Improving polygenic risk prediction in admixed populations by explicitly modeling ancestral-differential effects via GAUDI

Diversifying Psychiatric Genomics: Globally Inclusive Strategies Toward Health Equity

The underrepresentation of non-European researchers, participants, and datasets in psychiatric genetics hinders the understanding of mental health conditions and perpetuates health inequities. Ancestral diversity in research is crucial for advancing insights into disease etiology and achieving equity in precision medicine. Key strategies include optimizing data use, fostering global collaboration for capacity building, and adopting best practices in research methods. Ensuring clinical impact, accountability, and multi-agency commitment is vital. A more inclusive approach will enhance understanding of genetic and environmental factors in mental health, leading to equitable and accessible health care outcomes for all populations.

Tracing human genetic histories and natural selection with precise local ancestry inference

Local ancestry inference is crucial for unraveling demographic histories, discovering selection signals, and including admixed individuals in genomic studies for improved equity and portability. To date, the precision and resolution of local ancestry inference were limited by technical and dataset issues. To address them, we present Orchestra, a model we train on over 10,000 single-origin individuals from 35 worldwide populations that demonstrates superior accuracy in benchmarking analyzes. We employ Orchestra to shed light on the demographic history of Latin Americans, finding trace ancestries supported by historical records. We then deploy it to offer insight on the debated Ashkenazi Jewish origins, highlighting their South European heritage. Finally, Orchestra enables us to map selection signatures, identifying trace Scandinavian ancestry in British samples and unveiling an immune-rich region linked to respiratory infections passed down from the Viking conquests. Our work significantly advances the field of local ancestry inference, highlighting its use in admixed populations.

Methodological opportunities in genomic data analysis to advance health equity

The causes and consequences of inequities in genomic research and medicine are complex and widespread. However, it is widely acknowledged that underrepresentation of diverse populations in human genetics research risks exacerbating existing health disparities. Efforts to improve diversity are ongoing, but an often-overlooked source of inequity is the choice of analytical methods used to process, analyse and interpret genomic data. This choice can influence all areas of genomic research, from genome-wide association studies and polygenic score development to variant prioritization and functional genomics. New statistical and machine learning techniques to understand, quantify and correct for the impact of biases in genomic data are emerging within the wider genomic research and genomic medicine ecosystems. At this crucial time point, it is important to clarify where improvements in methods and practices can, or cannot, have a role in improving equity in genomics. Here, we review existing approaches to promote equity and fairness in statistical analysis for genomics, and propose future methodological developments that are likely to yield the most impact for equity.

Opportunities and challenges of local ancestry in genetic association analyses

Recently, admixed populations make up an increasing percentage of the US and global populations, and the admixture is not uniform over space or time or across genomes. Therefore, it becomes indispensable to evaluate local ancestry in addition to global ancestry to improve genetic epidemiological studies. Recent advances in representing human genome diversity, coupled with large-scale whole-genome sequencing initiatives and improved tools for local ancestry inference, have enabled studies to demonstrate that incorporating local ancestry information enhances both genetic association analyses and polygenic risk predictions. Along with the opportunities that local ancestry provides, there exist challenges preventing its full usage in genetic analyses. In this review, we first summarize methods for local ancestry inference and illustrate how local ancestry can be utilized in various analyses, including admixture mapping, association testing, and polygenic risk score construction. In addition, we discuss current challenges in research involving local ancestry, both in terms of the inference itself and its role in genetic association studies. We further pinpoint some future study directions and methodology development opportunities to help more effectively incorporate local ancestry in genetic analyses. It is worth the effort to pursue those future directions and address these analytical challenges because the appropriate use of local ancestry estimates could help mitigate inequality in genomic medicine and improve our understanding of health and disease outcomes.

Psychiatric genetics in the diverse landscape of Latin American populations

Psychiatric disorders are highly heritable and polygenic, influenced by environmental factors and often comorbid. Large-scale genome-wide association studies (GWASs) through consortium efforts have identified genetic risk loci and revealed the underlying biology of psychiatric disorders and traits. However, over 85% of psychiatric GWAS participants are of European ancestry, limiting the applicability of these findings to non-European populations. Latin America and the Caribbean, regions marked by diverse genetic admixture, distinct environments and healthcare disparities, remain critically understudied in psychiatric genomics. This threatens access to precision psychiatry, where diversity is crucial for innovation and equity. This Review evaluates the current state and advancements in psychiatric genomics within Latin America and the Caribbean, discusses the prevalence and burden of psychiatric disorders, explores contributions to psychiatric GWASs from these regions and highlights methods that account for genetic diversity. We also identify existing gaps and challenges and propose recommendations to promote equity in psychiatric genomics.

