The UCLA ATLAS Community Health Initiative: Promoting precision health research in a diverse biobank

Association between plausible genetic factors and weight loss from GLP1-RA and bariatric surgery

Obesity is a major public health challenge. Glucagon-like peptide-1 receptor agonists (GLP1-RA) and bariatric surgery (BS) are effective weight loss interventions; however, the genetic factors influencing treatment response remain largely unexplored. Moreover, most previous studies have focused on race and ethnicity rather than genetic ancestry. Here we analyzed 10,960 individuals from 9 multiancestry biobank studies across 6 countries to assess the impact of known genetic factors on weight loss. Between 6 and 12 months, GLP1-RA users had an average weight change of -3.93% or -6.00%, depending on the outcome definition, with modest ancestry-based differences. BS patients experienced -21.17% weight change between 6 and 48 months. We found no significant associations between GLP1-RA-induced weight loss and polygenic scores for body mass index or type 2 diabetes, nor with missense variants in GLP1R. A higher body mass index polygenic score was modestly linked to lower weight loss after BS (+0.7% per s.d., P = 1.24 × 10-4), but the effect attenuated in sensitivity analyses. Our findings suggest known genetic factors have limited impact on GLP1-RA effectiveness with respect to weight change and confirm treatment efficacy across ancestry groups.

Inclusion bias affects common variant discovery and replication in a health-system linked biobank

Electronic Health Records (EHR) -linked biobanks have emerged as promising tools for precision medicine, enabling the integration of clinical and molecular data for individual risk assessment. Association studies performed in biobank studies can connect common genetic variation to clinical phenotypes, such as through the use of polygenic scores (PGS), which are starting to have utility in aiding clinician decision making. However, while biobanks aggregate large amounts of data effectively for such studies, most employ various opt-in consent protocols, and, as a result, are expected to be subject to participation and recruitment biases. The extent to which biases affect genetic analyses in biobanks remains unstudied. In this study, we quantify bias and evaluate its impact on genetic analyses, using the UCLA ATLAS Community Health Initiative as a case study. Our analyses reveal that a wide array of factors, particularly socio-demographic characteristics and healthcare utilization patterns, influence participation, effectively differentiating biobank participants from the broader patient population (AUROC = 0.85, AUPRC = 0.82). Through weighting the sample using inverse probability weights derived from probabilities of enrollment, we replicated 54% more known GWAS variants than models that did not take bias into account (e.g. associations between variants in the PPARG gene and type 2 diabetes). We further show that PGS-Phenome wide associations are affected by the weighting scheme, and suggest associations corroborated by weighted analyses to be more robust. Our results highlight that genetic analyses within biobanks should account for inclusion biases, and suggest inverse probability weighting as a potential approach.

Reduced Penetrance is Common Among Predicted Loss-of-Function Variants and is Likely Driven by Residual Allelic Activity

Loss-of-function genetic variants (LoFs) often result in severe phenotypes, including autosomal dominant diseases driven by haploinsufficiency. Due to low carrier frequencies, their penetrance is generally unknown but typically variable. Here, we investigate the penetrance of >6,000 predicted LoFs (pLoFs) linked to 91 haploinsufficient diseases using a cohort of ≈24,000 carriers with linked electronic health record data. We find evidence for widespread reduced penetrance, which persisted after accounting for variant annotation artifacts, missed diagnoses, and incomplete clinical data. We thus hypothesized that many pLoFs have incomplete penetrance, which may be driven by residual allelic activity. To test this, we trained machine learning models to predict pLoF penetrance using variant-specific genomic features that may correlate with incomplete loss-of-function. The models were predictive of pLoF penetrance across a range of diseases and variant types, including those with prior clinical evidence for pathogenicity. This suggests that many pLoFs have incomplete penetrance due to residual allelic activity, complicating disease prognostication in asymptomatic carriers.

Phenotypic evaluation of deep learning models for classifying germline variant pathogenicity

Deep learning models for predicting variant pathogenicity have not been thoroughly evaluated on real-world clinical phenotypes. Here, we apply state-of-the-art pathogenicity prediction models to hereditary breast cancer gene variants in UK Biobank participants. Model predictions for missense variants in BRCA1, BRCA2 and PALB2, but not ATM and CHEK2, were associated with breast cancer risk. However, deep learning models had limited clinical utility when specifically applied to variants of uncertain significance.

