Modeling Linkage Disequilibrium Increases Accuracy of Polygenic Risk Scores

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.

Distinct explanations underlie gene-environment interactions in the UK Biobank

The role of gene-environment (GxE) interaction in disease and complex trait architectures is widely hypothesized, but currently unknown. Here, we apply three statistical approaches to quantify and distinguish three different types of GxE interaction for a given trait and E variable. First, we detect locus-specific GxE interaction by testing for genetic correlation r g < 1 across E bins. Second, we detect genome-wide effects of the E variable on genetic variance by leveraging polygenic risk scores (PRS) to test for significant PRSxE in a regression of phenotypes on PRS, E, and PRSxE, together with differences in SNP-heritability across E bins. Third, we detect genome-wide proportional amplification of genetic and environmental effects as a function of the E variable by testing for significant PRSxE with no differences in SNP-heritability across E bins. Simulations show that these approaches achieve high sensitivity and specificity in distinguishing these three GxE scenarios. We applied our framework to 33 UK Biobank traits (25 quantitative traits and 8 diseases; average N = 325 K ) and 10 E variables spanning lifestyle, diet, and other environmental exposures. First, we identified 19 trait-E pairs with r g significantly < 1 (FDR<5%) (average r g = 0.95 ); for example, white blood cell count had r g = 0.95 (s.e. 0.01) between smokers and non-smokers. Second, we identified 28 trait-E pairs with significant PRSxE and significant SNP-heritability differences across E bins; for example, BMI had a significant PRSxE for physical activity (P=4.6e-5) with 5% larger SNP-heritability in the largest versus smallest quintiles of physical activity (P=7e-4). Third, we identified 15 trait-E pairs with significant PRSxE with no SNP-heritability differences across E bins; for example, waist-hip ratio adjusted for BMI had a significant PRSxE effect for time spent watching television (P=5e-3) with no SNP-heritability differences. Across the three scenarios, 8 of the trait-E pairs involved disease traits, whose interpretation is complicated by scale effects. Analyses using biological sex as the E variable produced additional significant findings in each of the three scenarios. Overall, we infer a significant contribution of GxE and GxSex effects to complex trait and disease variance.

Polygenic risk scores associate with blood pressure traits across the lifespan

AIMS

Hypertension is a major modifiable cause of morbidity and mortality that affects over 1 billion people worldwide. Blood pressure (BP) traits have a strong genetic component that can be quantified with polygenic risk scores (PRSs). To date, the performance of BP PRSs has mainly been assessed in adults, and less is known about polygenic hypertension risk in childhood.

METHODS AND RESULTS

Multiple PRSs for systolic BP (SBP), diastolic BP (DBP), and pulse pressure were developed using either genome-wide significant weights, pruning and thresholding, or Bayesian regression. Among 87 total PRSs, the top performer for each trait was applied in independent cohorts of children and adult to assess genotype-phenotype associations and disease risk across the lifespan. Differences between those with low (1st decile), average (2nd-9th decile), and high (10th decile) PRS emerge in the first years of life and are maintained throughout adulthood. These diverging BP trajectories also seem to affect cardiovascular and renal disease risk, with increased risk observed among those in the top decile and reduced risk among those in the bottom decile of the polygenic risk distribution compared with the rest of the population.

CONCLUSION

Genetic risk factors are associated with BP traits across the lifespan, beginning in the first years of life. Given the importance of exposure time in disease pathogenesis and the early rise in BP levels among those genetically susceptible, PRSs may help identify high-risk individuals prior to hypertension onset, facilitate primordial prevention, and reduce the burden of this public health challenge.

No gene by stressful life events interaction on individual differences in adults' self-control

Background

Difficulty with self-control, or the ability to alter impulses and behavior in a goal-directed way, predicts interpersonal conflict, lower socioeconomic attainments, and more adverse health outcomes. Etiological understanding, and intervention for low self-control is, therefore, a public health goal. A prominent developmental theory proposes that individuals with high genetic propensity for low self-control that are also exposed to stressful environments may be most at-risk of low levels of self-control. Here we examine if polygenic measures associated with behaviors marked by low self-control interact with stressful life events in predicting self-control.

Methods

Leveraging molecular data from a large population-based Dutch sample (N = 7,090, Mage = 41.2) to test for effects of genetics (i.e., polygenic scores for ADHD and aggression), stressful life events (e.g., traffic accident, violent assault, financial problems), and a gene-by-stress interaction on self-control (measured with the ASEBA Self-Control Scale).

Results

Both genetics (β =.03 -.04, p <.001) and stressful life events (β = .11 -.14, p <.001) were associated with individual differences in self-control. We find no evidence of a gene-by-stressful life events interaction on individual differences in adults' self-control.

Conclusion

Our findings are consistent with the notion that genetic influences and stressful life events exert largely independent effects on adult self-control. However, the small effect sizes of polygenic scores increases the likelihood of null results. Genetically-informed longitudinal research in large samples can further inform the etiology of individual differences in self-control from early childhood into later adulthood and its downstream implications for public health.

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.

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.

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.

Integrative polygenic risk score improves the prediction accuracy of complex traits and diseases

Polygenic risk scores (PRSs) are an emerging tool to predict the clinical phenotypes and outcomes of individuals. We propose PRSmix, a framework that leverages the PRS corpus of a target trait to improve prediction accuracy, and PRSmix+, which incorporates genetically correlated traits to better capture the human genetic architecture for 47 and 32 diseases/traits in European and South Asian ancestries, respectively. PRSmix demonstrated a mean prediction accuracy improvement of 1.20-fold (95% confidence interval [CI], [1.10; 1.3]; p = 9.17 × 10-5) and 1.19-fold (95% CI, [1.11; 1.27]; p = 1.92 × 10-6), and PRSmix+ improved the prediction accuracy by 1.72-fold (95% CI, [1.40; 2.04]; p = 7.58 × 10-6) and 1.42-fold (95% CI, [1.25; 1.59]; p = 8.01 × 10-7) in European and South Asian ancestries, respectively. Compared to the previously cross-trait-combination methods with scores from pre-defined correlated traits, we demonstrated that our method improved prediction accuracy for coronary artery disease up to 3.27-fold (95% CI, [2.1; 4.44]; p value after false discovery rate (FDR) correction = 2.6 × 10-4). Our method provides a comprehensive framework to benchmark and leverage the combined power of PRS for maximal performance in a desired target population.

