Characterization of blood group variants in an Omani population by comparison of whole genome sequencing and serology

Abstract

Background: Although blood group variation was first described over a century ago, our understanding of the genetic variation affecting antigenic expression on the red blood cell surface in many populations is lacking. This deficit limits the ability to accurately type patients, especially as serological testing is not available for all described blood groups, and targeted genotyping panels may lack rare or population-specific variants.

Study design and methods: Here, we perform serological assays across 24 antigens and whole genome sequencing on 100 Omanis, a population underrepresented in genomic databases. We inferred blood group phenotypes using known genetic variants underlying antigen expression.

Results: Comparison of serological to genetically inferred phenotypes resulted in an average prediction accuracy of 98.7%. By investigating 12 discordances, we describe candidate variants in the Lewis, Lutheran, MNS, and P1 blood groups that could affect antigenic expression, although further functional confirmation is required. We identify blood group alleles that, to our knowledge, have not been previously reported in Omanis, including several most common in African populations, likely introduced to Oman by gene flow over the last thousand years.

Discussion: These findings highlight the need to evaluate individual populations and their population history when considering variants to include in genotype panels for blood group typing. This research will inform future genetic work in blood banks and transfusion services, ensuring that testing strategies are optimized for diverse genetic backgrounds.

FIGURE 1

Comparison of allele frequencies of select blood group variants newly described in the Omani population (OM) with frequencies reported for Middle Eastern (ME), African (AFR), South Asian (SAS), and European (EUR) populations in gnomAD v4.1. [Color figure can be viewed at wileyonlinelibrary.com]

FIGURE 2

Copy number inference across the glycophorin genes based on read coverage. Position on chromosome 4 is shown on the x‐axis, with the locations of GYPE, GYPB and GYPA genes indicated and demarcated with black vertical lines. The inferred copy number is indicated by color for the four individuals in this dataset with deletions or duplications affecting GYPB or GYPA along with HG02554, an individual from the 1000 Genomes dataset known to carry the Dantu structural variant. Orange indicates one copy (heterozygous deletion), white indicates two copies (no copy number variant), light blue indicates three copies (heterozygous duplication), and dark blue indicates four copies (heterozygous triplication). Dotted horizontal gray lines separate the inference for each sample, with sample names indicated on the left along with the serological results for M, N, S, and s antigens and genotype for the SNV determining expression of the S or s antigen (rs7683365, p.Thr48Met, c.143C>T). [Color figure can be viewed at wileyonlinelibrary.com]