Population Performance and Individual Agreement of Coronary Artery Disease Polygenic Risk Scores

Importance

Polygenic risk scores (PRSs) for coronary artery disease (CAD) are a growing clinical and commercial reality. Whether existing scores provide similar individual-level assessments of disease liability is a critical consideration for clinical implementation that remains uncharacterized.

Objective

Characterize the reliability of CAD PRSs that perform equivalently at the population level at predicting individual-level risk.

Design

Cross-sectional Study.

Setting

All of Us Research Program (AOU), Penn Medicine Biobank (PMBB), and UCLA ATLAS Precision Health Biobank.

Participants

Volunteers of diverse genetic backgrounds enrolled in AOU, PMBB, and UCLA with available electronic health record and genotyping data.

Exposures

Polygenic risk for CAD from previously published PRSs and new PRSs developed separately from the testing cohorts.

Main Outcomes and Measures

Sets of CAD PRSs that perform population prediction equivalently were identified by comparing calibration and discrimination (Brier score and AUROC) of generalized linear models of prevalent CAD using Bayesian analysis of variance. Among equivalently performing scores, individual-level agreement between risk estimates was tested with intraclass correlation (ICC) and Light's Kappa, measures of inter-rater reliability.

Results

50 PRSs were calculated for 171,095 AOU participants. When included in a model of prevalent CAD, 48 scores had practically equivalent Brier scores and AUROCs (region of practical equivalence = 0.02). Across these scores, 84% of participants had at least one score in both the top and bottom risk quintile. Continuous agreement of individual risk predictions from the 48 scores was poor, with an ICC of 0.351 (95% CI; 0.349, 0.352). Agreement between two statistically equivalent scores was moderate, with an ICC of 0.649 (95% CI; 0.646, 0.652). Light's Kappa, used to evaluate consistency of assignment to high-risk thresholds, did not exceed 0.56 (interpreted as 'fair') across statistically and practically equivalent scores. Repeating the analysis among 41,193 PMBB and 50,748 UCLA participants yielded different sets of statistically and practically equivalent scores which also lacked strong individual agreement.

Conclusions and Relevance

Across three diverse biobanks, CAD PRSs that performed equivalently at the population level produced unreliable individual risk estimates. Approaches to clinical implementation of CAD PRSs must consider the potential for discordant individual risk estimates from otherwise indistinguishable scores.

Scalable summary-statistics-based heritability estimation method with individual genotype level accuracy

SNP heritability, the proportion of phenotypic variation explained by genotyped SNPs, is an important parameter in understanding the genetic architecture underlying various diseases and traits. Methods that aim to estimate SNP heritability from individual genotype and phenotype data are limited by their ability to scale to Biobank-scale data sets and by the restrictions in access to individual-level data. These limitations have motivated the development of methods that only require summary statistics. Although the availability of publicly accessible summary statistics makes them widely applicable, these methods lack the accuracy of methods that utilize individual genotypes. Here we present a SUMmary-statistics-based Randomized Haseman-Elston regression (SUM-RHE), a method that can estimate the SNP heritability of complex phenotypes with accuracies comparable to approaches that require individual genotypes, while exclusively relying on summary statistics. SUM-RHE employs Genome-Wide Association Study (GWAS) summary statistics and statistics obtained on a reference population, which can be efficiently estimated and readily shared for public use. Our results demonstrate that SUM-RHE obtains estimates of SNP heritability that are substantially more accurate compared with other summary statistic methods and on par with methods that rely on individual-level data.

Improving genetic risk modeling of dementia from real-world data in underrepresented populations

Genetic risk modeling for dementia offers significant benefits, but studies based on real-world data, particularly for underrepresented populations, are limited. We employ an Elastic Net model for dementia risk prediction using single-nucleotide polymorphisms prioritized by functional genomic data from multiple neurodegenerative disease genome-wide association studies. We compare this model with APOE and polygenic risk score models across genetic ancestry groups (Hispanic Latino American sample: 610 patients with 126 cases; African American sample: 440 patients with 84 cases; East Asian American sample: 673 patients with 75 cases), using electronic health records from UCLA Health for discovery and the All of Us cohort for validation. Our model significantly outperforms other models across multiple ancestries, improving the area-under-precision-recall curve by 31-84% (Wilcoxon signed-rank test p-value <0.05) and the area-under-the-receiver-operating characteristic by 11-17% (DeLong test p-value <0.05) compared to the APOE and the polygenic risk score models. We identify shared and ancestry-specific risk genes and biological pathways, reinforcing and adding to existing knowledge. Our study highlights the benefits of integrating functional mapping, multiple neurodegenerative diseases, and machine learning for genetic risk models in diverse populations. Our findings hold potential for refining precision medicine strategies in dementia diagnosis.

