Polygenic adaptation on height is overestimated due to uncorrected stratification in genome-wide association studies

Sohail, Mashaal; Maier, Robert; Ganna, Andrea; Bloemendal, Alex; Martin, Alicia R.; Turchin, Michael C.; Chiang, Charleston W.K.; Hirschhorn, Joel N.; Daly, Mark J.; Patterson, Nick; Neale, Benjamin M.; Mathieson, Iain; Reich, David; Sunyaev, Shamil

doi:10.7554/elife.39702

Cited by 364 publications

(446 citation statements)

References 43 publications

Supporting

Mentioning

420

Contrasting

Unclassified

Order By: Relevance

“…For the 2014 height GWAS, that inflation is reported as the LDSC intercept is 2.09 in [77], indicating considerable inflation; for the 2018 dataset we find σ 2 0 =2.5, while the LD score regression intercept is 2.1116 (se 0.0458). Given the various estimates of inflation and the controversy over population structure in the height data [50,51], it is not clear what results are definitely incorrect.…”

Section: Discussionmentioning

confidence: 99%

“…For height, we focused on the 2014 GWAS [46], not the more recent 2018 GWAS [49], although we also report below model results for the latter. There are issues pertaining to population structure in the various height GWAS [50,51], and the 2018 GWAS is a combination of GIANT and UKB GWAS, so some caution is warranted in interpreting results for these data.…”

Section: Data Preparationmentioning

confidence: 99%

See 1 more Smart Citation

Beyond SNP Heritability: Polygenicity and Discoverability of Phenotypes Estimated with a Univariate Gaussian Mixture Model

Holland¹,

Frei

Desikan³

et al. 2017

Preprint

View full text Add to dashboard Cite

Of signal interest in the genetics of traits are estimating the proportion, π 1 , of causally associated single nucleotide polymorphisms (SNPs), and their effect size variance, σ 2 β , which are components of the mean heritabilities captured by the causal SNP. Here we present the first model, using detailed linkage disequilibrium structure, to estimate these quantities from genome-wide association studies (GWAS) summary statistics, assuming a Gaussian distribution of SNP effect sizes, β. We apply the model to three diverse phenotypes -schizophrenia, putamen volume, and educational attainment -and validate it with extensive simulations. We find that schizophrenia is highly polygenic, with ≃ 5 × 10 4 causal SNPs distributed with small effect size variance, σ 2 β = 3.5 × 10 −5 (in units where the phenotype variance is normalized to 1), requiring a GWAS study with more than 1/2-million samples in each arm for full discovery. In contrast, putamen volume involves only ≃ 3 × 10 2 causal SNPs, but with σ 2 β = 1.2 × 10 −3 , indicating a much larger proportion of the causal SNPs that are strongly associated. Educational attainment has similar polygenicity to schizophrenia, but with effects that are substantially weaker, σ 2 β = 5 × 10 −6 , leading to much lower heritability. Thus the model is able to describe the broad genetic architecture of phenotypes where both polygenicity and effect size variance range over several orders of magnitude, shows why only small proportions of heritability have been explained for discovered SNPs, and provides a roadmap for future GWAS discoveries.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Data Preparationmentioning

confidence: 99%

Beyond SNP Heritability: Polygenicity and Discoverability of Phenotypes Estimated with a Univariate Gaussian Mixture Model

Holland¹,

Frei

Desikan³

et al. 2017

Preprint

View full text Add to dashboard Cite

show abstract

“…Human height is controlled by more than 500 loci (Wood et al, ), although recent estimates suggest that this number is probably too high since population structure has not been adequately considered (Berg et al, ; Sohail et al ). We choose the following parameter ranges: 0.001–0.01 for the effect sizes (measured in units of the standard deviation, where in the case of human height 1 SD

\approx

6.5 cm; Turchin et al, ), s around 0.1 (Turchin et al, ), the number of loci affecting the trait l = 200, and mutation rate of about 10 −5 per generation.…”

Section: Modelmentioning

confidence: 99%

Important role of genetic drift in rapid polygenic adaptation

John

Stephan

2020

Ecology and Evolution

View full text Add to dashboard Cite

We analyzed a model to determine the factors that facilitate or limit rapid polygenic adaptation. This model includes population genetic terms of mutation and both directional and stabilizing selection on a highly polygenic trait in a diploid population of finite size. First, we derived the equilibrium distribution of the allele frequencies of the multilocus model by diffusion approximation. This formula describing the equilibrium allele frequencies as a mutation‐selection‐drift balance was examined by computer simulation using parameter values inferred for human height, a well‐studied polygenic trait. Second, assuming that a sudden environmental shift of the fitness optimum occurs while the population is in equilibrium, we analyzed the adaptation of the trait to the new optimum. The speed at which the trait mean approaches the new optimum increases with the equilibrium genetic variance. Thus, large population size and/or large mutation rate may facilitate rapid adaptation. Third, the contribution of an individual locus i to polygenic adaptation depends on the compound parameter γipi0qi0, where γi is the effect size, pi0 the equilibrium frequency of the trait‐increasing allele of this locus, and qi0=1-pi0. Thus, only loci with large values of this parameter contribute coherently to polygenic adaptation. Given that mutation rates are relatively small, this is more likely in large populations, in which the effects of drift are limited.

show abstract

“…A similar GWAS meta-analysis was also conducted for cortical thickness, but given the massive expansion of SA in modern humans and only subtle increases in cortical thickness as compared to extant mammalian species 5 , we chose to focus on SA for this study. Association was assessed in 35,660 individuals from 55 cohorts across the lifespan, including data from the UK Biobank 16 (UKBB, N = 9,923). From standard structural magnetic resonance imaging (MRI) scans of the brain, the ENIGMA consortium automatically measured total SA across the cortex, as well as average bilateral SA for each of 34 cortical regions defined by gyral anatomy 30 .…”

Section: Genome-wide Association To Human Cortical Surface Areamentioning

confidence: 99%

The Evolutionary History of Common Genetic Variants Influencing Human Cortical Surface Area

Tilot

Khramtsova

Grasby

et al. 2019

Preprint

View full text Add to dashboard Cite

Structural brain changes along the lineage that led to modern Homo sapiens have contributed to our unique cognitive and social abilities. However, the evolutionarily relevant molecular variants impacting key aspects of neuroanatomy are largely unknown. Here, we integrate evolutionary annotations of the genome at diverse timescales with common variant associations from large-scale neuroimaging genetic screens in living humans, to reveal how selective pressures have shaped neocortical surface area. We show that variation within human gained enhancers active in the developing brain is associated with global surface area as well as that of specific regions. Moreover, we find evidence of recent polygenic selection over the past 2,000 years influencing surface area of multiple cortical regions, including those involved in spoken language and visual processing.

show abstract

Polygenic adaptation on height is overestimated due to uncorrected stratification in genome-wide association studies

Cited by 364 publications

References 43 publications

Beyond SNP Heritability: Polygenicity and Discoverability of Phenotypes Estimated with a Univariate Gaussian Mixture Model

Beyond SNP Heritability: Polygenicity and Discoverability of Phenotypes Estimated with a Univariate Gaussian Mixture Model

Important role of genetic drift in rapid polygenic adaptation

The Evolutionary History of Common Genetic Variants Influencing Human Cortical Surface Area

Contact Info

Product

Resources

About