Stability of polygenic scores across discovery genome-wide association studies

Low Stability and Specificity of Polygenic Risk Scores for Major Psychiatric Disorders Limit their Clinical Utility

BACKGROUND

There has been little examination of the stability and validity of polygenic risk scores (PRS) i.e., whether individuals identified as high-risk for a disorder with one PRS are identified as high-risk with another PRS, and whether high-risk individuals have the disorder.

METHODS

The UK Biobank recruited 502,534 individuals aged 37-73 in the UK between 2006-2010. PRS were calculated for 408,853 white British individuals. PRS-CS, which uses SNP effect sizes under continuous shrinkage, was used to calculate three different PRS for major depressive disorder (MDD), alcohol use disorder (AUD), and type 2 diabetes (T2D), and two different PRS for schizophrenia (SCZ). PRS stability was measured using correlations between different PRS for the same disorder, and percentage of individuals consistently identified as high-risk (top 5% PRS). Sensitivity and specificity were used to measure PRS validity.

RESULTS

Correlations between PRS ranged from low to high (SCZ: r=0.78; MDD: r=0.16-0.78; AUD: r=0.13-0.90; T2D r=0.29-0.77). Percentage of individuals consistently identified as high-risk (top 5% PRS) for schizophrenia with different SCZ PRS was 47.7%, i.e., less than half of individuals identified as high-risk with one PRS were identified as high-risk with another PRS. Percentages of individuals consistently identified as high-risk were 9.5-47.0% for MDD, 8.3-63.5% for AUD, and 14.1-45.2% for T2D. Sensitivity of PRS was moderate for MDD (66.1-74.4%) and AUD (72.3-74.2%), moderate/good for T2D (77.3-96.3%), and good for SCZ (90.2-93.3%). Specificity was low for all PRS (50.7-56.4%).

CONCLUSIONS

Limited stability and specificity of PRS highlight their lack of clinical utility in psychiatry.

Distinguishing clinical and genetic risk factors for suicidal ideation and behavior in a diverse hospital population

Suicidal ideation (SI) and behavior (SB) are major public health concerns, but risk factors for their development and progression are poorly understood. We used ICD codes and a natural language processing algorithm to identify individuals in a hospital biobank with SI-only, SB, and controls without either. We compared the profiles of SB and SI-only patients to controls, and each other, using phenome-wide association studies (PheWAS) and polygenic risk scores (PRS). PheWAS identified many risk factors for SB and SI-only, plus specific psychiatric disorders which may be involved in progression from SI-only to SB. PRS for suicide attempt were only associated with SB, and even after accounting for psychiatric disorder PRS. SI PRS were only associated with SI-only, although not after accounting for psychiatric disorder PRS. These findings advance understanding of distinct genetic and clinical risk factors for SB and SI-only, which will aid in early detection and intervention efforts.

Genetic predisposition for negative affect predicts mental health burden during the COVID-19 pandemic

The coronavirus disease 2019 (COVID-19) pandemic was accompanied by an increase in mental health challenges including depression, stress, loneliness, and anxiety. Common genetic variants can contribute to the risk for psychiatric disorders and may present a risk factor in times of crises. However, it is unclear to what extent polygenic risk played a role in the mental health response to the COVID-19 pandemic. In this study, we investigate whether polygenic scores (PGSs) for mental health-related traits can distinguish between four resilience-vulnerability trajectories identified during the COVID-19 pandemic and associated lockdowns in 2020/21. We used multinomial regression in a genotyped subsample (n = 1316) of the CovSocial project. The most resilient trajectory characterized by the lowest mental health burden and the highest recovery rates served as the reference group. Compared to this most resilient trajectory, a higher value on the PGS for the well-being spectrum decreased the odds for individuals to be in one of the more vulnerable trajectories (adjusted R-square = 0.3%). Conversely, a higher value on the PGS for neuroticism increased the odds for individuals to be in one of the more vulnerable trajectories (adjusted R-square = 0.2%). Latent change in mental health burden extracted from the resilience-vulnerability trajectories was not associated with any PGS. Although our findings support an influence of PGS on mental health during COVID-19, the small added explained variance suggests limited utility of such genetic markers for the identification of vulnerable individuals in the general population.

