Stability of Polygenic Scores Across Discovery Genome-Wide Association Studies

Polygenic scores (PGS) are commonly evaluated in terms of their predictive accuracy at the population level by the proportion of phenotypic variance they explain. To be useful for precision medicine applications, they also need to be evaluated at the individual patient level when phenotypes are not necessarily already known. Hence, we investigated the stability of PGS in European-American (EUR)- and African-American (AFR)-ancestry individuals from the Philadelphia Neurodevelopmental Cohort (PNC) and the Adolescent Brain Cognitive Development Study (ABCD) using different discovery GWAS for post-traumatic stress disorder (PTSD), type-2 diabetes (T2D), and height. We found that pairs of EUR-ancestry GWAS for the same trait had genetic correlations > 0.92. However, PGS calculated from pairs of same-ancestry and different-ancestry GWAS had correlations that ranged from <0.01 to 0.74. PGS stability was higher for GWAS that explained more of the trait variance, with height PGS being more stable than PTSD or T2D PGS. Focusing on the upper end of the PGS distribution, different discovery GWAS do not consistently identify the same individuals in the upper quantiles, with the best case being 60% of individuals above the 80th percentile of PGS overlapping from one height GWAS to another. The degree of overlap decreases sharply as higher quantiles, less heritable traits, and different-ancestry GWAS are considered. PGS computed from different discovery GWAS have only modest correlation at the level of the individual patient, underscoring the need to proceed cautiously with integrating PGS into precision medicine applications.

Download Full-text

Sociology, Genetics, and the Coming of Age of Sociogenomics

Annual Review of Sociology ◽

10.1146/annurev-soc-121919-054756 ◽

2020 ◽

Vol 46 (1) ◽

pp. 553-581 ◽

Cited By ~ 1

Author(s):

Melinda C. Mills ◽

Felix C. Tropf

Keyword(s):

Coming Of Age ◽

Association Studies ◽

Molecular Genetic ◽

Genetic Correlations ◽

Genetic Research ◽

Well Being ◽

Environment Interaction ◽

Genome Wide Association Studies ◽

Molecular Genetic Data ◽

Polygenic Scores

Recent years have seen the birth of sociogenomics via the infusion of molecular genetic data. We chronicle the history of genetics, focusing particularly on post-2005 genome-wide association studies, the post-2015 big data era, and the emergence of polygenic scores. We argue that understanding polygenic scores, including their genetic correlations with each other, causation, and underlying biological architecture, is vital. We show how genetics can be introduced to understand a myriad of topics such as fertility, educational attainment, intergenerational social mobility, well-being, addiction, risky behavior, and longevity. Although models of gene-environment interaction and correlation mirror agency and structure models in sociology, genetics is yet to be fully discovered by this discipline. We conclude with a critical reflection on the lack of diversity, nonrepresentative samples, precision policy applications, ethics, and genetic determinism. We argue that sociogenomics can speak to long-standing sociological questions and that sociologists can offer innovative theoretical, measurement, and methodological innovations to genetic research.

Download Full-text

Accurate genetic profiling of anthropometric traits using a big data approach

10.1101/033134 ◽

2015 ◽

Cited By ~ 1

Author(s):

Oriol Canela-Xandri ◽

Konrad Rawlik ◽

John A. Woolliams ◽

Albert Tenesa

Keyword(s):

