A haplotype reference panel constructed from 490,319 UK Biobank genomes improves genotype imputation for global populations

Haplotype reference panel built from WGS of 490,319 UK Biobank participants, phasing and imputation with EagleImp2 and EagleImp-RAP

The authors built an imputation reference panel from whole-genome sequencing of 490,319 UK Biobank participants, usable standalone or within the UK Biobank Research Analysis Platform through EagleImp2 and EagleImp-RAP. Compared with the TOPMed r3 panel (133,597 individuals), it lowers phasing switch error rates for all global superpopulations except AMR and for 25 of 26 1000 Genomes Project subpopulations. Imputation accuracy (MAE) improved for 4 of 5 superpopulations and 12 subpopulations for variants with minor allele frequencies down to 0.01%, and estimated squared correlation improved across all 5 superpopulations, with the largest gains in East and South Asian populations. Reimputation of COVID-19 genome-wide association datasets from Italy, Germany, Spain and Norway showed that common-variant imputation remains unsaturated, enhancing locus discovery and fine-mapping resolution.

Synthesis written by Geno'X. For the full original abstract, please refer to the source publication.

Primarily a research resource: imputation serves association studies first, not individual Mendelian diagnosis. The measured gain in East and South Asian populations is the notable point, and the weaker phasing for the AMR superpopulation is a reminder that panel size does not replace the ancestral diversity it contains.

Analysis by Dr Thibaut Benquey