Considerations for the application of polygenic scores to clinical care of individuals with substance use disorders

Substance use disorders (SUDs) are highly prevalent and associated with excess morbidity, mortality, and economic costs. Thus, there is considerable interest in the early identification of individuals who may be more susceptible to developing SUDs and in improving personalized treatment decisions for those who have SUDs. SUDs are known to be influenced by both genetic and environmental factors. Polygenic scores (PGSs) provide a single measure of genetic liability that could be used as a biomarker in predicting disease development, progression, and treatment response. Although PGSs are rapidly being integrated into clinical practice, there is little information to guide clinicians in their responsible use and interpretation. In this Review, we discuss the potential benefits and pitfalls of the use of PGSs in the clinical care of SUDs, highlighting current research. We also provide suggestions for important considerations prior to implementing the clinical use of PGSs and recommend future directions for research.

Clinical and genetic contributions to medical comorbidity in bipolar disorder: a study using electronic health records-linked biobank data

Bipolar disorder is a chronic and complex polygenic disease with high rates of comorbidity. However, the independent contribution of either diagnosis or genetic risk of bipolar disorder to the medical comorbidity profile of individuals with the disease remains unresolved. Here, we conducted a multi-step phenome-wide association study (PheWAS) of bipolar disorder using phenomes derived from the electronic health records of participants enrolled in the Mayo Clinic Biobank and the Mayo Clinic Bipolar Disorder Biobank. First, we explored the conditions associated with a diagnosis of bipolar disorder by conducting a phenotype-based PheWAS followed by LASSO-penalized regression to account for correlations within the phenome. Then, we explored the conditions associated with bipolar disorder polygenic risk score (BD-PRS) using a PRS-based PheWAS with a sequential exclusion approach to account for the possibility that diagnosis, instead of genetic risk, may drive such associations. 53,386 participants (58.7% women) with a mean age at analysis of 67.8 years (SD = 15.6) were included. A bipolar disorder diagnosis (n = 1479) was associated with higher rates of psychiatric conditions, injuries and poisonings, endocrine/metabolic and neurological conditions, viral hepatitis C, and asthma. BD-PRS was associated with psychiatric comorbidities but, in contrast, had no positive associations with general medical conditions. While our findings warrant confirmation with longitudinal-prospective studies, the limited associations between bipolar disorder genetics and medical conditions suggest that shared environmental effects or environmental consequences of diagnosis may have a greater impact on the general medical comorbidity profile of individuals with bipolar disorder than its genetic risk.

All of Us diversity and scale improve polygenic prediction contextually with greatest improvements for under-represented populations

Recent studies have demonstrated that polygenic risk scores (PRS) trained on multi-ancestry data can improve prediction accuracy in groups historically underrepresented in genomic studies, but the availability of linked health and genetic data from large-scale diverse cohorts representative of a wide spectrum of human diversity remains limited. To address this need, the All of Us research program (AoU) generated whole-genome sequences of 245,388 individuals who collectively reflect the diversity of the USA. Leveraging this resource and another widely-used population-scale biobank, the UK Biobank (UKB) with a half million participants, we developed PRS trained on multi-ancestry and multi-biobank data with up to ~750,000 participants for 32 common, complex traits and diseases across a range of genetic architectures. We then compared effects of ancestry, PRS methodology, and genetic architecture on PRS accuracy across a held out subset of ancestrally diverse AoU participants. Due to the more heterogeneous study design of AoU, we found lower heritability on average compared to UKB (0.075 vs 0.165), which limited the maximal achievable PRS accuracy in AoU. Overall, we found that the increased diversity of AoU significantly improved PRS performance in some participants in AoU, especially underrepresented individuals, across multiple phenotypes. Notably, maximizing sample size by combining discovery data across AoU and UKB is not the optimal approach for predicting some phenotypes in African ancestry populations; rather, using data from only AoU for these traits resulted in the greatest accuracy. This was especially true for less polygenic traits with large ancestry-enriched effects, such as neutrophil count (R 2: 0.055 vs. 0.035 using AoU vs. cross-biobank meta-analysis, respectively, because of e.g. DARC). Lastly, we calculated individual-level PRS accuracies rather than grouping by continental ancestry, a critical step towards interpretability in precision medicine. Individualized PRS accuracy decays linearly as a function of ancestry divergence, but the slope was smaller using multi-ancestry GWAS compared to using European GWAS. Our results highlight the potential of biobanks with more balanced representations of human diversity to facilitate more accurate PRS for the individuals least represented in genomic studies.