A perspective on genetic and polygenic risk scores-advances and limitations and overview of associated tools

Polygenetic Risk Scores are used to evaluate an individual's vulnerability to developing specific diseases or conditions based on their genetic composition, by taking into account numerous genetic variations. This article provides an overview of the concept of Polygenic Risk Scores (PRS). We elucidate the historical advancements of PRS, their advantages and shortcomings in comparison with other predictive methods, and discuss their conceptual limitations in light of the complexity of biological systems. Furthermore, we provide a survey of published tools for computing PRS and associated resources. The various tools and software packages are categorized based on their technical utility for users or prospective developers. Understanding the array of available tools and their limitations is crucial for accurately assessing and predicting disease risks, facilitating early interventions, and guiding personalized healthcare decisions. Additionally, we also identify potential new avenues for future bioinformatic analyzes and advancements related to PRS.

Genetic similarity between relatives provides evidence on the presence and history of assortative mating

Assortative mating - the non-random mating of individuals with similar traits - is known to increase trait-specific genetic variance and genetic similarity between relatives. However, empirical evidence is limited for many traits, and the implications hinge on whether assortative mating has started recently or many generations ago. Here we show theoretically and empirically that genetic similarity between relatives can provide evidence on the presence and history of assortative mating. First, we employed path analysis to understand how assortative mating affects genetic similarity between family members across generations, finding that similarity between distant relatives is more affected than close relatives. Next, we correlated polygenic indices of 47,135 co-parents from the Norwegian Mother, Father, and Child Cohort Study (MoBa) and found genetic evidence of assortative mating in nine out of sixteen examined traits. The same traits showed elevated similarity between relatives, especially distant relatives. Six of the nine traits, including educational attainment, showed greater genetic variance among offspring, which is inconsistent with stable assortative mating over many generations. These results suggest an ongoing increase in familial similarity for these traits. The implications of this research extend to genetic methodology and the understanding of social and economic disparities.

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.

AI-enabled evaluation of genome-wide association relevance and polygenic risk score prediction in Alzheimer's disease

GWAS focuses on significance loosing false positives; machine learning probes sub-significant features relying on predictivity. Yet, these are far from orthogonal. We sought to explore how these inform each other in sub-genome-wide significant situations to define relevance for predictive features. We introduce the SVM-based RubricOE that selects heavily cross-validated feature sets, and LDpred2 PRS as a strong contrast to SVM, to explore significance and predictivity. Our Alzheimer's test case notoriously lacks strong genetic signals except for few very strong phenotype-SNP associations, which suits the problem we are exploring. We found that the most significant SNPs among ML and PRS-selected SNPs captured most of the predictivity, while weaker associations tend also to contribute weakly to predictivity. SNPs with weak associations tend not to contribute to predictivity, but deletion of these features does not injure it. Significance provides a ranking that helps identify weakly predictive features.

Cumulative Effects of Resting-State Connectivity Across All Brain Networks Significantly Correlate with Attention-Deficit Hyperactivity Disorder Symptoms

Identification of replicable neuroimaging correlates of attention-deficit hyperactivity disorder (ADHD) has been hindered by small sample sizes, small effects, and heterogeneity of methods. Given evidence that ADHD is associated with alterations in widely distributed brain networks and the small effects of individual brain features, a whole-brain perspective focusing on cumulative effects is warranted. The use of large, multisite samples is crucial for improving reproducibility and clinical utility of brain-wide MRI association studies. To address this, a polyneuro risk score (PNRS) representing cumulative, brain-wide, ADHD-associated resting-state functional connectivity was constructed and validated using data from the Adolescent Brain Cognitive Development (ABCD, N = 5,543, 51.5% female) study, and was further tested in the independent Oregon-ADHD-1000 case-control cohort (N = 553, 37.4% female). The ADHD PNRS was significantly associated with ADHD symptoms in both cohorts after accounting for relevant covariates (p < 0.001). The most predictive PNRS involved all brain networks, though the strongest effects were concentrated among the default mode and cingulo-opercular networks. In the longitudinal Oregon-ADHD-1000, non-ADHD youth had significantly lower PNRS (Cohen's d = -0.318, robust p = 5.5 × 10-4) than those with persistent ADHD (age 7-19). The PNRS, however, did not mediate polygenic risk for ADHD. Brain-wide connectivity was robustly associated with ADHD symptoms in two independent cohorts, providing further evidence of widespread dysconnectivity in ADHD. Evaluation in enriched samples demonstrates the promise of the PNRS approach for improving reproducibility in neuroimaging studies and unraveling the complex relationships between brain connectivity and behavioral disorders.

Using encrypted genotypes and phenotypes for collaborative genomic analyses to maintain data confidentiality

To adhere to and capitalize on the benefits of the FAIR (findable, accessible, interoperable, and reusable) principles in agricultural genome-to-phenome studies, it is crucial to address privacy and intellectual property issues that prevent sharing and reuse of data in research and industry. Direct sharing of genotype and phenotype data is often prohibited due to intellectual property and privacy concerns. Thus, there is a pressing need for encryption methods that obscure confidential aspects of the data, without affecting the outcomes of certain statistical analyses. A homomorphic encryption method for genotypes and phenotypes (HEGP) has been proposed for single-marker regression in genome-wide association studies (GWAS) using linear mixed models with Gaussian errors. This methodology permits frequentist likelihood-based parameter estimation and inference. In this paper, we extend HEGP to broader applications in genome-to-phenome analyses. We show that HEGP is suited to commonly used linear mixed models for genetic analyses of quantitative traits including genomic best linear unbiased prediction (GBLUP) and ridge-regression best linear unbiased prediction (RR-BLUP), as well as Bayesian variable selection methods (e.g. those in Bayesian Alphabet), for genetic parameter estimation, genomic prediction, and GWAS. By advancing the capabilities of HEGP, we offer researchers and industry professionals a secure and efficient approach for collaborative genomic analyses while preserving data confidentiality.

A linear adjustment-based approach to posterior drift in transfer learning

We present new models and methods for the posterior drift problem where the regression function in the target domain is modelled as a linear adjustment, on an appropriate scale, of that in the source domain, and study the theoretical properties of our proposed estimators in the binary classification problem. The core idea of our model inherits the simplicity and the usefulness of generalized linear models and accelerated failure time models from the classical statistics literature. Our approach is shown to be flexible and applicable in a variety of statistical settings, and can be adopted for transfer learning problems in various domains including epidemiology, genetics and biomedicine. As concrete applications, we illustrate the power of our approach (i) through mortality prediction for British Asians by borrowing strength from similar data from the larger pool of British Caucasians, using the UK Biobank data, and (ii) in overcoming a spurious correlation present in the source domain of the Waterbirds dataset.

