Back

Calling for diversity: improving transfusion safety through high-throughput blood group microarray genotyping

Wittig, M.; Steiert, T. A.; ElAbd, H.; Degenhardt, F.; Valenti, L.; Prati, D.; Ronzoni, L.; Bujanda, L.; Banales, J. M.; Blay, N.; Invernizzi, P.; Buti, M.; Albillos, A.; Fernandez, J.; Sacchi, N.; Julia, A.; Latiano, A.; de Cid, R.; D'Amato, M.; Asselta, R.; Laudes, M.; Lieb, W.; Juhl, D.; Gassner, C.; Franke, A.

2023-12-18 hematology
10.1101/2023.12.15.23299980 medRxiv
Show abstract

Blood transfusions, conducted between donors compatible in their red blood cell (RBC) antigens, play a life-saving role in transfusion medicine. Genetic differences at blood group loci between ethnicities result in diversity and altered frequency of RBC antigens that need to be considered in blood transfusion. Consequently, comprehensive, and accurate blood group antigen typing is especially relevant for inter-ethnic blood transfusions and for minorities underrepresented in the donor population. Blood group microarray genotyping is a cost-efficient and scalable method for comprehensive blood group typing. Previously, however, microarray typing has been challenging for the clinically important blood group systems Rh and MNS, as these feature highly paralogous genomic loci leading to mixed signals. We here present an approach for accurately typing blood group systems, including Rh and MNS variations, that we benchmarked in an ethnically diverse cohort. We tested its performance using gold-standard, diagnostic-grade MALDI-TOF data from 1,052-samples, including 334 CEPH diversity samples and applied the approach to 4,999 samples of a COVID-19 genetics study. Overall, we obtained a 99.95% benchmarking concordance and 99.65% call rate. In summary, we provide a highly accurate and cost-efficient high-throughput genotyping method for comprehensive blood group analysis that is also suitable for ethnically diverse sample sets.

Published in Genomics, Proteomics & Bioinformatics · not in our set (fewer than 10 published preprints to learn from) · training set

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.