Disease Management ◽

Prediction Models ◽

Disease Risk ◽

Association Studies ◽

Genomic Medicine ◽

Full Potential ◽

Phenotypic Variance ◽

Genome Wide Association Studies ◽

Starting Point ◽

The Individual

Genome-wide association studies (GWAS) promised to translate their findings into clinically beneficial improvements of patient management by tailoring disease management to the individual through the prediction of disease risk. However, the ability to translate genetic findings from GWAS into predictive tools that are of clinical utility and which may inform clinical practice has, so far, been encouraging but limited. Here we propose to use a more powerful statistical approach that enables the prediction of multiple medically relevant phenotypes without the costs associated with developing a genetic test for each of them. As a proof of principle, we used a common panel of 319,038 SNPs to train the prediction models in 114,264 unrelated White-British for height and four obesity related traits (body mass index, basal metabolic rate, body fat percentage, and waist-to-hip ratio). We obtained prediction accuracies that ranged between 46% and 75% of the maximum achievable given their explained heritable component. This represents an improvement of up to 75% over the phenotypic variance explained by the predictors developed through large collaborations, which used more than twice as many training samples. Across-population predictions in White non-British individuals were similar to those of White-British whilst those in Asian and Black individuals were informative but less accurate. The genotyping of circa 500,000 UK Biobank participants will yield predictions ranging between 66% and 83% of the maximum. We anticipate that our models and a common panel of genetic markers, which can be used across multiple traits and diseases, will be the starting point to tailor disease management to the individual. Ultimately, we will be able to capitalise on whole-genome sequence and environmental risk factors to realise the full potential of genomic medicine.

Download Full-text

Learning polygenic scores for human blood cell traits

10.1101/2020.02.17.952788 ◽

2020 ◽

Author(s):

Yu Xu ◽

Dragana Vuckovic ◽

Scott C Ritchie ◽

Parsa Akbari ◽

Tao Jiang ◽

...

Keyword(s):

Machine Learning ◽

Blood Cell ◽

Association Studies ◽

Relative Effectiveness ◽

Univariate Analysis ◽

Genetic Correlations ◽

Genome Wide Association Studies ◽

Learning Methods ◽

Polygenic Scores ◽

Using Data

AbstractPolygenic scores (PGSs) for blood cell traits can be constructed using summary statistics from genome-wide association studies. As the selection of variants and the modelling of their interactions in PGSs may be limited by univariate analysis, therefore, such a conventional method may yield sub-optional performance. This study evaluated the relative effectiveness of four machine learning and deep learning methods, as well as a univariate method, in the construction of PGSs for 26 blood cell traits, using data from UK Biobank (n=~400,000) and INTERVAL (n=~40,000). Our results showed that learning methods can improve PGSs construction for nearly every blood cell trait considered, with this superiority explained by the ability of machine learning methods to capture interactions among variants. This study also demonstrated that populations can be well stratified by the PGSs of these blood cell traits, even for traits that exhibit large differences between ages and sexes, suggesting potential for disease prevention. As our study found genetic correlations between the PGSs for blood cell traits and PGSs for several common human diseases (recapitulating well-known associations between the blood cell traits themselves and certain diseases), it suggests that blood cell traits may be indicators or/and mediators for a variety of common disorders via shared genetic variants and functional pathways.

Download Full-text

Polygenic scores for height in admixed populations

10.1101/2020.04.08.030361 ◽

2020 ◽

Cited By ~ 2

Author(s):

Bárbara D. Bitarello ◽

Iain Mathieson

Keyword(s):

Predictive Power ◽

Predictive Accuracy ◽

Association Studies ◽

Effect Sizes ◽

European Ancestry ◽

Marginal Effect ◽

Genome Wide Association Studies ◽

Major Barrier ◽

Cohort Differences ◽

Polygenic Scores

AbstractPolygenic risk scores (PRS) use the results of genome-wide association studies (GWAS) to predict quantitative phenotypes or disease risk at an individual level. This provides a potential route to the use of genetic data in personalized medical care. However, a major barrier to the use of PRS is that the majority of GWAS come from cohorts of European ancestry. The predictive power of PRS constructed from these studies is substantially lower in non-European ancestry cohorts, although the reasons for this are unclear. To address this question, we investigate the performance of PRS for height in cohorts with admixed African and European ancestry, allowing us to evaluate ancestry-related differences in PRS predictive accuracy while controlling for environment and cohort differences. We first show that that the predictive accuracy of height PRS increases linearly with European ancestry and is largely explained by European ancestry segments of the admixed genomes. We show that differences in allele frequencies, recombination rate, and marginal effect sizes across ancestries all contribute to the decrease in predictive power, but none of these effects explain the decrease on its own. Finally, we demonstrate that prediction for admixed individuals can be improved by using a linear combination of PRS that includes ancestry-specific effect sizes, although this approach is at present limited by the small size of non-European ancestry discovery cohorts.