The interaction between polygenic risk and environmental influences: A direct test of the 3P model of insomnia in adolescents

BACKGROUND

Stress is a universal phenomenon and one of the most common precipitants of insomnia. However, not everyone develops insomnia after experiencing a stressful life event. This study aims to test aspects of Spielman's '3P model of insomnia' (during adolescence) by exploring the extent to which: (a) insomnia symptoms are predicted by polygenic scores (PGS); (b) life events predict insomnia symptoms; (c) the interaction between PGS and life events contribute to the prediction of insomnia symptoms; (d) gene-environment interaction effects remain after controlling for sex.

METHODS

The sample comprised 4,629 twins aged 16 from the Twin Early Development Study who reported on their insomnia symptoms and life events. PGS for insomnia were calculated. In order to test the main hypothesis of this study (a significant interaction between PGS and negative life events), we fitted a series of mixed effect regressions.

RESULTS

The best fit was provided by the model including sex, PGS for insomnia, negative life events, and their interactions (AIC = 26,158.7). Our results show that the association between insomnia symptoms and negative life events is stronger for those with a higher genetic risk for insomnia.

CONCLUSIONS

This work sheds light on the complex relationship between genetic and environmental factors implicated for insomnia. This study has tested for the first time the interaction between genetic predisposition (PGS) for insomnia and environmental stressors (negative life events) in adolescents. This work represents a direct test of components of Spielman's 3P model for insomnia which is supported by our results.

Canadian COVID-19 host genetics cohort replicates known severity associations

The HostSeq initiative recruited 10,059 Canadians infected with SARS-CoV-2 between March 2020 and March 2023, obtained clinical information on their disease experience and whole genome sequenced (WGS) their DNA. We analyzed the WGS data for genetic contributors to severe COVID-19 (considering 3,499 hospitalized cases and 4,975 non-hospitalized after quality control). We investigated the evidence for replication of loci reported by the International Host Genetics Initiative (HGI); analyzed the X chromosome; conducted rare variant gene-based analysis and polygenic risk score testing. Population stratification was adjusted for using meta-analysis across ancestry groups. We replicated two loci identified by the HGI for COVID-19 severity: the LZTFL1/SLC6A20 locus on chromosome 3 and the FOXP4 locus on chromosome 6 (the latter with a variant significant at P < 5E-8). We found novel significant associations with MRAS and WDR89 in gene-based analyses, and constructed a polygenic risk score that explained 1.01% of the variance in severe COVID-19. This study provides independent evidence confirming the robustness of previously identified COVID-19 severity loci by the HGI and identifies novel genes for further investigation.

Using Instrumental Variables to Measure Causation over Time in Cross-Lagged Panel Models

Cross-lagged panel models (CLPMs) are commonly used to estimate causal influences between two variables with repeated assessments. The lagged effects in a CLPM depend on the time interval between assessments, eventually becoming undetectable at longer intervals. To address this limitation, we incorporate instrumental variables (IVs) into the CLPM with two study waves and two variables. Doing so enables estimation of both the lagged (i.e., "distal") effects and the bidirectional cross-sectional (i.e., "proximal") effects at each wave. The distal effects reflect Granger-causal influences across time, which decay with increasing time intervals. The proximal effects capture causal influences that accrue over time and can help infer causality when the distal effects become undetectable at longer intervals. Significant proximal effects, with a negligible distal effect, would imply that the time interval is too long to estimate a lagged effect at that time interval using the standard CLPM. Through simulations and an empirical application, we demonstrate the impact of time intervals on causal inference in the CLPM and present modeling strategies to detect causal influences regardless of the time interval in a study. Furthermore, to motivate empirical applications of the proposed model, we highlight the utility and limitations of using genetic variables as IVs in large-scale panel studies.

Distinguishing different psychiatric disorders using DDx-PRS

Despite great progress on methods for case-control polygenic prediction (e.g. schizophrenia vs. control), there remains an unmet need for a method that genetically distinguishes clinically related disorders (e.g. schizophrenia (SCZ) vs. bipolar disorder (BIP) vs. depression (MDD) vs. control); such a method could have important clinical value, especially at disorder onset when differential diagnosis can be challenging. Here, we introduce a method, Differential Diagnosis-Polygenic Risk Score (DDx-PRS), that jointly estimates posterior probabilities of each possible diagnostic category (e.g. SCZ=50%, BIP=25%, MDD=15%, control=10%) by modeling variance/covariance structure across disorders, leveraging case-control polygenic risk scores (PRS) for each disorder (computed using existing methods) and prior clinical probabilities for each diagnostic category. DDx-PRS uses only summary-level training data and does not use tuning data, facilitating implementation in clinical settings. In simulations, DDx-PRS was well-calibrated (whereas a simpler approach that analyzes each disorder marginally was poorly calibrated), and effective in distinguishing each diagnostic category vs. the rest. We then applied DDx-PRS to Psychiatric Genomics Consortium SCZ/BIP/MDD/control data, including summary-level training data from 3 case-control GWAS ( N =41,917-173,140 cases; total N =1,048,683) and held-out test data from different cohorts with equal numbers of each diagnostic category (total N =11,460). DDx-PRS was well-calibrated and well-powered relative to these training sample sizes, attaining AUCs of 0.66 for SCZ vs. rest, 0.64 for BIP vs. rest, 0.59 for MDD vs. rest, and 0.68 for control vs. rest. DDx-PRS produced comparable results to methods that leverage tuning data, confirming that DDx-PRS is an effective method. True diagnosis probabilities in top deciles of predicted diagnosis probabilities were considerably larger than prior baseline probabilities, particularly in projections to larger training sample sizes, implying considerable potential for clinical utility under certain circumstances. In conclusion, DDx-PRS is an effective method for distinguishing clinically related disorders.

A primer on the use of machine learning to distil knowledge from data in biological psychiatry

Applications of machine learning in the biomedical sciences are growing rapidly. This growth has been spurred by diverse cross-institutional and interdisciplinary collaborations, public availability of large datasets, an increase in the accessibility of analytic routines, and the availability of powerful computing resources. With this increased access and exposure to machine learning comes a responsibility for education and a deeper understanding of its bases and bounds, borne equally by data scientists seeking to ply their analytic wares in medical research and by biomedical scientists seeking to harness such methods to glean knowledge from data. This article provides an accessible and critical review of machine learning for a biomedically informed audience, as well as its applications in psychiatry. The review covers definitions and expositions of commonly used machine learning methods, and historical trends of their use in psychiatry. We also provide a set of standards, namely Guidelines for REporting Machine Learning Investigations in Neuropsychiatry (GREMLIN), for designing and reporting studies that use machine learning as a primary data-analysis approach. Lastly, we propose the establishment of the Machine Learning in Psychiatry (MLPsych) Consortium, enumerate its objectives, and identify areas of opportunity for future applications of machine learning in biological psychiatry. This review serves as a cautiously optimistic primer on machine learning for those on the precipice as they prepare to dive into the field, either as methodological practitioners or well-informed consumers.

