A new algorithm based on whole-genome sequencing (bloodTyper) allows automated typing of red blood cell (RBC) and platelet antigens, researchers report. This finding would have the potential to inform patient care in pregnancy and future blood transfusions.
"The most surprising result was how accurately the algorithm typed for the ABO and Rh blood groups, which are both the most clinically important blood groups, but also the most complex," said Dr. William J. Lane from Brigham and Women's Hospital and Harvard Medical School, in Boston.
"I think one of the most interesting results was that we identified an uncommon platelet antigen change, HPA-1(a-), found in 1/50 people and a rare blood group antigen change, Lu(b-), found in only 1/500 people," he told.
Extended antigen matching for more than the traditional ABO and RhD blood-group antigens is not commonly done, because the conventional antibody-based serological method is labor-intensive and costly, and reagent antibodies are not available for many clinically important antigens.
Dr. Lane and colleagues used whole-genome sequencing data from the MedSeq Project to develop an automated antigen-typing algorithm and compared the results with conventional serology and single nucleotide polymorphism (SNP) methods for 38 RBC and 22 platelet antigens.
In the first 20 genomes studied, the algorithm made 1,194 correct calls out of 1,200 individual antigen typing, for a concordance of 99.5%, the researchers report in The Lancet Haematology, online.
With further modifications, the algorithm was 99.8% concordant for the remaining 90 genomes studied, with only 10 (out of 5,400) discordant typings for RBC antigens and none for platelet antigens.
After the second round of improvements, the final algorithm was 99.2% concordant with serological methods in a validation set of 200 genomes from the INTERVAL study. When adjusted for the depth of coverage, the final algorithm was 100% concordant with serology in typing for ABO and D antigens and 99.5% accurate at typing the C antigen.
The researchers say that this approach could "(transform) the way in which safe blood products are provided to patients." Dr. Robert Flower from Australian Red Cross Blood Services, in Brisbane, who co-authored an accompanying editorial, told.
Blood group profile
"A complete blood group profile can be automatically extracted from sequencing your DNA. This tells you not just about the well-known ABO and Rh blood groups but for all 36 blood groups and the multitude of blood group antigens (over 316 now identified)," said Flower.
"This could be especially useful for patients who are dependent on transfusion to manage chronic diseases," he said. "This would (also) be of benefit for cancer patients and hemoglobinopathy patients (sickle cell patients, thalassemia patients, etc.)."
"For variants for which there are no commercially available typing sera, genetics and allele matching is the only way to identify compatible blood for transfusion," Dr. Lane said. "Patients who are already having whole-genome sequencing performed for other diagnoses may also benefit (or their clinicians may benefit) by having a comprehensive blood group genotype available in case of future need."