AI interpretationAI-generated
Can whole-genome data reliably type blood groups? A 79-sample test says yes, with caveats.
This study asked whether whole-genome sequencing (WGS) can replace targeted methods for red blood cell antigen typing. The authors analysed WGS data from 79 individuals across nine blood group genes (KEL, FY, JK, YT, DO, CO, IN, DI, LW) plus HLA-DRB1, using a custom database of 1648 variable positions. Results matched older amplicon and SNaPshot typing about 93% of the time for blood groups and 91% for HLA-DRB1. Accuracy depended heavily on sequencing depth: reliable calls needed coverage above 15x.
- Blood group typing from WGS reached 93% concordance with established amplicon and SNaPshot methods.
- HLA-DRB1 typing from the same WGS data reached 91% concordance.
- Reliable typing required read depth strictly above 15x; lower coverage caused errors and unresolved calls.
- WGS also showed potential for screening donors carrying rare antigens such as weak JK alleles.
Sample interpretation
The study drew on whole-genome data from 79 individuals, all from a single site (UZ8_90). Paternal lineages are dominated by R1b-related branches: R-M87 (5), R-FT414862 (4), R-FT412111 (4), R-FT414871 (3), R-BY127338 (3), plus smaller R-FT414897, R-FT353743 and R-FT414889 groups. Maternal lines are more varied, led by Z7 (3) and U2b2 (3), then HV2a, R5a2, C4a1a-T195C! and M34'57 (2 each). This is a modest, geographically narrow cohort, so the blood group findings are methodological rather than population-level.
For genealogy enthusiasts
For ancestry hobbyists, this shows WGS raw data can yield medically useful blood group and HLA information, not just deep ancestry. But treat low-coverage calls with caution: without sufficient depth, antigen typing can be wrong. Rare antigen screening is a promising bonus.
Abstract
Many questions can be explored thanks to whole-genome data. The aim of this study was to overcome their main limits, software availability and database accuracy, and estimate the feasibility of red blood cell (RBC) antigen typing from whole-genome sequencing (WGS) data. We analyzed whole-genome data from 79 individuals for HLA-DRB1 and 9 RBC antigens. Whole-genome sequencing data was analyzed with software allowing phasing of variable positions to define alleles or haplotypes and validated for HLA typing from next-generation sequencing data. A dedicated database was set up with 1648 variable positions analyzed in KEL (KEL), ACKR1 (FY), SLC14A1 (JK), ACHE (YT), ART4 (DO), AQP1 (CO), CD44 (IN), SLC4A1 (DI) and ICAM4 (LW). Whole-genome sequencing typing was compared to that previously obtained by amplicon-based monoallelic sequencing and by SNaPshot analysis. Whole-genome sequencing data were also explored for other alleles. Our results showed 93% of concordance for blood group polymorphisms and 91% for HLA-DRB1. Incorrect typing and unresolved results confirm that WGS should be considered reliable with read depths strictly above 15x. Our results supported that RBC antigen typing from WGS is feasible but requires improvements in read depth for SNV polymorphisms typing accuracy. We also showed the potential for WGS in screening donors with rare blood antigens, such as weak JK alleles. The development of WGS analysis in immunogenetics laboratories would offer personalized care in the management of RBC disorders.