Type 2 diabetes and its genetic susceptibility are associated with increased severity and mortality of COVID-19 in UK Biobank

Type 2 diabetes (T2D) is known as one of the important risk factors for the severity and mortality of COVID-19. Here, we evaluate the impact of T2D and its genetic susceptibility on the severity and mortality of COVID-19, using 459,119 individuals in UK Biobank. Utilizing the polygenic risk scores (PRS) for T2D, we identified a significant association between T2D or T2D PRS, and COVID-19 severity. We further discovered the efficacy of vaccination and the pivotal role of T2D-related genetics in the pathogenesis of severe COVID-19. Moreover, we found that individuals with T2D or those in the high T2D PRS group had a significantly increased mortality rate. We also observed that the mortality rate for SARS-CoV-2-infected patients was approximately 2 to 7 times higher than for those not infected, depending on the time of infection. These findings emphasize the potential of T2D PRS in estimating the severity and mortality of COVID-19.

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.

Genetically determined telomere length and its association with chronic obstructive pulmonary disease and interstitial lung disease in biobank Japan: A Mendelian randomization study

Importance

Chronic obstructive pulmonary disease (COPD) and interstitial lung disease (ILD) have been linked to shorter telomere length (TL). While understanding this association has critical clinical implications for respiratory diseases, previous studies exploring these associations were conducted in European populations. The present study aims to investigate this relationship in an Asian population.

Objective

To examine the causal relationship between leukocyte TL and COPD and ILD in an Asian population.

Design

Setting, and Participants: We used a genome-wide association study summary statistics-based two-sample Mendelian randomization (MR) design to investigate the association between leukocyte TL, genetically predicted by nine single-nucleotide polymorphisms and the risk of COPD and ILD. Participants were Japanese individuals enrolled in the Biobank Japan Project, including 3315 COPD patients and 806 ILD patients.

Exposure

Leukocyte TL was genetically predicted by nine single-nucleotide polymorphisms.

Results

The inverse-variance weighted estimates showed a significant inverse association between leukocyte TL and COPD (odds ratio [OR] = 0.78; 95 % confidence interval [CI]: 0.64, 0.95; P = 0.01) and ILD (OR = 0.29; 95 % CI: 0.14, 0.61; P = 0.001), respectively. All sensitivity analyses yielded consistent results. The MR-Egger regression intercept test showed no evidence of horizontal pleiotropy (Pintercept: COPD, 0.56; ILD: 0.70).

Conclusion

and Relevance: Our findings suggest that leukocyte telomere shortening may causally increase the risk of COPD and ILD. These results highlight the potential importance of TL for these respiratory diseases.

Genomic risk prediction of cardiovascular diseases among type 2 diabetes patients in the UK Biobank

Background: Polygenic risk score (PRS) has proved useful in predicting the risk of cardiovascular diseases (CVD) based on the genotypes of an individual, but most analyses have focused on disease onset in the general population. The usefulness of PRS to predict CVD risk among type 2 diabetes (T2D) patients remains unclear. Methods: We built a meta-PRSCVD upon the candidate PRSs developed from state-of-the-art PRS methods for three CVD subtypes of significant importance: coronary artery disease (CAD), ischemic stroke (IS), and heart failure (HF). To evaluate the prediction performance of the meta-PRSCVD, we restricted our analysis to 21,092 white British T2D patients in the UK Biobank, among which 4,015 had CVD events. Results: Results showed that the meta-PRSCVD was significantly associated with CVD risk with a hazard ratio per standard deviation increase of 1.28 (95% CI: 1.23-1.33). The meta-PRSCVD alone predicted the CVD incidence with an area under the receiver operating characteristic curve (AUC) of 0.57 (95% CI: 0.54-0.59). When restricted to the early-onset patients (onset age ≤ 55), the AUC was further increased to 0.61 (95% CI 0.56-0.67). Conclusion: Our results highlight the potential role of genomic screening for secondary preventions of CVD among T2D patients, especially among early-onset patients.

Tuning parameters for polygenic risk score methods using GWAS summary statistics from training data

Various polygenic risk scores (PRS) methods have been proposed to combine the estimated effects of single nucleotide polymorphisms (SNPs) to predict genetic risks for common diseases, using data collected from genome-wide association studies (GWAS). Some methods require external individual-level GWAS dataset for parameter tuning, posing privacy and security-related concerns. Leaving out partial data for parameter tuning can also reduce model prediction accuracy. In this article, we propose PRStuning, a method that tunes parameters for different PRS methods using GWAS summary statistics from the training data. PRStuning predicts the PRS performance with different parameters, and then selects the best-performing parameters. Because directly using training data effects tends to overestimate the performance in the testing data, we adopt an empirical Bayes approach to shrinking the predicted performance in accordance with the genetic architecture of the disease. Extensive simulations and real data applications demonstrate PRStuning's accuracy across PRS methods and parameters.

Improving fine-mapping by modeling infinitesimal effects

Fine-mapping aims to identify causal genetic variants for phenotypes. Bayesian fine-mapping algorithms (for example, SuSiE, FINEMAP, ABF and COJO-ABF) are widely used, but assessing posterior probability calibration remains challenging in real data, where model misspecification probably exists, and true causal variants are unknown. We introduce replication failure rate (RFR), a metric to assess fine-mapping consistency by downsampling. SuSiE, FINEMAP and COJO-ABF show high RFR, indicating potential overconfidence in their output. Simulations reveal that nonsparse genetic architecture can lead to miscalibration, while imputation noise, nonuniform distribution of causal variants and quality control filters have minimal impact. Here we present SuSiE-inf and FINEMAP-inf, fine-mapping methods modeling infinitesimal effects alongside fewer larger causal effects. Our methods show improved calibration, RFR and functional enrichment, competitive recall and computational efficiency. Notably, using our methods' posterior effect sizes substantially increases polygenic risk score accuracy over SuSiE and FINEMAP. Our work improves causal variant identification for complex traits, a fundamental goal of human genetics.

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.