Polygenic Scores of Cardiometabolic Risk Factors in American Indian Adults

Importance

Numerous efforts have been made to include diverse populations in genetic studies, but American Indian populations are still severely underrepresented. Polygenic scores derived from genetic data have been proposed in clinical care, but how polygenic scores perform in American Indian individuals and whether they can predict disease risk in this population remains unknown.

Objective

To study the performance of polygenic scores for cardiometabolic risk factors of lipid traits and C-reactive protein in American Indian adults and to determine whether such scores are helpful in clinical prediction for cardiometabolic diseases.

Design, Setting, and Participants

The Strong Heart Study (SHS) is a large American Indian cohort recruited from 1989 to 1991, with ongoing follow-up (phase VII). In this genetic association study, data from SHS American Indian participants were used in addition to data from 2 large-scale, external, ancestry-mismatched genome-wide association studies (GWASs; 450 865 individuals from a European GWAS and 33 096 individuals from a multi-ancestry GWAS) and 1 small-scale internal ancestry-matched American Indian GWAS (2000 individuals). Analyses were conducted from February 2023 to August 2024.

Exposure

Genetic risk score for cardiometabolic disease risk factors from 6 traits including 5 lipids (apolipoprotein A, apolipoprotein B, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, and triglycerides), and an inflammatory biomarker (C-reactive protein [CRP]).

Main Outcomes and Measures

Data from SHS participants and the 2 GWASs were used to construct 8 polygenic scores. The association of polygenic scores with cardiometabolic disease was assessed using 2-sided z tests and 1-sided likelihood ratio tests.

Results

In the 3157 SHS participants (mean [SD] age, 56.44 [8.12] years; 1845 female [58.4%]), a large European-based polygenic score had the most robust performance (mean [SD] R2 = 5.0% [1.7%]), but adding a small-scale ancestry-matched GWAS using American Indian data helped improve polygenic score prediction for 5 of 6 traits (all but CRP; mean [SD] R2, 7.6% [3.2%]). Lipid polygenic scores developed in American Indian individuals improved prediction of diabetes compared with baseline clinical risk factors (area under the curve for absolute improvement, 0.86%; 95% CI, 0.78%-0.93%; likelihood ratio test P = 3.8 × 10-3).

Conclusions and Relevance

In this genetic association study of lipids and CRP among American Indian individuals, polygenic scores of lipid traits were found to improve prediction of diabetes when added to clinical risk factors, although the magnitude of improvement was small. The transferability of polygenic scores derived from other populations is still a concern, with implications for the advancement of precision medicine and the potential of perpetuating health disparities, particularly in this underrepresented population.

Old vs. New Local Ancestry Inference in HCHS/SOL: A Comparative Study

Hispanic/Latino populations are admixed, with genetic contributions from multiple ancestral populations. Studies of genetic association in these admixed populations often use methods such as admixture mapping, which relies on inferred counts of "local" ancestry, i.e., of the source ancestral population at a locus. Local ancestries are inferred using external reference panels that represent ancestral populations, making the choice of inference method and reference panel critical. This study used a dataset of Hispanic/Latino individuals from the Hispanic Community Health Study/Study of Latinos (HCHS/SOL) to evaluate the "old" local ancestry inference performed using the state-of-the-art inference method, RFMix, alongside "new" inferences performed using Fast Local Ancestry Estimation (FLARE), which also used an updated reference panel. We compared their performance in terms of global and local ancestry correlations, as well as admixture mapping-based associations. Overall, the old RFMix and new FLARE inferences were highly similar for both global and local ancestries, with FLARE-inferred datasets yielding admixture mapping results consistent with those computed from RFMix. However, in some genomic regions the old and new local ancestries have relatively lower correlations (Pearson R < 0.9). Most of these genomic regions (86.42%) were mapped to either ENCODE blacklist regions, or to gene clusters, compared to 7.67% of randomly-matched regions with high correlations (Pearson R > 0.97) between old and new local ancestries.

