The following links list how search determinants are assigned to each HLA allele by the different registries; The Anthony Nolan, the NMDP, BMDW, Canadian Unrelated Bone Marrow Donor Registry, France Greffe de Moelle, BBMR and also the ZKRD approach to search determinants.
Bone Marrow Donors Worldwide (BMDW)
Bone Marrow Donors Worldwide (BMDW) includes two columns for search determinants. One for validating serology and DNA, and one for assigning a SD from a DNA typing. When validating a donor with serology A26 and DNA-A*3401, you would not report this as a problem. Based on the three resources mentioned below, we have one column with the combination 26(10)/34(10)/66(10) for the allele A*3401. So a combination of either of these three antigens with this allele would be ok. When assigning a SD for a donor or patient without any serology in its history, we look in a second column. This column has a 34(10) for the A*3401. So the SD for this donor or patient would be A34.The column for validating serology and DNA is built based on the following three resources:
- Nomenclature for factors of the HLA system web site
- NMDP multiple allele code list
- WMDA HLA DNA-Serology Dictionary
The following example explains why we need two columns one for extracting search determinants, and one for validating serology (or serologic equivalents/determinants) and DNA data. Allele B*4016 has a WHO assigned equivalent of B61(40) in the WMDA HLA Dictionary. In our tables we have a B60(40) / B61(40) - meaning B60(40) or B61(40) - since we included the UCLA and NMDP information from the Dictionary, which states that in one third of the cases the B*4016 is assigned as B60. When extracting one search determinant for this allele, our software would assign B40 for B*4016 (based on B60(40) or B61(40)). When extracting a search determinant for a patient that is of course not what we want - a B61 should be assigned. For this purpose we need a separate column which is more precise. However, if a donor or patient is submitted with the combination B*4016 and B61, we would not want to reject this submission as an error, since in many cases the B*4016 is a B61. For these kinds of validations, we would need the second column.