Impact of individual level uncertainty of lung cancer polygenic risk score (PRS) on risk stratification

BACKGROUND

Although polygenic risk score (PRS) has emerged as a promising tool for predicting cancer risk from genome-wide association studies (GWAS), the individual-level accuracy of lung cancer PRS and the extent to which its impact on subsequent clinical applications remains largely unexplored.

METHODS

Lung cancer PRSs and confidence/credible interval (CI) were constructed using two statistical approaches for each individual: (1) the weighted sum of 16 GWAS-derived significant SNP loci and the CI through the bootstrapping method (PRS-16-CV) and (2) LDpred2 and the CI through posteriors sampling (PRS-Bayes), among 17,166 lung cancer cases and 12,894 controls with European ancestry from the International Lung Cancer Consortium. Individuals were classified into different genetic risk subgroups based on the relationship between their own PRS mean/PRS CI and the population level threshold.

RESULTS

Considerable variances in PRS point estimates at the individual level were observed for both methods, with an average standard deviation (s.d.) of 0.12 for PRS-16-CV and a much larger s.d. of 0.88 for PRS-Bayes. Using PRS-16-CV, only 25.0% of individuals with PRS point estimates in the lowest decile of PRS and 16.8% in the highest decile have their entire 95% CI fully contained in the lowest and highest decile, respectively, while PRS-Bayes was unable to find any eligible individuals. Only 19% of the individuals were concordantly identified as having high genetic risk (> 90th percentile) using the two PRS estimators. An increased relative risk of lung cancer comparing the highest PRS percentile to the lowest was observed when taking the CI into account (OR = 2.73, 95% CI: 2.12-3.50, P-value = 4.13 × 10-15) compared to using PRS-16-CV mean (OR = 2.23, 95% CI: 1.99-2.49, P-value = 5.70 × 10-46). Improved risk prediction performance with higher AUC was consistently observed in individuals identified by PRS-16-CV CI, and the best performance was achieved by incorporating age, gender, and detailed smoking pack-years (AUC: 0.73, 95% CI = 0.72-0.74).

CONCLUSIONS

Lung cancer PRS estimates using different methods have modest correlations at the individual level, highlighting the importance of considering individual-level uncertainty when evaluating the practical utility of PRS.

Allostatic load in early adolescence: gene / environment contributions and relevance for mental health

Background

Allostatic load is the cumulative "wear and tear" on the body due to chronic adversity. We aimed to test poly-environmental (exposomic) and polygenic contributions to allostatic load and their combined contribution to early adolescent mental health.

Methods

We analyzed data on N = 5,035 diverse youth (mean age 12) from the Adolescent Brain Cognitive Development Study (ABCD). Using dimensionality reduction method, we calculated and overall allostatic load score (AL) using body mass index [BMI], waist circumference, blood pressure, blood glycemia, blood cholesterol, and salivary DHEA. Childhood exposomic risk was quantified using multi-level environmental exposures before age 11. Genetic risk was quantified using polygenic risk scores (PRS) for metabolic system susceptibility (type 2 diabetes [T2D]) and stress-related psychiatric disease (major depressive disorder [MDD]). We used linear mixed effects models to test main, additive, and interactive effects of exposomic and polygenic risk (independent variables) on AL (dependent variable). Mediation models tested the mediating role of AL on the pathway from exposomic and polygenic risk to youth mental health. Models adjusted for demographics and genetic principal components.

