UK Biobank whole-genome sequencing reveals robust contributions of rare variants to complex-trait heritability

The missing heritability of complex traits remains a key challenge in human genetics. Recent studies have highlighted the potential role of rare variants in addressing this gap. However, their contribution has been difficult to quantify due to small sample sizes and large standard errors in large-scale whole-genome sequencing data.

Leveraging whole-genome sequencing data from 348,977 unrelated White British participants in the UK Biobank, we analyze 5,378,300 linkage disequilibrium-independent rare variants to estimate their heritability for five complex traits. This dataset is more than tenfold larger than previous whole-genome sequencing studies, enabling substantially reduced standard errors and more precise estimates. Using genome-wide complex trait analysis and BOLT-REML, we find that rare variants explain 0.37, 0.15, 0.20, 0.13, and 0.11 of the heritability for standing height, body mass index, platelet count, systolic blood pressure, and glycated hemoglobin, respectively. After adjusting for a polygenic score based on common variants, the estimated rare-variant heritability decreased, but standing height, BMI, platelet count, and systolic blood pressure retained contributions (0.23, 0.10, 0.05, and 0.06), suggesting that rare-variant effects may not be fully explained by common-variant architecture, although their magnitude and stability vary across traits.

Our findings demonstrate that rare variants explain a trait-specific proportion of heritability, refining our understanding of the genetic architecture of complex traits. By leveraging the largest whole genome sequencing dataset to date, this study provides a precise estimate of the contribution of rare variants to the heritability of multiple complex traits and highlights the value of large-scale whole-genome sequencing in resolving the sources of missing heritability.