-wide association study (GWAS) and a reference panel of genotype data. Besides estimation, a distinct feature of the method is its summary statistics-based likelihood ratio test on directed edges. We applied our method to estimate a causal network of 29 cardiovascular-related proteins and linked the estimated network to Alzheimer's disease (AD). A simulation study was conducted to demonstrate the effectiveness of this method. An R package sumdag implementing the proposed method, all relevant code, and a Shiny application are available.

Dementia with Lewy Bodies: Genomics, Transcriptomics, and Its Future with Data Science

Dementia with Lewy bodies (DLB) is a significant public health issue. It is the second most common neurodegenerative dementia and presents with severe neuropsychiatric symptoms. Genomic and transcriptomic analyses have provided some insight into disease pathology. Variants within SNCA, GBA, APOE, SNCB, and MAPT have been shown to be associated with DLB in repeated genomic studies. Transcriptomic analysis, conducted predominantly on candidate genes, has identified signatures of synuclein aggregation, protein degradation, amyloid deposition, neuroinflammation, mitochondrial dysfunction, and the upregulation of heat-shock proteins in DLB. Yet, the understanding of DLB molecular pathology is incomplete. This precipitates the current clinical position whereby there are no available disease-modifying treatments or blood-based diagnostic biomarkers. Data science methods have the potential to improve disease understanding, optimising therapeutic intervention and drug development, to reduce disease burden. Genomic prediction will facilitate the early identification of cases and the timely application of future disease-modifying treatments. Transcript-level analyses across the entire transcriptome and machine learning analysis of multi-omic data will uncover novel signatures that may provide clues to DLB pathology and improve drug development. This review will discuss the current genomic and transcriptomic understanding of DLB, highlight gaps in the literature, and describe data science methods that may advance the field.

A high-resolution PheWAS approach to alcohol-related polygenic risk scores reveals mechanistic influences of alcohol reinforcing value and drinking motives

AIMS

This study uses a high-resolution phenome-wide approach to evaluate the motivational mechanisms of polygenic risk scores (PRSs) that have been robustly associated with coarse alcohol phenotypes in large-scale studies.

METHODS

In a community-based sample of 1534 Europeans, we examined genome-wide PRSs for the Alcohol Use Disorders Identification Test (AUDIT), drinks per week, alcohol use disorder (AUD), problematic alcohol use (PAU), and general addiction, in relation to 42 curated phenotypes. The curated phenotypes were in seven categories: alcohol consumption, alcohol reinforcing value, drinking motives, other addictive behaviors, commonly comorbid psychiatric syndromes, impulsivity, and personality traits.

RESULTS

The PRS for each alcohol phenotype was validated via its within-sample association with the corresponding phenotype (adjusted R2s = 0.35-1.68%, Ps = 0.012-3.6 × 10-7) with the exception of AUD. All PRSs were positively associated with alcohol reinforcing value and drinking motives, with the strongest effects from AUDIT-consumption (adjusted R2s = 0.45-1.33%, Ps = 0.006-3.6 × 10-5) and drinks per week PRSs (adjusted R2s = 0.52-2.28%, Ps = 0.004-6.6 × 10-9). Furthermore, the PAU and drinks per week PRSs were positively associated with adverse childhood experiences (adjusted R2s = 0.6-0.7%, Ps = 0.0001-4.8 × 10-4).

CONCLUSIONS

These results implicate alcohol reinforcing value and drinking motives as genetically-influenced mechanisms using PRSs for the first time. The findings also highlight the value of dissecting genetic influence on alcohol involvement through diverse phenotypic risk pathways but also the need for future studies with both phenotypic richness and larger samples.