Interpreting SNP heritability in admixed populations

SNP heritabilityh s n p 2 is defined as the proportion of phenotypic variance explained by genotyped SNPs and is believed to be a lower bound of heritability (h 2 ), being equal to it if all causal variants are known. Despite the simple intuition behindh s n p 2 , its interpretation and equivalence toh 2 is unclear, particularly in the presence of population structure and assortative mating. It is well known that population structure can lead to inflation inh ˆ s n p 2 estimates because of confounding due to linkage disequilibrium (LD) or shared environment. Here we use analytical theory and simulations to demonstrate thath s n p 2 estimates can be biased in admixed populations, even in the absence of confounding and even if all causal variants are known. This is because admixture generates LD, which contributes to the genetic variance, and therefore to heritability. Genome-wide restricted maximum likelihood (GREML) does not capture this contribution leading to under- or over-estimates ofh s n p 2 relative toh 2 , depending on the genetic architecture. In contrast, Haseman-Elston (HE) regression exaggerates the LD contribution leading to biases in the opposite direction. For the same reason, GREML and HE estimates of local ancestry heritabilityh γ 2 are also biased. We describe this bias inh ˆ s n p 2 andh ˆ γ 2 as a function of admixture history and the genetic architecture of the trait and show that it can be recovered under some conditions. We clarify the interpretation ofh ˆ s n p 2 in admixed populations and discuss its implication for genome-wide association studies and polygenic prediction.

Improving genetic risk modeling of dementia from real-world data in underrepresented populations

BACKGROUND

Genetic risk modeling for dementia offers significant benefits, but studies based on real-world data, particularly for underrepresented populations, are limited.

METHODS

We employed an Elastic Net model for dementia risk prediction using single-nucleotide polymorphisms prioritized by functional genomic data from multiple neurodegenerative disease genome-wide association studies. We compared this model with APOE and polygenic risk score models across genetic ancestry groups, using electronic health records from UCLA Health for discovery and All of Us cohort for validation.

RESULTS

Our model significantly outperforms other models across multiple ancestries, improving the area-under-precision-recall curve by 21-61% and the area-under-the-receiver-operating characteristic by 10-21% compared to the APOEand the polygenic risk score models. We identified shared and ancestry-specific risk genes and biological pathways, reinforcing and adding to existing knowledge.

CONCLUSIONS

Our study highlights benefits of integrating functional mapping, multiple neurodegenerative diseases, and machine learning for genetic risk models in diverse populations. Our findings hold potential for refining precision medicine strategies in dementia diagnosis.

Improving genetic risk modeling of dementia from real-world data in underrepresented populations

BACKGROUND

Genetic risk modeling for dementia offers significant benefits, but studies based on real-world data, particularly for underrepresented populations, are limited.

METHODS

We employed an Elastic Net model for dementia risk prediction using single-nucleotide polymorphisms prioritized by functional genomic data from multiple neurodegenerative disease genome-wide association studies. We compared this model with APOE and polygenic risk score models across genetic ancestry groups, using electronic health records from UCLA Health for discovery and All of Us cohort for validation.

RESULTS

Our model significantly outperforms other models across multiple ancestries, improving the area-under-precision-recall curve by 21-61% and the area-under-the-receiver-operating characteristic by 10-21% compared to the APOE and the polygenic risk score models. We identified shared and ancestry-specific risk genes and biological pathways, reinforcing and adding to existing knowledge.

CONCLUSIONS

Our study highlights benefits of integrating functional mapping, multiple neurodegenerative diseases, and machine learning for genetic risk models in diverse populations. Our findings hold potential for refining precision medicine strategies in dementia diagnosis.

