Bycroft · Nature 2018 · prospective cohort study · n=approximately 500,000

The UK Biobank resource with deep phenotyping and genomic data.

Cited 10015 times in the scientific literature.

Level 3 - non-randomized controlled study

Prospective cohort study resource profile and genetic methodology description.

PubMed 30305743 · doi:10.1038/s41586-018-0579-z · record verified 2026-08-29

What was done

The authors describe the UK Biobank prospective cohort resource and its centralized genetic data processing. The cohort recruited approximately 500,000 individuals aged 40 to 69 from across the United Kingdom. Data collected include biological measurements, lifestyle indicators, blood and urine biomarkers, brain and body imaging, linked medical records, and genome-wide genotype data across all participants. Centralized genetic workflows evaluated genotype quality, characterized population structure and relatedness, performed phasing, and imputed variants and classical human leukocyte antigen (HLA) alleles.

What was found

Phasing and genotype imputation increased the number of testable genetic variants to around 96 million across the ~500,000 participants. Classical allelic variation was imputed at 11 HLA genes, which successfully recovered signals of established associations between HLA alleles and various diseases.

Why it matters

This resource provides an unprecedented combination of sample size, deep phenotyping, longitudinal health record linkage, and dense genome-wide genetic data to power the discovery of genetic associations with complex human traits and diseases.

Limits

Participant recruitment was restricted to individuals aged 40 to 69 within the United Kingdom, which may limit generalizability to other age groups and globally diverse ancestral populations. In addition, the abstract provides descriptive methodology and imputation outputs rather than new disease-specific risk estimates or clinical outcome trials.

Cited by