The accuracy of polygenic score models for anthropometric traits and Type II Diabetes in the Native Hawaiian Population

Polygenic scores (PGS) are promising in stratifying individuals based on the genetic susceptibility to complex diseases or traits. However, the accuracy of PGS models, typically trained in European- or East Asian-ancestry populations, tend to perform poorly in other ethnic minority populations, and their accuracies have not been evaluated for Native Hawaiians. Using body mass index, height, and type-2 diabetes as examples of highly polygenic traits, we evaluated the prediction accuracies of PGS models in a large Native Hawaiian sample from the Multiethnic Cohort with up to 5,300 individuals. We evaluated both publicly available PGS models or genome-wide PGS models trained in this study using the largest available GWAS. We found evidence of lowered prediction accuracies for the PGS models in some cases, particularly for height. We also found that using the Native Hawaiian samples as an optimization cohort during training did not consistently improve PGS performance. Moreover, even the best performing PGS models among Native Hawaiians would have lowered prediction accuracy among the subset of individuals most enriched with Polynesian ancestry. Our findings indicate that factors such as admixture histories, sample size and diversity in GWAS can influence PGS performance for complex traits among Native Hawaiian samples. This study provides an initial survey of PGS performance among Native Hawaiians and exposes the current gaps and challenges associated with improving polygenic prediction models for underrepresented minority populations.

Efficient estimation for large-scale linkage disequilibrium patterns of the human genome

In this study, we proposed an efficient algorithm (X-LD) for estimating linkage disequilibrium (LD) patterns for a genomic grid, which can be of inter-chromosomal scale or of small segments. Compared with conventional methods, the proposed method was significantly faster, dropped from O(nm2) to O(n2m)-n the sample size and m the number of SNPs, and consequently we were permitted to explore in depth unknown or reveal long-anticipated LD features of the human genome. Having applied the algorithm for 1000 Genome Project (1KG), we found (1) the extended LD, driven by population structure, universally existed, and the strength of inter-chromosomal LD was about 10% of their respective intra-chromosomal LD in relatively homogeneous cohorts, such as FIN, and to nearly 56% in admixed cohort, such as ASW. (2) After splitting each chromosome into upmost of more than a half million grids, we elucidated the LD of the HLA region was nearly 42 folders higher than chromosome 6 in CEU and 11.58 in ASW; on chromosome 11, we observed that the LD of its centromere was nearly 94.05 folders higher than chromosome 11 in YRI and 42.73 in ASW. (3) We uncovered the long-anticipated inversely proportional linear relationship between the length of a chromosome and the strength of chromosomal LD, and their Pearson's correlation was on average over 0.80 for 26 1KG cohorts. However, this linear norm was so far perturbed by chromosome 11 given its more completely sequenced centromere region. Uniquely chromosome 8 of ASW was found most deviated from the linear norm than any other autosomes. The proposed algorithm has been realized in C++ (called X-LD) and is available at https://github.com/gc5k/gear2, and can be applied to explore LD features in any sequenced populations.

Longitudinal multi-omics study reveals common etiology underlying association between plasma proteome and BMI trajectories in adolescent and young adult twins

BACKGROUND

The influence of genetics and environment on the association of the plasma proteome with body mass index (BMI) and changes in BMI remains underexplored, and the links to other omics in these associations remain to be investigated. We characterized protein-BMI trajectory associations in adolescents and adults and how these connect to other omics layers.

METHODS

Our study included two cohorts of longitudinally followed twins: FinnTwin12 (N = 651) and the Netherlands Twin Register (NTR) (N = 665). Follow-up comprised 4 BMI measurements over approximately 6 (NTR: 23-27 years old) to 10 years (FinnTwin12: 12-22 years old), with omics data collected at the last BMI measurement. BMI changes were calculated in latent growth curve models. Mixed-effects models were used to quantify the associations between the abundance of 439 plasma proteins with BMI at blood sampling and changes in BMI. In FinnTwin12, the sources of genetic and environmental variation underlying the protein abundances were quantified by twin models, as were the associations of proteins with BMI and BMI changes. In NTR, we investigated the association of gene expression of genes encoding proteins identified in FinnTwin12 with BMI and changes in BMI. We linked identified proteins and their coding genes to plasma metabolites and polygenic risk scores (PRS) applying mixed-effects models and correlation networks.

RESULTS

We identified 66 and 14 proteins associated with BMI at blood sampling and changes in BMI, respectively. The average heritability of these proteins was 35%. Of the 66 BMI-protein associations, 43 and 12 showed genetic and environmental correlations, respectively, including 8 proteins showing both. Similarly, we observed 7 and 3 genetic and environmental correlations between changes in BMI and protein abundance, respectively. S100A8 gene expression was associated with BMI at blood sampling, and the PRG4 and CFI genes were associated with BMI changes. Proteins showed strong connections with metabolites and PRSs, but we observed no multi-omics connections among gene expression and other omics layers.

CONCLUSIONS

Associations between the proteome and BMI trajectories are characterized by shared genetic, environmental, and metabolic etiologies. We observed few gene-protein pairs associated with BMI or changes in BMI at the proteome and transcriptome levels.

Multimodal Data Integration Advances Longitudinal Prediction of the Naturalistic Course of Depression and Reveals a Multimodal Signature of Remission During 2-Year Follow-up

BACKGROUND

The ability to predict the disease course of individuals with major depressive disorder (MDD) is essential for optimal treatment planning. Here, we used a data-driven machine learning approach to assess the predictive value of different sets of biological data (whole-blood proteomics, lipid metabolomics, transcriptomics, genetics), both separately and added to clinical baseline variables, for the longitudinal prediction of 2-year remission status in MDD at the individual-subject level.

METHODS

Prediction models were trained and cross-validated in a sample of 643 patients with current MDD (2-year remission n = 325) and subsequently tested for performance in 161 individuals with MDD (2-year remission n = 82).

RESULTS

Proteomics data showed the best unimodal data predictions (area under the receiver operating characteristic curve = 0.68). Adding proteomic to clinical data at baseline significantly improved 2-year MDD remission predictions (area under the receiver operating characteristic curve = 0.63 vs. 0.78, p = .013), while the addition of other omics data to clinical data did not yield significantly improved model performance. Feature importance and enrichment analysis revealed that proteomic analytes were involved in inflammatory response and lipid metabolism, with fibrinogen levels showing the highest variable importance, followed by symptom severity. Machine learning models outperformed psychiatrists' ability to predict 2-year remission status (balanced accuracy = 71% vs. 55%).

CONCLUSIONS

This study showed the added predictive value of combining proteomic data, but not other omics data, with clinical data for the prediction of 2-year remission status in MDD. Our results reveal a novel multimodal signature of 2-year MDD remission status that shows clinical potential for individual MDD disease course predictions from baseline measurements.