Leveraging genetic ancestry continuum information to interpolate PRS for admixed populations

The relatively low representation of admixed populations in both discovery and fine-tuning individual-level datasets limits polygenic risk score (PRS) development and equitable clinical translation for admixed populations. Under the assumption that the most informative PRS weight for a homogeneous sample varies linearly in an ancestry continuum space, we introduce a Genetic Distance-assisted PRS Combination Pipeline for Diverse Genetic Ancestries (DiscoDivas) to interpolate a harmonized PRS for diverse, especially admixed, ancestries, leveraging multiple PRS weights fine-tuned within single-ancestry samples and genetic distance. DiscoDivas treats ancestry as a continuous variable and does not require shifting between different models when calculating PRS for different ancestries. We generated PRS with DiscoDivas and the current conventional method, i.e. fine-tuning multiple GWAS PRS using the matched or similar ancestry samples. DiscoDivas generated a harmonized PRS of the accuracy comparable to or higher than the conventional approach, with the greatest advantage exhibited in admixed individuals.

Integrating electronic health records and GWAS summary statistics to predict the progression of autoimmune diseases from preclinical stages

Autoimmune diseases often exhibit a preclinical stage before diagnosis. Electronic health record (EHR) based-biobanks contain genetic data and diagnostic information, which can identify preclinical individuals at risk for progression. Biobanks typically have small numbers of cases, which are not sufficient to construct accurate polygenic risk scores (PRS). Importantly, progression and case-control phenotypes may have shared genetic basis, which we can exploit to improve prediction accuracy. We propose a novel method Genetic Progression Score (GPS) that integrates biobank and case-control study to predict the disease progression risk. Via penalized regression, GPS incorporates PRS weights for case-control studies as prior and forces model parameters to be similar to the prior if the prior improves prediction accuracy. In simulations, GPS consistently yields better prediction accuracy than alternative strategies relying on biobank or case-control samples only and those combining biobank and case-control samples. The improvement is particularly evident when biobank sample is smaller or the genetic correlation is lower. We derive PRS for the progression from preclinical rheumatoid arthritis and systemic lupus erythematosus in the BioVU biobank and validate them in All of Us. For both diseases, GPS achieves the highest predictionR 2 and the resulting PRS yields the strongest correlation with progression prevalence.

Polygenic scores and their applications in kidney disease

Genome-wide association studies (GWAS) have uncovered thousands of risk variants that individually have small effects on the risk of human diseases, including chronic kidney disease, type 2 diabetes, heart diseases and inflammatory disorders, but cumulatively explain a substantial fraction of disease risk, underscoring the complexity and pervasive polygenicity of common disorders. This complexity poses unique challenges to the clinical translation of GWAS findings. Polygenic scores combine small effects of individual GWAS risk variants across the genome to improve personalized risk prediction. Several polygenic scores have now been developed that exhibit sufficiently large effects to be considered clinically actionable. However, their clinical use is limited by their partial transferability across ancestries and a lack of validated models that combine polygenic, monogenic, family history and clinical risk factors. Moreover, prospective studies are still needed to demonstrate the clinical utility and cost-effectiveness of polygenic scores in clinical practice. Here, we discuss evolving methods for developing polygenic scores, best practices for validating and reporting their performance, and the study designs that will empower their clinical implementation. We specifically focus on the polygenic scores relevant to nephrology and other chronic, complex diseases and review their key limitations, necessary refinements and potential clinical applications.

The PRIMED Consortium: Reducing disparities in polygenic risk assessment

By improving disease risk prediction, polygenic risk scores (PRSs) could have a significant impact on health promotion and disease prevention. Due to the historical oversampling of populations with European ancestry for genome-wide association studies, PRSs perform less well in other, understudied populations, leading to concerns that clinical use in their current forms could widen health care disparities. The PRIMED Consortium was established to develop methods to improve the performance of PRSs in global populations and individuals of diverse genetic ancestry. To this end, PRIMED is aggregating and harmonizing multiple phenotype and genotype datasets on AnVIL, an interoperable secure cloud-based platform, to perform individual- and summary-level analyses using population and statistical genetics approaches. Study sites, the coordinating center, and representatives from the NIH work alongside other NHGRI and global consortia to achieve these goals. PRIMED is also evaluating ethical and social implications of PRS implementation and investigating the joint modeling of social determinants of health and PRS in computing disease risk. The phenotypes of interest are primarily cardiometabolic diseases and cancer, the leading causes of death and disability worldwide. Early deliverables of the consortium include methods for data sharing on AnVIL, development of a common data model to harmonize phenotype and genotype data from cohort studies as well as electronic health records, adaptation of recent guidelines for population descriptors to global cohorts, and sharing of PRS methods/tools. As a multisite collaboration, PRIMED aims to foster equity in the development and use of polygenic risk assessment.