Evaluating the Efficacy of Type 2 Diabetes Polygenic Risk Scores in an Independent European Population

Numerous type 2 diabetes (T2D) polygenic risk scores (PGSs) have been developed to predict individuals' predisposition to the disease. An independent assessment and verification of the best-performing PGS are warranted to allow for a rapid application of developed models. To date, only 3% of T2D PGSs have been evaluated. In this study, we assessed all (n = 102) presently published T2D PGSs in an independent cohort of 3718 individuals, which has not been included in the construction or fine-tuning of any T2D PGS so far. We further chose the best-performing PGS, assessed its performance across major population principal component analysis (PCA) clusters, and compared it with newly developed population-specific T2D PGS. Our findings revealed that 88% of the published PGSs were significantly associated with T2D; however, their performance was lower than what had been previously reported. We found a positive association of PGS improvement over the years (p-value = 8.01 × 10-4 with PGS002771 currently showing the best discriminatory power (area under the receiver operating characteristic (AUROC) = 0.669) and PGS003443 exhibiting the strongest association PGS003443 (odds ratio (OR) = 1.899). Further investigation revealed no difference in PGS performance across major population PCA clusters and when compared with newly developed population-specific PGS. Our findings revealed a positive trend in T2D PGS performance, consistently identifying high-T2D-risk individuals in an independent European population.

Estimating Disorder Probability Based on Polygenic Prediction Using the BPC Approach

Polygenic Scores (PGSs) summarize an individual's genetic propensity for a given trait in a single value, based on SNP effect sizes derived from Genome-Wide Association Study (GWAS) results. Methods have been developed that apply Bayesian approaches to improve the prediction accuracy of PGSs through optimization of estimated effect sizes. While these methods are generally well-calibrated for continuous traits (implying the predicted values are on average equal to the true trait values), they are not well-calibrated for binary disorder traits in ascertained samples. This is a problem because well-calibrated PGSs are needed to reliably compute the absolute disorder probability for an individual to facilitate future clinical implementation. Here we introduce the Bayesian polygenic score Probability Conversion (BPC) approach, which computes an individual's predicted disorder probability using GWAS summary statistics, an existing Bayesian PGS method (e.g. PRScs, SBayesR), the individual's genotype data, and a prior disorder probability. The BPC approach transforms the PGS to its underlying liability scale, computes the variances of the PGS in cases and controls, and applies Bayes' Theorem to compute the absolute disorder probability; it is practical in its application as it does not require a tuning dataset with both genotype and phenotype data. We applied the BPC approach to extensive simulated data and empirical data of nine disorders. The BPC approach yielded well-calibrated results that were consistently better than the results of another recently published approach.

An X Chromosome Transcriptome Wide Association Study Implicates ARMCX6 in Alzheimer's Disease

Background

The X chromosome is often omitted in disease association studies despite containing thousands of genes that may provide insight into well-known sex differences in the risk of Alzheimer's disease (AD).

Objective

To model the expression of X chromosome genes and evaluate their impact on AD risk in a sex-stratified manner.

Methods

Using elastic net, we evaluated multiple modeling strategies in a set of 175 whole blood samples and 126 brain cortex samples, with whole genome sequencing and RNA-seq data. SNPs (MAF > 0.05) within the cis-regulatory window were used to train tissue-specific models of each gene. We apply the best models in both tissues to sex-stratified summary statistics from a meta-analysis of Alzheimer's Disease Genetics Consortium (ADGC) studies to identify AD-related genes on the X chromosome.

Results

Across different model parameters, sample sex, and tissue types, we modeled the expression of 217 genes (95 genes in blood and 135 genes in brain cortex). The average model R2 was 0.12 (range from 0.03 to 0.34). We also compared sex-stratified and sex-combined models on the X chromosome. We further investigated genes that escaped X chromosome inactivation (XCI) to determine if their genetic regulation patterns were distinct. We found ten genes associated with AD at p < 0.05, with only ARMCX6 in female brain cortex (p = 0.008) nearing the significance threshold after adjusting for multiple testing (α = 0.002).

Conclusions

We optimized the expression prediction of X chromosome genes, applied these models to sex-stratified AD GWAS summary statistics, and identified one putative AD risk gene, ARMCX6.

Principles and methods for transferring polygenic risk scores across global populations

Polygenic risk scores (PRSs) summarize the genetic predisposition of a complex human trait or disease and may become a valuable tool for advancing precision medicine. However, PRSs that are developed in populations of predominantly European genetic ancestries can increase health disparities due to poor predictive performance in individuals of diverse and complex genetic ancestries. We describe genetic and modifiable risk factors that limit the transferability of PRSs across populations and review the strengths and weaknesses of existing PRS construction methods for diverse ancestries. Developing PRSs that benefit global populations in research and clinical settings provides an opportunity for innovation and is essential for health equity.