Multi-class Modeling Identifies Shared Genetic Risk for Late-onset Epilepsy and Alzheimer's Disease

Background

Previous studies have established a strong link between late-onset epilepsy (LOE) and Alzheimer's disease (AD). However, their shared genetic risk beyond the APOE gene remains unclear. Our study sought to examine the shared genetic factors of AD and LOE, interpret the biological pathways involved, and evaluate how AD onset may be mediated by LOE and shared genetic risks.

Methods

We defined phenotypes using phecodes mapped from diagnosis codes, with patients' records aged 60-90. A two-step Least Absolute Shrinkage and Selection Operator (LASSO) workflow was used to identify shared genetic variants based on prior AD GWAS integrated with functional genomic data. We calculated an AD-LOE shared risk score and used it as a proxy in a causal mediation analysis. We used electronic health records from an academic health center (UCLA Health) for discovery analyses and validated our findings in a multi-institutional EHR database (All of Us).

Results

The two-step LASSO method identified 34 shared genetic loci between AD and LOE, including the APOE region. These loci were mapped to 65 genes, which showed enrichment in molecular functions and pathways such as tau protein binding and lipoprotein metabolism. Individuals with high predicted shared risk scores have a higher risk of developing AD, LOE, or both in their later life compared to those with low-risk scores. LOE partially mediates the effect of AD-LOE shared genetic risk on AD (15% proportion mediated on average). Validation results from All of Us were consistent with findings from the UCLA sample.

Conclusions

We employed a machine learning approach to identify shared genetic risks of AD and LOE. In addition to providing substantial evidence for the significant contribution of the APOE-TOMM40-APOC1 gene cluster to shared risk, we uncovered novel genes that may contribute. Our study is one of the first to utilize All of Us genetic data to investigate AD, and provides valuable insights into the potential common and disease-specific mechanisms underlying AD and LOE, which could have profound implications for the future of disease prevention and the development of targeted treatment strategies to combat the co-occurrence of these two diseases.

Polygenic scores for tobacco use provide insights into systemic health risks in a diverse EHR-linked biobank in Los Angeles

Tobacco use is a major risk factor for many diseases and is heavily influenced by environmental factors with significant underlying genetic contributions. Here, we evaluated the predictive performance, risk stratification, and potential systemic health effects of tobacco use disorder (TUD) predisposing germline variants using a European- ancestry-derived polygenic score (PGS) in 24,202 participants from the multi-ancestry, hospital-based UCLA ATLAS biobank. Among genetically inferred ancestry groups (GIAs), TUD-PGS was significantly associated with TUD in European American (EA) (OR: 1.20, CI: [1.16, 1.24]), Hispanic/Latin American (HL) (OR:1.19, CI: [1.11, 1.28]), and East Asian American (EAA) (OR: 1.18, CI: [1.06, 1.31]) GIAs but not in African American (AA) GIA (OR: 1.04, CI: [0.93, 1.17]). Similarly, TUD-PGS offered strong risk stratification across PGS quantiles in EA and HL GIAs and inconsistently in EAA and AA GIAs. In a cross-ancestry phenome-wide association meta-analysis, TUD-PGS was associated with cardiometabolic, respiratory, and psychiatric phecodes (17 phecodes at P < 2.7E-05). In individuals with no history of smoking, the top TUD-PGS associations with obesity and alcohol-related disorders (P = 3.54E-07, 1.61E-06) persist. Mendelian Randomization (MR) analysis provides evidence of a causal association between adiposity measures and tobacco use. Inconsistent predictive performance of the TUD-PGS across GIAs motivates the inclusion of multiple ancestry populations at all levels of genetic research of tobacco use for equitable clinical translation of TUD-PGS. Phenome associations suggest that TUD-predisposed individuals may require comprehensive tobacco use prevention and management approaches to address underlying addictive tendencies.