Download Full-text

Polygenic Scores for Height in Admixed Populations

G3 Genes|Genome|Genetics ◽

10.1534/g3.120.401658 ◽

2020 ◽

Vol 10 (11) ◽

pp. 4027-4036 ◽

Cited By ~ 2

Author(s):

Bárbara D. Bitarello ◽

Iain Mathieson

Keyword(s):

Predictive Power ◽

Predictive Accuracy ◽

Association Studies ◽

Effect Sizes ◽

European Ancestry ◽

Marginal Effect ◽

Genome Wide Association Studies ◽

Major Barrier ◽

Cohort Differences ◽

Polygenic Scores

Polygenic risk scores (PRS) use the results of genome-wide association studies (GWAS) to predict quantitative phenotypes or disease risk at an individual level, and provide a potential route to the use of genetic data in personalized medical care. However, a major barrier to the use of PRS is that the majority of GWAS come from cohorts of European ancestry. The predictive power of PRS constructed from these studies is substantially lower in non-European ancestry cohorts, although the reasons for this are unclear. To address this question, we investigate the performance of PRS for height in cohorts with admixed African and European ancestry, allowing us to evaluate ancestry-related differences in PRS predictive accuracy while controlling for environment and cohort differences. We first show that the predictive accuracy of height PRS increases linearly with European ancestry and is partially explained by European ancestry segments of the admixed genomes. We show that recombination rate, differences in allele frequencies, and differences in marginal effect sizes across ancestries all contribute to the decrease in predictive power, but none of these effects explain the decrease on its own. Finally, we demonstrate that prediction for admixed individuals can be improved by using a linear combination of PRS that includes ancestry-specific effect sizes, although this approach is at present limited by the small size of non-European ancestry discovery cohorts.

Download Full-text

Largest genome-wide association study for PTSD identifies genetic risk loci in European and African ancestries and implicates novel biological pathways

10.1101/458562 ◽

2018 ◽

Cited By ~ 6

Author(s):

Caroline M. Nievergelt ◽

Adam X. Maihofer ◽

Torsten Klengel ◽

Elizabeth G. Atkinson ◽

Chia-Yen Chen ◽

...

Keyword(s):

Genome Wide Association Study ◽

Association Studies ◽

Meta Analysis ◽

African Ancestry ◽

Genetic Correlations ◽

Genome Wide Association ◽

European Ancestry ◽

Genome Wide Association Studies ◽

Post Traumatic Stress ◽

Genome Wide

AbstractPost-traumatic stress disorder (PTSD) is a common and debilitating disorder. The risk of PTSD following trauma is heritable, but robust common variants have yet to be identified by genome-wide association studies (GWAS). We have collected a multi-ethnic cohort including over 30,000 PTSD cases and 170,000 controls. We first demonstrate significant genetic correlations across 60 PTSD cohorts to evaluate the comparability of these phenotypically heterogeneous studies. In this largest GWAS meta-analysis of PTSD to date we identify a total of 6 genome-wide significant loci, 4 in European and 2 in African-ancestry analyses. Follow-up analyses incorporated local ancestry and sex-specific effects, and functional studies. Along with other novel genes, a non-coding RNA (ncRNA) and a Parkinson’s Disease gene,PARK2, were associated with PTSD. Consistent with previous reports, SNP-based heritability estimates for PTSD range between 10-20%. Despite a significant shared liability between PTSD and major depressive disorder, we show evidence that some of our loci may be specific to PTSD. These results demonstrate the role of genetic variation contributing to the biology of differential risk for PTSD and the necessity of expanding GWAS beyond European ancestry.

