Aim DNA-based registry typing is commonly done by SSOP and only a few registries employ sequencing at the time of recruitment, allowing to understand allele and haplotype frequencies and evaluating the size and composition of the registry, to assess the likelihood of finding matched donors. We explored the impact of the increasing number of HLA alleles in the genetic diversity of Mexicans, which will in turn impact the possibility of finding a donor for patients with a similar ethnic background. Methods Healthy Mexicans enrolled at the DONORMO-The Mexican Unrelated Bone Marrow Donor Registry, born in different geographic areas of Mexico were HLA typed at the allele level, for A∗, B∗, C∗, DRB1∗, DQB1∗ loci. DNA samples were sequenced in an ABI Prism 310 or in a 3500 sequencer. The SPSS 12 software was used for analysis. For A locus, 127 individuals were typed; 523 for B, 81 for C; 952 were SBT for DRB1 and 3343 for DQB1. Results A total of 273 common and well documented alleles were detected; 91 were unique observations; 40 at A∗ locus; 26 at B∗; 8 at C∗; 14 at DRB1∗ & 3 at DQB1∗. 65 different A∗ alleles are present, of which 32 are A2 different variants: A∗02:01; A2:06; A2:05; (13.4–5.5%) prevail. The prevalent B∗ alleles were 49:01; 35:12; 35:17; 15:15; 14:02; 14:01;39:06; 39:05; 40:05; 40:08; 50:01; 52:01 51:02; 53:01; (>1.5%). Only 10 C∗ alleles are frequent: 04:01, 07:02; 01:02; 03:04; 07:01; 16:01, 03:06; 5:01; 08:01; 08:02; (>2.5%). 16 frequent DRB1∗ alleles are present: 04:07; 08:02; 14:06; 01:02; 04:02; 04:11; 04:04; 19:01; 14:02; 07:01: 03:01; 16:02; 01:03; 04:04; 10:01; 15:03; 15:01, 15:03 (1.8%). Unique observations were found at DQB1∗ for 06:11; 03:08; 02:04. Mediterranean, Amerindian, Spanish, Caucasian, Semitic and African influences are clear. Conclusion The molecular variation shown, and a one-step typing strategy using SBT since donor recruitment, is highly relevant to optimizing donor search algorithms in HSCT, for identification of potential matches, and in understanding the genetic background of this highly heterogeneous country. The data are also very useful for applications that depend on allele prevalence information such as: resolution of ambiguity during HLA typing, research and clinical practice, and the management of the general data, in the face of an ever growing list of recognized allele sequences.