Systematic single-variant and gene-based association testing of thousands of phenotypes in 426,370 UK Biobank exomes

Abstract

Genome-wide association studies have successfully discovered thousands of common variants associated with human diseases and traits, but the landscape of rare variation in human disease has not been explored at scale. Exome sequencing studies of population biobanks provide an opportunity to systematically evaluate the impact of rare coding variation across a wide range of phenotypes to discover genes and allelic series relevant to human health and disease. Here, we present results from systematic association analyses of 3,700 phenotypes using single-variant and gene tests of 281,850 individuals in the UK Biobank with exome sequence data. We find that the discovery of genetic associations is tightly linked to frequency as well as correlated with metrics of deleteriousness and natural selection. We highlight biological findings elucidated by these data and release the dataset as a public resource alongside a browser framework for rapidly exploring rare variant association results.

Genetic associations of protein-coding variants in human disease

“…Our approach benefits considerably from the Finnish genetic background, where certain alleles are stochastically enriched to unusually high allele frequencies [6][7][8] , at times exceeding population frequencies in the UK Biobank by more than 50-fold. Our theoretical and empirical results suggest the increasing utility of enriched variants for identifying associations quantitatively towards lower allelic frequencies.…”

“…For instance, the UK Biobank (UKB) is a resource that captures detailed phenotype information matched to genetic data for more than 500,000 individuals and, since its inception, has facilitated biomedical discoveries at an unprecedented scale 2 . We and others have recently reported on the ongoing efforts to sequence the exomes of all UKB participants and link genetic findings to a broad range of phenotypes [3][4][5][6] . We also established FinnGen (FG) (https://www.finngen.fi), an academic-industry collaboration to identify genotype-phenotype correlations in the Finnish founder population with the aim of better understanding how the genome affects health.…”

Association of ultra‐rare coding variants with genetic generalized epilepsy: A case–control whole exome sequencing study

“…GABRG2 was also the lead association in a burden analysis of pLoF URVs (p = 6.9 × 10 −5 ) in a recent study investigating the exomes of 3999 individuals with epilepsy (without further phenotypic subclassification) versus 277 586 controls from the UK Biobank. 41 The different definitions of QVs in these studies might also explain the variable outcomes. Our PPh2 analysis model is similar to the prior model we used to analyze a subset of our samples (thus allowing for comparisons of outcomes with the increase in sample size).…”

Set-based rare variant association tests for biobank scale sequencing data sets

“…For example, in our evaluation, the most recent set-based test, STAAR 2 , cannot control for type I error rates in the presence of case-control imbalance (Supplementary Figure 1). SAIGE-GENE has been recently used for exome-wide association analysis for 3,700 phenotypes of 281,850 individuals in the UKBB WES data 5 . In the evaluation of the method using the UKBB WES data with 160K white British samples, we have found that all the tests (Burden, SKAT 6 and SKAT-O 7 ) in SAIGE-GENE performed well when testing all rare variants with MAF  1% (Figure 1A), but inflation was observed in SKAT and SKAT-O tests in SAIGE-GENE when restricting to variants with MAF  0.1% or MAF  0.01% and the case-control rates were more unbalanced than 1:30 (Figure 1A, Supplementary Figure 2).…”