Download Full-text

Genetics of Sleep and Insights into Its Relationship with Obesity

Annual Review of Nutrition ◽

10.1146/annurev-nutr-082018-124258 ◽

2021 ◽

Vol 41 (1) ◽

Author(s):

Hassan S. Dashti ◽

José M. Ordovás

Keyword(s):

Sleep Disorders ◽

Healthy Lifestyle ◽

Association Studies ◽

Genetic Correlations ◽

European Ancestry ◽

Annual Review ◽

Publication Date ◽

Genome Wide Association Studies ◽

Genome Wide ◽

Polygenic Scores

Considerable recent advancements in elucidating the genetic architecture of sleep traits and sleep disorders may provide insight into the relationship between sleep and obesity. Despite the considerable involvement of the circadian clock in sleep and metabolism, few shared genes, including FTO, were implicated in genome-wide association studies (GWASs) of sleep and obesity. Polygenic scores composed of signals from GWASs of sleep traits show largely null associations with obesity, suggesting lead variants are unique to sleep. Modest genome-wide genetic correlations are observed between many sleep traits and obesity and are largest for snoring.Notably, U-shaped positive genetic correlations with body mass index (BMI) exist for both short and long sleep durations. Findings from Mendelian randomization suggest robust causal effects of insomnia on higher BMI and, conversely, of higher BMI on snoring and daytime sleepiness. Bidirectional effects between sleep duration and daytime napping with obesity may also exist. Limited gene-sleep interaction studies suggest that achieving favorable sleep, as part of a healthy lifestyle, may attenuate genetic predisposition to obesity, but whether these improvements produce clinically meaningful reductions in obesity risk remains unclear. Investigations of the genetic link between sleep and obesity for sleep disorders other than insomnia and in populations of non-European ancestry are currently limited. Expected final online publication date for the Annual Review of Nutrition, Volume 41 is September 2021. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.

Download Full-text

GWAS of 19,629 individuals identifies novel genetic variants for regional brain volumes and refines their genetic co-architecture with cognitive and mental health traits

10.1101/586339 ◽

2019 ◽

Cited By ~ 5

Author(s):

Bingxin Zhao ◽

Tianyou Luo ◽

Tengfei Li ◽

Yun Li ◽

Jingwen Zhang ◽

...

Keyword(s):

Mental Health ◽

Sample Size ◽

Genetic Variants ◽

Association Studies ◽

Genetic Correlations ◽

Joint Analysis ◽

Phenotypic Variance ◽

Genome Wide Association Studies ◽

Brain Volumes ◽

Genome Wide

AbstractVolumetric variations of human brain are heritable and are associated with many brain-related complex traits. Here we performed genome-wide association studies (GWAS) and post-GWAS analyses of 101 brain volumetric phenotypes using the UK Biobank (UKB) sample including 19,629 participants. GWAS identified 287 independent SNPs exceeding genome-wide significance threshold of 4.9*10−10, adjusted for testing multiple phenotypes. Gene-based association study found 142 associated genes (113 new) and functional gene mapping analysis linked 122 more genes. Many of the discovered genetic variants have previously been implicated with cognitive and mental health traits (such as cognitive performance, education, mental disease/disorders), and significant genetic correlations were detected for 29 pairs of traits. The significant SNPs discovered in the UKB sample were supported by a joint analysis with other four independent studies (total sample size 2,192), and we performed a meta-analysis of five samples to provide GWAS summary statistics with sample size larger than 20,000. Using genome-wide polygenic risk scores prediction, up to 4.36% of phenotypic variance (p-value=2.97*10−22) in the four independent studies can be explained by the UKB GWAS results. In conclusion, our study identifies many new genetic variants at SNP, locus and gene levels and advances our understanding of the pleiotropy and genetic co-architecture between brain volumes and other traits.