Inferring disease architecture and predictive ability with LDpred2-auto

LDpred2 is a widely used Bayesian method for building polygenic scores (PGSs). LDpred2-auto can infer the two parameters from the LDpred model, the SNP heritability h2 and polygenicity p, so that it does not require an additional validation dataset to choose best-performing parameters. The main aim of this paper is to properly validate the use of LDpred2-auto for inferring multiple genetic parameters. Here, we present a new version of LDpred2-auto that adds an optional third parameter α to its model, for modeling negative selection. We then validate the inference of these three parameters (or two, when using the previous model). We also show that LDpred2-auto provides per-variant probabilities of being causal that are well calibrated and can therefore be used for fine-mapping purposes. We also introduce a formula to infer the out-of-sample predictive performance r2 of the resulting PGS directly from the Gibbs sampler of LDpred2-auto. Finally, we extend the set of HapMap3 variants recommended to use with LDpred2 with 37% more variants to improve the coverage of this set, and we show that this new set of variants captures 12% more heritability and provides 6% more predictive performance, on average, in UK Biobank analyses.

An in-depth association analysis of genetic variants within nicotine-related loci: Meeting in middle of GWAS and genetic fine-mapping

In the last two decades of Genome-wide association studies (GWAS), nicotine-dependence-related genetic loci (e.g., nicotinic acetylcholine receptor - nAChR subunit genes) are among the most replicable genetic findings. Although GWAS results have reported tens of thousands of SNPs within these loci, further analysis (e.g., fine-mapping) is required to identify the causal variants. However, it is computationally challenging for existing fine-mapping methods to reliably identify causal variants from thousands of candidate SNPs based on the posterior inclusion probability. To address this challenge, we propose a new method to select SNPs by jointly modeling the SNP-wise inference results and the underlying structured network patterns of the linkage disequilibrium (LD) matrix. We use adaptive dense subgraph extraction method to recognize the latent network patterns of the LD matrix and then apply group LASSO to select causal variant candidates. We applied this new method to the UK biobank data to identify the causal variant candidates for nicotine addiction. Eighty-one nicotine addiction-related SNPs (i.e.,-log(p) > 50) of nAChR were selected, which are highly correlated (average r2>0.8) although they are physically distant (e.g., >200 kilobase away) and from various genes. These findings revealed that distant SNPs from different genes can show higher LD r2 than their neighboring SNPs, and jointly contribute to a complex trait like nicotine addiction.

Inference of causal metabolite networks in the presence of invalid instrumental variables with GWAS summary data

We propose structural equation models (SEMs) as a general framework to infer causal networks for metabolites and other complex traits. Traditionally SEMs are used only for individual-level data under the assumption that all instrumental variables (IVs) are valid. To overcome these limitations, we propose both one- and two-sample approaches for causal network inference based on SEMs that can: (1) perform causal analysis and discover causal relationships among multiple traits; (2) account for the possible presence of some invalid IVs; (3) allow for data analysis using only genome-wide association studies (GWAS) summary statistics when individual-level data are not available; (4) consider the possibility of bidirectional relationships between traits. Our method employs a simple stepwise selection to identify invalid IVs, thus avoiding false positives while possibly increasing true discoveries based on two-stage least squares (2SLS). We use both real GWAS data and simulated data to demonstrate the superior performance of our method over the standard 2SLS/SEMs. For real data analysis, our proposed approach is applied to a human blood metabolite GWAS summary data set to uncover putative causal relationships among the metabolites; we also identify some metabolites (putative) causal to Alzheimer's disease (AD), which, along with the inferred causal metabolite network, suggest some possible pathways of metabolites involved in AD.

Polygenic risk scores identify heterogeneity in asthma and chronic obstructive pulmonary disease

BACKGROUND

Asthma and chronic obstructive pulmonary disease (COPD) have distinct and overlapping genetic and clinical features.

OBJECTIVE

We sought to test the hypothesis that polygenic risk scores (PRSs) for asthma (PRSAsthma) and spirometry (FEV1 and FEV1/forced vital capacity; PRSspiro) would demonstrate differential associations with asthma, COPD, and asthma-COPD overlap (ACO).