One score to rule them all: regularized ensemble polygenic risk prediction with GWAS summary statistics

Ensemble learning has been increasingly popular for boosting the predictive power of polygenic risk scores (PRS), with almost every recent multi-ancestry PRS approach employing ensemble learning as a final step. Existing ensemble approaches rely on individual-level data for model training, which severely limits their real-world applications, especially in non-European populations without sufficient genomic samples. Here, we introduce a statistical framework to construct regularized ensemble PRS, which allows us to combine a large number of candidate PRS models using only summary statistics from genome-wide association studies. We demonstrate its robust and substantial improvement over many existing PRS models in both within- and cross-ancestry applications. We believe this is truly "one score to rule them all" due to its capability to continuously combine newly developed PRS models with existing models to improve prediction performance, which makes it a universal approach that should always be employed in future PRS applications.

Methodologies underpinning polygenic risk scores estimation: a comprehensive overview

Polygenic risk scores (PRS) have emerged as a promising tool for predicting disease risk and treatment outcomes using genomic data. Thousands of genome-wide association studies (GWAS), primarily involving populations of European ancestry, have supported the development of PRS models. However, these models have not been adequately evaluated in non-European populations, raising concerns about their clinical validity and predictive power across diverse groups. Addressing this issue requires developing novel risk prediction frameworks that leverage genetic characteristics across diverse populations, considering host-microbiome interactions and a broad range of health measures. One of the key aspects in evaluating PRS is understanding the strengths and limitations of various methods for constructing them. In this review, we analyze strengths and limitations of different methods for constructing PRS, including traditional weighted approaches and new methods such as Bayesian and Frequentist penalized regression approaches. Finally, we summarize recent advances in PRS calculation methods development, and highlight key areas for future research, including development of models robust across diverse populations by underlining the complex interplay between genetic variants across diverse ancestral backgrounds in disease risk as well as treatment response prediction. PRS hold great promise for improving disease risk prediction and personalized medicine; therefore, their implementation must be guided by careful consideration of their limitations, biases, and ethical implications to ensure that they are used in a fair, equitable, and responsible manner.

SPLENDID incorporates continuous genetic ancestry in biobank-scale data to improve polygenic risk prediction across diverse populations

Polygenic risk scores are widely used in disease risk stratification, but their accuracy varies across diverse populations. Recent methods large-scale leverage multi-ancestry data to improve accuracy in under-represented populations but require labelling individuals by ancestry for prediction. This poses challenges for practical use, as clinical practices are typically not based on ancestry. We propose SPLENDID, a novel penalized regression framework for diverse biobank-scale data. Our method utilizes ancestry principal component interactions to model genetic ancestry as a continuum within a single prediction model for all ancestries, eliminating the need for discrete labels. In extensive simulations and analyses of 9 traits from the All of Us Research Program (N=224,364) and UK Biobank (N=340,140), SPLENDID significantly outperformed existing methods in prediction accuracy and model sparsity. By directly incorporating continuous genetic ancestry in model training, SPLENDID stands as a valuable tool for robust risk prediction across diverse populations and fairer clinical implementation.

Powerful mapping of cis-genetic effects on gene expression across diverse populations reveals novel disease-critical genes

While disease-associated variants identified by genome-wide association studies (GWAS) most likely regulate gene expression levels, linking variants to target genes is critical to determining the functional mechanisms of these variants. Genetic effects on gene expression have been extensively characterized by expression quantitative trait loci (eQTL) studies, yet data from non-European populations is limited. This restricts our understanding of disease to genes whose regulatory variants are common in European populations. While previous work has leveraged data from multiple populations to improve GWAS power and polygenic risk score (PRS) accuracy, multi-ancestry data has not yet been used to better estimate cis-genetic effects on gene expression. Here, we present a new method, Multi-Ancestry Gene Expression Prediction Regularized Optimization (MAGEPRO), which constructs robust genetic models of gene expression in understudied populations or cell types by fitting a regularized linear combination of eQTL summary data across diverse cohorts. In simulations, our tool generates more accurate models of gene expression than widely-used LASSO and the state-of-the-art multi-ancestry PRS method, PRS-CSx, adapted to gene expression prediction. We attribute this improvement to MAGEPRO's ability to more accurately estimate causal eQTL effect sizes (p < 3.98 × 10-4, two-sided paired t-test). With real data, we applied MAGEPRO to 8 eQTL cohorts representing 3 ancestries (average n = 355) and consistently outperformed each of 6 competing methods in gene expression prediction tasks. Integration with GWAS summary statistics across 66 complex traits (representing 22 phenotypes and 3 ancestries) resulted in 2,331 new gene-trait associations, many of which replicate across multiple ancestries, including PHTF1 linked to white blood cell count, a gene which is overexpressed in leukemia patients. MAGEPRO also identified biologically plausible novel findings, such as PIGB, an essential component of GPI biosynthesis, associated with heart failure, which has been previously evidenced by clinical outcome data. Overall, MAGEPRO is a powerful tool to enhance inference of gene regulatory effects in underpowered datasets and has improved our understanding of population-specific and shared genetic effects on complex traits.