Download Full-text

Power analysis of transcriptome-wide association study: Implications for practical protocol choice

PLoS Genetics ◽

10.1371/journal.pgen.1009405 ◽

2021 ◽

Vol 17 (2) ◽

pp. e1009405

Author(s):

Chen Cao ◽

Bowei Ding ◽

Qing Li ◽

Devin Kwok ◽

Jingjing Wu ◽

...

Keyword(s):

Association Study ◽

Statistical Power ◽

Predictive Accuracy ◽

Association Studies ◽

Mediating Effect ◽

Phenotypic Variance ◽

Genome Wide Association Studies ◽

Gene Expressions ◽

Specific Protocol ◽

Expression Model

The transcriptome-wide association study (TWAS) has emerged as one of several promising techniques for integrating multi-scale ‘omics’ data into traditional genome-wide association studies (GWAS). Unlike GWAS, which associates phenotypic variance directly with genetic variants, TWAS uses a reference dataset to train a predictive model for gene expressions, which allows it to associate phenotype with variants through the mediating effect of expressions. Although effective, this core innovation of TWAS is poorly understood, since the predictive accuracy of the genotype-expression model is generally low and further bounded by expression heritability. This raises the question: to what degree does the accuracy of the expression model affect the power of TWAS? Furthermore, would replacing predictions with actual, experimentally determined expressions improve power? To answer these questions, we compared the power of GWAS, TWAS, and a hypothetical protocol utilizing real expression data. We derived non-centrality parameters (NCPs) for linear mixed models (LMMs) to enable closed-form calculations of statistical power that do not rely on specific protocol implementations. We examined two representative scenarios: causality (genotype contributes to phenotype through expression) and pleiotropy (genotype contributes directly to both phenotype and expression), and also tested the effects of various properties including expression heritability. Our analysis reveals two main outcomes: (1) Under pleiotropy, the use of predicted expressions in TWAS is superior to actual expressions. This explains why TWAS can function with weak expression models, and shows that TWAS remains relevant even when real expressions are available. (2) GWAS outperforms TWAS when expression heritability is below a threshold of 0.04 under causality, or 0.06 under pleiotropy. Analysis of existing publications suggests that TWAS has been misapplied in place of GWAS, in situations where expression heritability is low.

Download Full-text

The Genetic Basis of Hypertriglyceridemia

Current Atherosclerosis Reports ◽

10.1007/s11883-021-00939-y ◽

2021 ◽

Vol 23 (8) ◽

Author(s):

Germán D. Carrasquilla ◽

Malene Revsbech Christiansen ◽

Tuomas O. Kilpeläinen

Keyword(s):

Genetic Basis ◽

Association Studies ◽

Cumulative Effect ◽

Genome Wide Association Studies ◽

Genetic Loci ◽

Risk Variants ◽

Genome Wide ◽

Severe Hypertriglyceridemia ◽

Increased Risk ◽

Polygenic Scores

Abstract Purpose of Review Hypertriglyceridemia is a common dyslipidemia associated with an increased risk of cardiovascular disease and pancreatitis. Severe hypertriglyceridemia may sometimes be a monogenic condition. However, in the vast majority of patients, hypertriglyceridemia is due to the cumulative effect of multiple genetic risk variants along with lifestyle factors, medications, and disease conditions that elevate triglyceride levels. In this review, we will summarize recent progress in the understanding of the genetic basis of hypertriglyceridemia. Recent Findings More than 300 genetic loci have been identified for association with triglyceride levels in large genome-wide association studies. Studies combining the loci into polygenic scores have demonstrated that some hypertriglyceridemia phenotypes previously attributed to monogenic inheritance have a polygenic basis. The new genetic discoveries have opened avenues for the development of more effective triglyceride-lowering treatments and raised interest towards genetic screening and tailored treatments against hypertriglyceridemia. Summary The discovery of multiple genetic loci associated with elevated triglyceride levels has led to improved understanding of the genetic basis of hypertriglyceridemia and opened new translational opportunities.

Download Full-text