METHODS

We developed and tested 2 asthma PRSs and applied the higher performing PRSAsthma and a previously published PRSspiro to research (Genetic Epidemiology of COPD study and Childhood Asthma Management Program, with spirometry) and electronic health record-based (Mass General Brigham Biobank and Genetic Epidemiology Research on Adult Health and Aging [GERA]) studies. We assessed the association of PRSs with COPD and asthma using modified random-effects and binary-effects meta-analyses, and ACO and asthma exacerbations in specific cohorts. Models were adjusted for confounders and genetic ancestry.

RESULTS

In meta-analyses of 102,477 participants, the PRSAsthma (odds ratio [OR] per SD, 1.16 [95% CI, 1.14-1.19]) and PRSspiro (OR per SD, 1.19 [95% CI, 1.17-1.22]) both predicted asthma, whereas the PRSspiro predicted COPD (OR per SD, 1.25 [95% CI, 1.21-1.30]). However, results differed by cohort. The PRSspiro was not associated with COPD in GERA and Mass General Brigham Biobank. In the Genetic Epidemiology of COPD study, the PRSAsthma (OR per SD: Whites, 1.3; African Americans, 1.2) and PRSspiro (OR per SD: Whites, 2.2; African Americans, 1.6) were both associated with ACO. In GERA, the PRSAsthma was associated with asthma exacerbations (OR, 1.18) in Whites; the PRSspiro was associated with asthma exacerbations in White, LatinX, and East Asian participants.

CONCLUSIONS

PRSs for asthma and spirometry are both associated with ACO and asthma exacerbations. Genetic prediction performance differs in research versus electronic health record-based cohorts.

Evaluation of European-based polygenic risk score for breast cancer in Ashkenazi Jewish women in Israel

BACKGROUND

Polygenic risk score (PRS), calculated based on genome-wide association studies (GWASs), can improve breast cancer (BC) risk assessment. To date, most BC GWASs have been performed in individuals of European (EUR) ancestry, and the generalisation of EUR-based PRS to other populations is a major challenge. In this study, we examined the performance of EUR-based BC PRS models in Ashkenazi Jewish (AJ) women.

METHODS

We generated PRSs based on data on EUR women from the Breast Cancer Association Consortium (BCAC). We tested the performance of the PRSs in a cohort of 2161 AJ women from Israel (1437 cases and 724 controls) from BCAC (BCAC cohort from Israel (BCAC-IL)). In addition, we tested the performance of these EUR-based BC PRSs, as well as the established 313-SNP EUR BC PRS, in an independent cohort of 181 AJ women from Hadassah Medical Center (HMC) in Israel.

RESULTS

In the BCAC-IL cohort, the highest OR per 1 SD was 1.56 (±0.09). The OR for AJ women at the top 10% of the PRS distribution compared with the middle quintile was 2.10 (±0.24). In the HMC cohort, the OR per 1 SD of the EUR-based PRS that performed best in the BCAC-IL cohort was 1.58±0.27. The OR per 1 SD of the commonly used 313-SNP BC PRS was 1.64 (±0.28).

CONCLUSIONS

Extant EUR GWAS data can be used for generating PRSs that identify AJ women with markedly elevated risk of BC and therefore hold promise for improving BC risk assessment in AJ women.

Inferring a directed acyclic graph of phenotypes from GWAS summary statistics

Estimating phenotype networks is a growing field in computational biology. It deepens the understanding of disease etiology and is useful in many applications. In this study, we present a method that constructs a phenotype network by assuming a Gaussian linear structure model embedding a directed acyclic graph (DAG). We utilize genetic variants as instrumental variables and show how our method only requires access to summary statistics from a genome-wide association study (GWAS) and a reference panel of genotype data. Besides estimation, a distinct feature of the method is its summary statistics-based likelihood ratio test on directed edges. We applied our method to estimate a causal network of 29 cardiovascular-related proteins and linked the estimated network to Alzheimer's disease (AD). A simulation study was conducted to demonstrate the effectiveness of this method. An R package sumdag implementing the proposed method, all relevant code, and a Shiny application are available at https://github.com/chunlinli/sumdag.