Prediction of Parkinson's Disease Using Machine Learning Methods

The detection of Parkinson's disease (PD) in its early stages is of great importance for its treatment and management, but consensus is lacking on what information is necessary and what models should be used to best predict PD risk. In our study, we first grouped PD-associated factors based on their cost and accessibility, and then gradually incorporated them into risk predictions, which were built using eight commonly used machine learning models to allow for comprehensive assessment. Finally, the Shapley Additive Explanations (SHAP) method was used to investigate the contributions of each factor. We found that models built with demographic variables, hospital admission examinations, clinical assessment, and polygenic risk score achieved the best prediction performance, and the inclusion of invasive biomarkers could not further enhance its accuracy. Among the eight machine learning models considered, penalized logistic regression and XGBoost were the most accurate algorithms for assessing PD risk, with penalized logistic regression achieving an area under the curve of 0.94 and a Brier score of 0.08. Olfactory function and polygenic risk scores were the most important predictors for PD risk. Our research has offered a practical framework for PD risk assessment, where necessary information and efficient machine learning tools were highlighted.

Accurate and efficient estimation of local heritability using summary statistics and the linkage disequilibrium matrix

Existing SNP-heritability estimators that leverage summary statistics from genome-wide association studies (GWAS) are much less efficient (i.e., have larger standard errors) than the restricted maximum likelihood (REML) estimators which require access to individual-level data. We introduce a new method for local heritability estimation-Heritability Estimation with high Efficiency using LD and association Summary Statistics (HEELS)-that significantly improves the statistical efficiency of summary-statistics-based heritability estimator and attains comparable statistical efficiency as REML (with a relative statistical efficiency >92%). Moreover, we propose representing the empirical LD matrix as the sum of a low-rank matrix and a banded matrix. We show that this way of modeling the LD can not only reduce the storage and memory cost, but also improve the computational efficiency of heritability estimation. We demonstrate the statistical efficiency of HEELS and the advantages of our proposed LD approximation strategies both in simulations and through empirical analyses of the UK Biobank data.

Integration of Biomarker Polygenic Risk Score Improves Prediction of Coronary Heart Disease

There are several established biomarkers for coronary heart disease (CHD), including blood pressure, cholesterol, and lipoproteins. It is of high interest to determine how a combined polygenic risk score (PRS) of CHD-associated biomarkers (BioPRS) can further improve genetic prediction of CHD. We developed CHDBioPRS, combining BioPRS with PRS of CHD in the UK Biobank and tested it on FinnGen. We found that BioPRS was clearly predictive of CHD and that CHDBioPRS improved the standard CHD PRS. The largest effect was observed with early onset cases in FinnGen, with HRs above 2 per standard deviation of CHDBioPRS.

The long-term effects of early-life pollution exposure: Evidence from the London smog

This paper uses a large UK cohort to investigate the impact of early-life pollution exposure on individuals' human capital and health outcomes in older age. We compare individuals who were exposed to the London smog in December 1952 whilst in utero or in infancy to those born after the smog and those born at the same time but in unaffected areas. We find that those exposed to the smog have substantially lower fluid intelligence and worse respiratory health, with some evidence of a reduction in years of schooling.

Polygenic Risk Score for Cardiovascular Diseases in Artificial Intelligence Paradigm: A Review

Cardiovascular disease (CVD) related mortality and morbidity heavily strain society. The relationship between external risk factors and our genetics have not been well established. It is widely acknowledged that environmental influence and individual behaviours play a significant role in CVD vulnerability, leading to the development of polygenic risk scores (PRS). We employed the PRISMA search method to locate pertinent research and literature to extensively review artificial intelligence (AI)-based PRS models for CVD risk prediction. Furthermore, we analyzed and compared conventional vs. AI-based solutions for PRS. We summarized the recent advances in our understanding of the use of AI-based PRS for risk prediction of CVD. Our study proposes three hypotheses: i) Multiple genetic variations and risk factors can be incorporated into AI-based PRS to improve the accuracy of CVD risk predicting. ii) AI-based PRS for CVD circumvents the drawbacks of conventional PRS calculators by incorporating a larger variety of genetic and non-genetic components, allowing for more precise and individualised risk estimations. iii) Using AI approaches, it is possible to significantly reduce the dimensionality of huge genomic datasets, resulting in more accurate and effective disease risk prediction models. Our study highlighted that the AI-PRS model outperformed traditional PRS calculators in predicting CVD risk. Furthermore, using AI-based methods to calculate PRS may increase the precision of risk predictions for CVD and have significant ramifications for individualized prevention and treatment plans.

Applying polygenic risk score methods to pharmacogenomics GWAS: challenges and opportunities

Polygenic risk scores (PRSs) have emerged as promising tools for the prediction of human diseases and complex traits in disease genome-wide association studies (GWAS). Applying PRSs to pharmacogenomics (PGx) studies has begun to show great potential for improving patient stratification and drug response prediction. However, there are unique challenges that arise when applying PRSs to PGx GWAS beyond those typically encountered in disease GWAS (e.g. Eurocentric or trans-ethnic bias). These challenges include: (i) the lack of knowledge about whether PGx or disease GWAS/variants should be used in the base cohort (BC); (ii) the small sample sizes in PGx GWAS with corresponding low power and (iii) the more complex PRS statistical modeling required for handling both prognostic and predictive effects simultaneously. To gain insights in this landscape about the general trends, challenges and possible solutions, we first conduct a systematic review of both PRS applications and PRS method development in PGx GWAS. To further address the challenges, we propose (i) a novel PRS application strategy by leveraging both PGx and disease GWAS summary statistics in the BC for PRS construction and (ii) a new Bayesian method (PRS-PGx-Bayesx) to reduce Eurocentric or cross-population PRS prediction bias. Extensive simulations are conducted to demonstrate their advantages over existing PRS methods applied in PGx GWAS. Our systematic review and methodology research work not only highlights current gaps and key considerations while applying PRS methods to PGx GWAS, but also provides possible solutions for better PGx PRS applications and future research.

Evaluation of optimal methods and ancestries for calculating polygenic risk scores in East Asian population

Polygenic risk scores (PRSs) have been studied for predicting human diseases, and various methods for PRS calculation have been developed. Most PRS studies to date have focused on European ancestry, and the performance of PRS has not been sufficiently assessed in East Asia. Herein, we evaluated the predictive performance of PRSs for East Asian populations under various conditions. Simulation studies using data from the Korean cohort, Health Examinees (HEXA), demonstrated that SBayesRC and PRS-CS outperformed other PRS methods (lassosum, LDpred-funct, and PRSice) in high fixed heritability (0.3 and 0.7). In addition, we generated PRSs using real-world data from HEXA for ten diseases: asthma, breast cancer, cataract, coronary artery disease, gastric cancer, glaucoma, hyperthyroidism, hypothyroidism, osteoporosis, and type 2 diabetes (T2D). We utilized the five previous PRS methods and genome-wide association study (GWAS) data from two biobank-scale datasets [European (UK Biobank) and East Asian (BioBank Japan) ancestry]. Additionally, we employed PRS-CSx, a PRS method that combines GWAS data from both ancestries, to generate a total of 110 PRS for ten diseases. Similar to the simulation results, SBayesRC showed better predictive performance for disease risk than the other methods. Furthermore, the East Asian GWAS data outperformed those from European ancestry for breast cancer, cataract, gastric cancer, and T2D, but neither of the two GWAS ancestries showed a significant advantage on PRS performance for the remaining six diseases. Based on simulation data and real data studies, it is expected that SBayesRC will offer superior performance for East Asian populations, and PRS generated using GWAS from non-East Asian may also yield good results.

