488,264 items of data are available, covering 488,264 participants.
Average compound dataset is 766b, largest is 962b.
Each entry contains all HLA loci (one line per person) in the following order:
A, B, C, DRB5, DRB4, DRB3, DRB1, DQB1, DQA1, DPB1, DPA1
There is one value (separated by commas) per imputed allele and only the best-guess alleles are reported. The quantities P(A1|D) and P(A2|D,A1) from the Q column of the HLA:IMP*2 output are reported for each locus. For further information on the calculation methodology and the Quality metric see Resource 182.
To extract this data into a single file, approved researchers in possession of a UKB dataset may use -s22182 as a parameter to the ukbconv utility. To download an example row of data see Resource 1520.