Results

We observed disparities in AL with non-Hispanic White youth having significantly lower AL compared to Hispanic and Non-Hispanic Black youth. In the diverse sample, childhood exposomic burden was associated with AL in adolescence (beta=0.25, 95%CI 0.22-0.29, P<.001). In European ancestry participants (n=2,928), polygenic risk of both T2D and depression was associated with AL (T2D-PRS beta=0.11, 95%CI 0.07-0.14, P<.001; MDD-PRS beta=0.05, 95%CI 0.02-0.09, P=.003). Both polygenic scores showed significant interaction with exposomic risk such that, with greater polygenic risk, the association between exposome and AL was stronger. AL partly mediated the pathway to youth mental health from exposomic risk and from MDD-PRS, and fully mediated the pathway from T2D-PRS.

Conclusions

AL can be quantified in youth using anthropometric and biological measures and is mapped to exposomic and polygenic risk. Main and interactive environmental and genetic effects support a diathesis-stress model. Findings suggest that both environmental and genetic risk be considered when modeling stress-related health conditions.

Behavior genetics and randomized controlled trials: A misleading analogy

Madole & Harden argue that just as the results of randomized controlled trials (RCTs) represent gains in causal knowledge and are useful, despite their limitations, so too are the findings of human behavior genetics. We argue that this analogy is misleading. Unlike RCTs, the results of human behavior genetics research cannot suggest efficacious interventions, nor point toward future research.

Comparing Pruning and Thresholding with Continuous Shrinkage Polygenic Score Methods in a Large Sample of Ancestrally Diverse Adolescents from the ABCD Study®

Using individuals' genetic data researchers can generate Polygenic Scores (PS) that are able to predict risk for diseases, variability in different behaviors as well as anthropomorphic measures. This is achieved by leveraging models learned from previously published large Genome-Wide Association Studies (GWASs) associating locations in the genome with a phenotype of interest. Previous GWASs have predominantly been performed in European ancestry individuals. This is of concern as PS generated in samples with a different ancestry to the original training GWAS have been shown to have lower performance and limited portability, and many efforts are now underway to collect genetic databases on individuals of diverse ancestries. In this study, we compare multiple methods of generating PS, including pruning and thresholding and Bayesian continuous shrinkage models, to determine which of them is best able to overcome these limitations. To do this we use the ABCD Study, a longitudinal cohort with deep phenotyping on individuals of diverse ancestry. We generate PS for anthropometric and psychiatric phenotypes using previously published GWAS summary statistics and examine their performance in three subsamples of ABCD: African ancestry individuals (n = 811), European ancestry Individuals (n = 6703), and admixed ancestry individuals (n = 3664). We find that the single ancestry continuous shrinkage method, PRScs (CS), and the multi ancestry meta method, PRScsx Meta (CSx Meta), show the best performance across ancestries and phenotypes.

Polygenic Contributions to Suicidal Thoughts and Behaviors in a Sample Ascertained for Alcohol Use Disorders

Introduction

Suicidal thoughts and behaviors have partially distinct genetic etiologies.

Methods

We used PRS-CS to create polygenic risk scores (PRSs) from GWAS of non-suicidal self-injury, broad-sense self-harm ideation, nonfatal suicide attempt, death by suicide, and depression. Using mixed-effect models, we estimated whether these PRSs were associated with a range of suicidal thoughts and behaviors in the Collaborative Study on the Genetics of Alcoholism (N = 7,526).