Factorizing polygenic epistasis improves prediction and uncovers biological pathways in complex traits

Epistasis is central in many domains of biology, but it has not yet been proven useful for understanding the etiology of complex traits. This is partly because complex-trait epistasis involves polygenic interactions that are poorly captured in current models. To address this gap, we developed a model called Epistasis Factor Analysis (EFA). EFA assumes that polygenic epistasis can be factorized into interactions between a few epistasis factors (EFs), which represent latent polygenic components of the observed complex trait. The statistical goals of EFA are to improve polygenic prediction and to increase power to detect epistasis, while the biological goal is to unravel genetic effects into more-homogeneous units. We mathematically characterize EFA and use simulations to show that EFA outperforms current epistasis models when its assumptions approximately hold. Applied to predicting yeast growth rates, EFA outperforms the additive model for several traits with large epistasis heritability and uniformly outperforms the standard epistasis model. We replicate these prediction improvements in a second dataset. We then apply EFA to four previously characterized traits in the UK Biobank and find statistically significant epistasis in all four, including two that are robust to scale transformation. Moreover, we find that the inferred EFs partly recover pre-defined biological pathways for two of the traits. Our results demonstrate that more realistic models can identify biologically and statistically meaningful epistasis in complex traits, indicating that epistasis has potential for precision medicine and characterizing the biology underlying GWAS results.

Expanded utility of the R package, qgg, with applications within genomic medicine

SUMMARY

Here, we present an expanded utility of the R package qgg for genetic analyses of complex traits and diseases. One of the major updates of the package is, that it now includes Bayesian linear regression modeling procedures, which provide a unified framework for mapping of genetic variants, estimation of heritability and genomic prediction from either individual level data or from genome-wide association study summary data. With this release, the qgg package now provides a wealth of the commonly used methods in analysis of complex traits and diseases, without the need to switch between software and data formats.

AVAILABILITY AND IMPLEMENTATION

The methodologies are implemented in the publicly available R software package, qgg, using fast and memory efficient algorithms in C++ and is available on CRAN or as a developer version at our GitHub page (https://github.com/psoerensen/qgg). Notes on the implemented statistical genetic models, tutorials and example scripts are available at our GitHub page https://psoerensen.github.io/qgg/.

Disentangling the interplay between genes, cognitive skills, and educational level in adolescent and young adult smoking - The TRAILS study

Recent studies suggest that smoking and lower educational attainment may have genetic influences in common. However, little is known about the mechanisms through which genetics contributes to educational inequalities in adolescent and young adult smoking. Common genetic liabilities may underlie cognitive skills associated with both smoking and education, such as IQ and effortful control, in line with indirect health-related selection explanations. Additionally, by affecting cognitive skills, genes may predict educational trajectories and hereby adolescents' social context, which may be associated with smoking, consistent with social causation explanations. Using data from the Dutch TRAILS Study (N = 1581), we estimated the extent to which polygenic scores (PGSs) for ever smoking regularly (PGSSMOK) and years of education (PGSEDU) predict IQ and effortful control, measured around age 11, and whether these cognitive skills then act as shared predictors of smoking and educational level around age 16, 19, 22, and 26. Second, we assessed if educational level mediated associations between PGSs and smoking. Both PGSs were associated with lower effortful control, and PGSEDU also with lower IQ. Lower IQ and effortful control, in turn, predicted having a lower educational level. However, neither of these cognitive skills were directly associated with smoking behaviour after controlling for covariates and PGSs. This suggests that IQ and effortful control are not shared predictors of smoking and education (i.e., no indirect health-related selection related to cognitive skills). Instead, PGSSMOK and PGSEDU, partly through their associations with lower cognitive skills, predicted selection into a lower educational track, which in turn was associated with more smoking, in line with social causation explanations. Our findings suggest that educational differences in the social context contribute to associations between genetic liabilities and educational inequalities in smoking.

Transfer learning with false negative control improves polygenic risk prediction

Polygenic risk score (PRS) is a quantity that aggregates the effects of variants across the genome and estimates an individual's genetic predisposition for a given trait. PRS analysis typically contains two input data sets: base data for effect size estimation and target data for individual-level prediction. Given the availability of large-scale base data, it becomes more common that the ancestral background of base and target data do not perfectly match. In this paper, we treat the GWAS summary information obtained in the base data as knowledge learned from a pre-trained model, and adopt a transfer learning framework to effectively leverage the knowledge learned from the base data that may or may not have similar ancestral background as the target samples to build prediction models for target individuals. Our proposed transfer learning framework consists of two main steps: (1) conducting false negative control (FNC) marginal screening to extract useful knowledge from the base data; and (2) performing joint model training to integrate the knowledge extracted from base data with the target training data for accurate trans-data prediction. This new approach can significantly enhance the computational and statistical efficiency of joint-model training, alleviate over-fitting, and facilitate more accurate trans-data prediction when heterogeneity level between target and base data sets is small or high.

Pragmatic Approach to Applying Polygenic Risk Scores to Diverse Populations

Polygenic risk scores (PRS) estimate genetic susceptibility of an individual to disease and have the potential of providing utility in multiple clinical contexts. However, their performance, computation, and reporting in diverse populations remain challenging. Here, we present a pragmatic approach to optimize a PRS for a population of interest that leverages publicly available data and methods and consists of seven steps that are easily implemented without the requirement of expertise in complex genetics: step 1, selecting source genome-wide association studies (GWAS) and imputation; step 2, selecting methods to compute polygenic score; step 3, adjusting scores using principal components of genetic ancestry; step 4, selecting the best performing score; step 5, defining percentiles of a population distribution; step 6, validating performance of the optimized polygenic score; and step 7, implementing the optimized polygenic score in clinical practice. © 2023 Wiley Periodicals LLC.