Gene expression and splicing QTL analysis of blood cells in African American participants from the Jackson Heart Study

Most gene expression and alternative splicing quantitative trait loci (eQTL/sQTL) studies have been biased toward European ancestry individuals. Here, we performed eQTL and sQTL analyses using TOPMed whole-genome sequencing-derived genotype data and RNA-sequencing data from stored peripheral blood mononuclear cells in 1,012 African American participants from the Jackson Heart Study (JHS). At a false discovery rate of 5%, we identified 17,630 unique eQTL credible sets covering 16,538 unique genes; and 24,525 unique sQTL credible sets covering 9,605 unique genes, with lead QTL at P < 5e-8. About 24% of independent eQTLs and independent sQTLs with a minor allele frequency > 1% in JHS were rare (minor allele frequency < 0.1%), and therefore unlikely to be detected, in European ancestry individuals. Finally, we created an open database, which is freely available online, allowing fast query and bulk download of our QTL results.

Cell-type specific inference from bulk RNA-sequencing data by integrating single cell reference profiles via EPIC-unmix

Cell type specific (CTS) analysis is essential to reveal biological insights obscured in bulk tissue data. However, single-cell (sc) or single-nuclei (sn) resolution data are still cost-prohibitive for large-scale samples. Thus, computational methods to perform deconvolution from bulk tissue data are highly valuable. We here present EPIC-unmix, a novel two-step empirical Bayesian method integrating reference sc/sn RNA-seq data and bulk RNA-seq data from target samples to enhance the accuracy of CTS inference. We demonstrate through comprehensive simulations across three tissues that EPIC-unmix achieved 4.6% - 109.8% higher accuracy compared to alternative methods. By applying EPIC-unmix to human bulk brain RNA-seq data from the ROSMAP and MSBB cohorts, we identified multiple genes differentially expressed between Alzheimer's disease (AD) cases versus controls in a CTS manner, including 57.4% novel genes not identified using similar sample size sc/snRNA-seq data, indicating the power of our in-silico approach. Among the 6-69% overlapping, 83%-100% are in consistent direction with those from sc/snRNA-seq data, supporting the reliability of our findings. EPIC-unmix inferred CTS expression profiles similarly empowers CTS eQTL analysis. Among the novel eQTLs, we highlight a microglia eQTL for AD risk gene AP3B2, obscured in bulk and missed by sc/snRNA-seq based eQTL analysis. The variant resides in a microglia-specific cCRE, forming chromatin loop with AP3B2 promoter region in microglia. Taken together, we believe EPIC-unmix will be a valuable tool to enable more powerful CTS analysis.

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.

Admix-kit: an integrated toolkit and pipeline for genetic analyses of admixed populations

SUMMARY

Admixed populations, with their unique and diverse genetic backgrounds, are often underrepresented in genetic studies. This oversight not only limits our understanding but also exacerbates existing health disparities. One major barrier has been the lack of efficient tools tailored for the special challenges of genetic studies of admixed populations. Here, we present admix-kit, an integrated toolkit and pipeline for genetic analyses of admixed populations. Admix-kit implements a suite of methods to facilitate genotype and phenotype simulation, association testing, genetic architecture inference, and polygenic scoring in admixed populations.

AVAILABILITY AND IMPLEMENTATION

Admix-kit package is open-source and available at https://github.com/KangchengHou/admix-kit. Additionally, users can use the pipeline designed for admixed genotype simulation available at https://github.com/UW-GAC/admix-kit_workflow.