Results

All PRSs were significantly associated with suicidal ideation and suicide attempt (betas = 0.08-0.44, false discovery rate [FDR] <0.023). All PRSs except non-suicidal self-injury PRS were associated with active suicidal ideation (betas = 0.14-0.22, FDR <0.003). Several associations remained significant in models where all significant PRSs were included as simultaneous predictors, and when all PRSs predicted suicide attempt, the PRS together explained 6.2% of the variance in suicide attempt. Significant associations were also observed between some PRSs and persistent suicidal ideation, non-suicidal self-injury, compounded suicide attempt, and desire to die.

Conclusion

Our findings suggest that PRS for depression does not explain the entirety of the variance in suicidal thoughts and behaviors, with PRS specifically for suicidal thoughts and behaviors making additional and sometimes unique contributions.

Improving the computation efficiency of polygenic risk score modeling: faster in Julia

To enable computationally efficient polygenic risk score (PRS) calculations, PRS.jl translates a field standard PRS construction method, PRS-CS, to the Julia programming language.

To enable large-scale application of polygenic risk scores (PRSs) in a computationally efficient manner, we translate a widely used PRS construction method, PRS–continuous shrinkage, to the Julia programming language, PRS.jl. On nine different traits with varying genetic architectures, we demonstrate that PRS.jl maintains accuracy of prediction while decreasing the average runtime by 5.5×. Additional programmatic modifications improve usability and robustness. This freely available software substantially improves work flow and democratizes usage of PRSs by lowering the computational burden of the PRS–continuous shrinkage method.

Challenges and Opportunities for Developing More Generalizable Polygenic Risk Scores

Polygenic risk scores (PRS) estimate an individual's genetic likelihood of complex traits and diseases by aggregating information across multiple genetic variants identified from genome-wide association studies. PRS can predict a broad spectrum of diseases and have therefore been widely used in research settings. Some work has investigated their potential applications as biomarkers in preventative medicine, but significant work is still needed to definitively establish and communicate absolute risk to patients for genetic and modifiable risk factors across demographic groups. However, the biggest limitation of PRS currently is that they show poor generalizability across diverse ancestries and cohorts. Major efforts are underway through methodological development and data generation initiatives to improve their generalizability. This review aims to comprehensively discuss current progress on the development of PRS, the factors that affect their generalizability, and promising areas for improving their accuracy, portability, and implementation.

Gene-based polygenic risk scores analysis of alcohol use disorder in African Americans

Genome-wide association studies (GWAS) in admixed populations such as African Americans (AA) have limited sample sizes, resulting in poor performance of polygenic risk scores (PRS). Based on the observations that many disease-causing genes are shared between AA and European ancestry (EA) populations, and some disease-causing variants are located within the boundaries of these genes, we proposed a novel gene-based PRS framework (PRSgene) by using variants located within disease-associated genes. Using the AA GWAS of alcohol use disorder (AUD) from the Million Veteran Program and the EA GWAS of problematic alcohol use as the discovery GWAS, we identified 858 variants from 410 genes that were AUD-related in both AA and EA. PRSgene calculated using these variants were significantly associated with AUD in three AA target datasets (P-values ranged from 7.61E-05 to 6.27E-03; Betas ranged from 0.15 to 0.21) and outperformed PRS calculated using all variants (P-values ranged from 7.28E-03 to 0.16; Betas ranged from 0.06 to 0.18). PRSgene were also associated with AUD in an EA target dataset (P-value = 0.02, Beta = 0.11). In AA, individuals in the highest PRSgene decile had an odds ratio of 1.76 (95% CI: 1.32-2.34) to develop AUD compared to those in the lowest decile. The 410 genes were enriched in 54 Gene Ontology biological processes, including ethanol oxidation and processes involving the synaptic system, which are known to be AUD-related. In addition, 26 genes were targets of drugs used to treat AUD or other diseases that might be considered for repurposing to treat AUD. Our study demonstrated that the gene-based PRS had improved performance in evaluating AUD risk in AA and provided new insight into AUD genetics.