Genetic contributions of noncognitive skills to academic development

Noncognitive skills such as motivation and self-regulation, are partly heritable and predict academic achievement beyond cognitive skills. However, how the relationship between noncognitive skills and academic achievement changes over development is unclear. The current study examined how cognitive and noncognitive skills contribute to academic achievement from ages 7 to 16 in a sample of over 10,000 children from England and Wales. Noncognitive skills were increasingly predictive of academic achievement across development. Twin and polygenic scores analyses found that the contribution of noncognitive genetics to academic achievement became stronger over the school years. Results from within-family analyses indicated that associations with noncognitive genetics could not simply be attributed to confounding by environmental differences between nuclear families and are consistent with a possible role for evocative/active gene-environment correlations. By studying genetic effects through a developmental lens, we provide novel insights into the role of noncognitive skills in academic development.

Minimal improvement in coronary artery disease risk prediction in Chinese population using polygenic risk scores: evidence from the China Kadoorie Biobank

BACKGROUND

Several studies have reported that polygenic risk scores (PRSs) can enhance risk prediction of coronary artery disease (CAD) in European populations. However, research on this topic is far from sufficient in non-European countries, including China. We aimed to evaluate the potential of PRS for predicting CAD for primary prevention in the Chinese population.

METHODS

Participants with genome-wide genotypic data from the China Kadoorie Biobank were divided into training ( n = 28,490) and testing sets ( n = 72,150). Ten previously developed PRSs were evaluated, and new ones were developed using clumping and thresholding or LDpred method. The PRS showing the strongest association with CAD in the training set was selected to further evaluate its effects on improving the traditional CAD risk-prediction model in the testing set. Genetic risk was computed by summing the product of the weights and allele dosages across genome-wide single-nucleotide polymorphisms. Prediction of the 10-year first CAD events was assessed using hazard ratios (HRs) and measures of model discrimination, calibration, and net reclassification improvement (NRI). Hard CAD (nonfatal I21-I23 and fatal I20-I25) and soft CAD (all fatal or nonfatal I20-I25) were analyzed separately.

RESULTS

In the testing set, 1214 hard and 7201 soft CAD cases were documented during a mean follow-up of 11.2 years. The HR per standard deviation of the optimal PRS was 1.26 (95% CI:1.19-1.33) for hard CAD. Based on a traditional CAD risk prediction model containing only non-laboratory-based information, the addition of PRS for hard CAD increased Harrell's C index by 0.001 (-0.001 to 0.003) in women and 0.003 (0.001 to 0.005) in men. Among the different high-risk thresholds ranging from 1% to 10%, the highest categorical NRI was 3.2% (95% CI: 0.4-6.0%) at a high-risk threshold of 10.0% in women. The association of the PRS with soft CAD was much weaker than with hard CAD, leading to minimal or no improvement in the soft CAD model.

CONCLUSIONS

In this Chinese population sample, the current PRSs minimally changed risk discrimination and offered little improvement in risk stratification for soft CAD. Therefore, this may not be suitable for promoting genetic screening in the general Chinese population to improve CAD risk prediction.

Profiling the inflammatory bowel diseases using genetics, serum biomarkers, and smoking information

Crohn's disease (CD) and ulcerative colitis (UC) are two etiologically related yet distinctive subtypes of the inflammatory bowel diseases (IBD). Differentiating CD from UC can be challenging using conventional clinical approaches in a subset of patients. We designed and evaluated a novel molecular-based prediction model aggregating genetics, serum biomarkers, and tobacco smoking information to assist the diagnosis of CD and UC in over 30,000 samples. A joint model combining genetics, serum biomarkers and smoking explains 46% (42-50%, 95% CI) of phenotypic variation. Despite modest overlaps with serum biomarkers, genetics makes unique contributions to distinguishing IBD subtypes. Smoking status only explains 1% (0-6%, 95% CI) of the phenotypic variance suggesting it may not be an effective biomarker. This study reveals that molecular-based models combining genetics, serum biomarkers, and smoking information could complement current diagnostic strategies and help classify patients based on biologic state rather than imperfect clinical parameters.

Consistent effects of the genetics of happiness across the lifespan and ancestries in multiple cohorts

Happiness is a fundamental human affective trait, but its biological basis is not well understood. Using a novel approach, we construct LDpred-inf polygenic scores of a general happiness measure in 2 cohorts: the Adolescent Brain Cognitive Development (ABCD) cohort (N = 15,924, age range 9.23-11.8 years), the Add Health cohort (N = 9129, age range 24.5-34.7) to determine associations with several well-being and happiness measures. Additionally, we investigated associations between genetic scores for happiness and brain structure in ABCD (N = 9626, age range (8.9-11) and UK Biobank (N = 16,957, age range 45-83). We detected significant (p.FDR < 0.05) associations between higher genetic scores vs. several well-being measures (best r2 = 0.019) in children of multiple ancestries in ABCD and small yet significant correlations with a happiness measure in European participants in Add Health (r2 = 0.004). Additionally, we show significant associations between lower genetic scores for happiness with smaller structural brain phenotypes in a white British subsample of UK Biobank and a white sub-sample group of ABCD. We demonstrate that the genetic basis for general happiness level appears to have a consistent effect on happiness and wellbeing measures throughout the lifespan, across multiple ancestral backgrounds, and multiple brain structures.

Evaluating genomic polygenic risk scores for childhood acute lymphoblastic leukemia in Latinos

The utility of polygenic risk score (PRS) models has not been comprehensively evaluated for childhood acute lymphoblastic leukemia (ALL), the most common type of cancer in children. Previous PRS models for ALL were based on significant loci observed in genome-wide association studies (GWASs), even though genomic PRS models have been shown to improve prediction performance for a number of complex diseases. In the United States, Latino (LAT) children have the highest risk of ALL, but the transferability of PRS models to LAT children has not been studied. In this study, we constructed and evaluated genomic PRS models based on either non-Latino White (NLW) GWAS or a multi-ancestry GWAS. We found that the best PRS models performed similarly between held-out NLW and LAT samples (PseudoR2 = 0.086 ± 0.023 in NLW vs. 0.060 ± 0.020 in LAT), and can be improved for LAT if we performed GWAS in LAT-only (PseudoR2 = 0.116 ± 0.026) or multi-ancestry samples (PseudoR2 = 0.131 ± 0.025). However, the best genomic models currently do not have better prediction accuracy than a conventional model using all known ALL-associated loci in the literature (PseudoR2 = 0.166 ± 0.025), which includes loci from GWAS populations that we could not access to train genomic PRS models. Our results suggest that larger and more inclusive GWASs may be needed for genomic PRS to be useful for ALL. Moreover, the comparable performance between populations may suggest a more oligogenic architecture for ALL, where some large effect loci may be shared between populations. Future PRS models that move away from the infinite causal loci assumption may further improve PRS for ALL.