487,180 items are available, covering 487,180 participants.
Units of measurement are blocks.
Detailed summary is not currently available.
Imputation from genotype using the TOPMed R2 panel, performed by the TOPMed Informatics Research Center. Primary description of the imputation data can be found in: Taliun, D., Harris, D.N., Kessler, M.D. et al. Sequencing of 53,831 diverse genomes from the NHLBI TOPMed Program. Nature 590, 290-299 (2021) https://doi.org/10.1038/s41586-021-03205-y
After phasing the UK Biobank genetic data (carried out on 81 chromosomal chunks using Eagle v.2.4), the phased data were converted from GRCh37 to GRCh38 using LiftOver. Imputation was performed using Minimac4 v1.0.2 (https://genome.sph.umich.edu/wiki/Minimac4
). The reference panel is the TOPMed R2 panel, which is available to public at https://imputation.biodatacatalyst.nhlbi.nih.gov
. Imputation was performed in 1Mb chunks and merged back together by chromosome. Due to large data size, markers with poor imputation were not retained (excluded Minimac4 imputation quality metric R^2 < 0.1). Across all the chromosomes, 677,037 markers were used for imputation.
For distribution to UK Biobank, output Minimac4 VCF files were converted to BGEN format using qctool v2.0.8. The conversion used the Genotype Probability (GP) field from source VCF and the probabilities are stored using 8 bit precision.
Details on contributing cohorts for the TOPMed Imputataion Panel can be seen in Extended Table 3 of Sequencing of 53,831 diverse genomes from the NHLBI TOPMed Program. Nature 590, 290-299 (2021). https://doi.org/10.1038/s41586-021-03205-y