Copy Number Variant Risk Scores Associated With Cognition, Psychopathology, and Brain Structure in Youths in the Philadelphia Neurodevelopmental Cohort

Importance

Psychiatric and cognitive phenotypes have been associated with a range of specific, rare copy number variants (CNVs). Moreover, IQ is strongly associated with CNV risk scores that model the predicted risk of CNVs across the genome. But the utility of CNV risk scores for psychiatric phenotypes has been sparsely examined.

Objective

To determine how CNV risk scores, common genetic variation indexed by polygenic scores (PGSs), and environmental factors combine to associate with cognition and psychopathology in a community sample.

Design, Setting, and Participants

The Philadelphia Neurodevelopmental Cohort is a community-based study examining genetics, psychopathology, neurocognition, and neuroimaging. Participants were recruited through the Children's Hospital of Philadelphia pediatric network. Participants with stable health and fluency in English underwent genotypic and phenotypic characterization from November 5, 2009, through December 30, 2011. Data were analyzed from January 1 through July 30, 2021.

Exposures

The study examined (1) CNV risk scores derived from models of burden, predicted intolerance, and gene dosage sensitivity; (2) PGSs from genomewide association studies related to developmental outcomes; and (3) environmental factors, including trauma exposure and neighborhood socioeconomic status.

Main Outcomes and Measures

The study examined (1) neurocognition, with the Penn Computerized Neurocognitive Battery; (2) psychopathology, with structured interviews based on the Schedule for Affective Disorders and Schizophrenia for School-Age Children; and (3) brain volume, with magnetic resonance imaging.

Results

Participants included 9498 youths aged 8 to 21 years; 4906 (51.7%) were female, and the mean (SD) age was 14.2 (3.7) years. After quality control, 18 185 total CNVs greater than 50 kilobases (10 517 deletions and 7668 duplications) were identified in 7101 unrelated participants genotyped on Illumina arrays. In these participants, elevated CNV risk scores were associated with lower overall accuracy on cognitive tests (standardized β = 0.12; 95% CI, 0.10-0.14; P = 7.41 × 10-26); lower accuracy across a range of cognitive subdomains; increased overall psychopathology; increased psychosis-spectrum symptoms; and higher deviation from a normative developmental model of brain volume. Statistical models of developmental outcomes were significantly improved when CNV risk scores were combined with PGSs and environmental factors.

Conclusions and Relevance

In this study, elevated CNV risk scores were associated with lower cognitive ability, higher psychopathology including psychosis-spectrum symptoms, and greater deviations from normative magnetic resonance imaging models of brain development. Together, these results represent a step toward synthesizing rare genetic, common genetic, and environmental factors to understand clinically relevant outcomes in youth.

THE FEATURES OF POSTTRAUMATIC STRESS DISORDER DEVELOPMENT IN PATIENTS WITH DIABETES MELLITUS 2 TYPE

OBJECTIVE

The aim: The revealing of the development of stress-related disorders in patients with type 2 diabetes mellitus (DM 2) to: identify the prevalence of stress-related disorders, particularly, posttraumatic stress disorder (PTSD); study the influence of psychosocial factors on the occurrence and course of stress-related disorders and increase the effectiveness of treatment in DM 2.

PATIENTS AND METHODS

Materials and methods: Research papers have been found by searching the PubMed database using the keywords ``ptsd and diabetes 2 type" with the result of 74 studies. Totally 25 of selected publications were analysed based on our criteria about the mechanisms through which the influence of psychosocial factors, permanent stressful or traumatic events on the probable risk of PTSD development and their analysis and relationships for the improvement of treatment effectiveness in DM 2 patients who have not been the veterans.

CONCLUSION

Conclusions: Given the complex neurophysiological relationships between the long-term stress and pathophysiological mechanisms of DM 2 - this group of patients has the higher risk of developing stress-related disorders, including PTSD.