Genetic risk prediction in Hispanics/Latinos: milestones, challenges, and social-ethical considerations

Genome-wide association studies (GWAS) have allowed the identification of disease-associated variants, which can be leveraged to build polygenic scores (PGSs). Even though PGSs can be a valuable tool in personalized medicine, their predictive power is limited in populations of non-European ancestry, particularly in admixed populations. Recent efforts have focused on increasing racial and ethnic diversity in GWAS, thus, addressing some of the limitations of genetic risk prediction in these populations. Even with these efforts, few studies focus exclusively on Hispanics/Latinos. Additionally, Hispanic/Latino populations are often considered a single population despite varying admixture proportions between and within ethnic groups, diverse genetic heterogeneity, and demographic history. Combined with highly heterogeneous environmental and socioeconomic exposures, this diversity can reduce the transferability of genetic risk prediction models. Given the recent increase of genomic studies that include Hispanics/Latinos, we review the milestones and efforts that focus on genetic risk prediction, summarize the potential for improving PGS transferability, and highlight the challenges yet to be addressed. Additionally, we summarize social-ethical considerations and provide ideas to promote genetic risk prediction models that can be implemented equitably.

Decoding Genetics, Ancestry, and Geospatial Context for Precision Health

Mass General Brigham, an integrated healthcare system based in the Greater Boston area of Massachusetts, annually serves 1.5 million patients. We established the Mass General Brigham Biobank (MGBB), encompassing 142,238 participants, to unravel the intricate relationships among genomic profiles, environmental context, and disease manifestations within clinical practice. In this study, we highlight the impact of ancestral diversity in the MGBB by employing population genetics, geospatial assessment, and association analyses of rare and common genetic variants. The population structures captured by the genetics mirror the sequential immigration to the Greater Boston area throughout American history, highlighting communities tied to shared genetic and environmental factors. Our investigation underscores the potency of unbiased, large-scale analyses in a healthcare-affiliated biobank, elucidating the dynamic interplay across genetics, immigration, structural geospatial factors, and health outcomes in one of the earliest American sites of European colonization.

Disease risk and healthcare utilization among ancestrally diverse groups in the Los Angeles region

An individual's disease risk is affected by the populations that they belong to, due to shared genetics and environmental factors. The study of fine-scale populations in clinical care is important for identifying and reducing health disparities and for developing personalized interventions. To assess patterns of clinical diagnoses and healthcare utilization by fine-scale populations, we leveraged genetic data and electronic medical records from 35,968 patients as part of the UCLA ATLAS Community Health Initiative. We defined clusters of individuals using identity by descent, a form of genetic relatedness that utilizes shared genomic segments arising due to a common ancestor. In total, we identified 376 clusters, including clusters with patients of Afro-Caribbean, Puerto Rican, Lebanese Christian, Iranian Jewish and Gujarati ancestry. Our analysis uncovered 1,218 significant associations between disease diagnoses and clusters and 124 significant associations with specialty visits. We also examined the distribution of pathogenic alleles and found 189 significant alleles at elevated frequency in particular clusters, including many that are not regularly included in population screening efforts. Overall, this work progresses the understanding of health in understudied communities and can provide the foundation for further study into health inequities.

Polygenic scoring accuracy varies across the genetic ancestry continuum

Polygenic scores (PGSs) have limited portability across different groupings of individuals (for example, by genetic ancestries and/or social determinants of health), preventing their equitable use1-3. PGS portability has typically been assessed using a single aggregate population-level statistic (for example, R2)4, ignoring inter-individual variation within the population. Here, using a large and diverse Los Angeles biobank5 (ATLAS, n = 36,778) along with the UK Biobank6 (UKBB, n = 487,409), we show that PGS accuracy decreases individual-to-individual along the continuum of genetic ancestries7 in all considered populations, even within traditionally labelled 'homogeneous' genetic ancestries. The decreasing trend is well captured by a continuous measure of genetic distance (GD) from the PGS training data: Pearson correlation of -0.95 between GD and PGS accuracy averaged across 84 traits. When applying PGS models trained on individuals labelled as white British in the UKBB to individuals with European ancestries in ATLAS, individuals in the furthest GD decile have 14% lower accuracy relative to the closest decile; notably, the closest GD decile of individuals with Hispanic Latino American ancestries show similar PGS performance to the furthest GD decile of individuals with European ancestries. GD is significantly correlated with PGS estimates themselves for 82 of 84 traits, further emphasizing the importance of incorporating the continuum of genetic ancestries in PGS interpretation. Our results highlight the need to move away from discrete genetic ancestry clusters towards the continuum of genetic ancestries when considering PGSs.