Search

Since their inception, the International HLA & Immunogenetics Workshops (IHIW) served as a collaborative platform for exchange of specimens, reference materials, experiences and best practices. In this report we present a subset of the results of human leukocyte antigen (HLA) haplotypes in families tested by next generation sequencing (NGS) under the 17th IHIW. We characterized 961 haplotypes in 921 subjects belonging to 250 families from 8 countries (Argentina, Austria, Egypt, Jamaica, Germany, Greece, Kuwait, and Switzerland). These samples were tested in a single core laboratory in a high throughput fashion using 6 different reagents/software platforms. Families tested included patients evaluated clinically as transplant recipients (kidney and hematopoietic cell transplant) and their respective family members. We identified 486 HLA alleles at the following loci HLA-A, -B, -C, -DRB1, -DRB3, -DRB4, -DRB5, -DQA1, -DQB1, -DPA1, -DPB1 (77, 115, 68, 69, 10, 6, 4, 44, 31, 20 and 42 alleles, respectively). We also identified nine novel alleles with polymorphisms in coding regions. This approach of testing samples from multiple laboratories across the world in different stages of technology implementation in a single core laboratory may be useful for future international workshops. Although data presented may not be reflective of allele and haplotype frequencies in the countries to which the families belong, they represent an extensive collection of 3rd and 4th field resolution level 11-locus haplotype associations of 486 alleles identified in families from 8 countries.

HLA haplotype mismatches have been associated with an elevated risk of acute graft-versus-host disease (aGVHD) in patients undergoing HLA-matched unrelated donor (URD) hematopoietic cell transplantation (HCT). The gamma block (GB) is located in the central MHC region between beta and delta blocks (encoding HLA-B and -C and HLA-DQ and -DR antigens, respectively) and contains numerous inflammatory and immune regulatory genes, including Bf, C2, and C4 genes. A single-center study showed that mismatches in SNPs c.2918+98G, c.3316C, and c.4385C in the GB block (C4 SNPs) were associated with higher risk of grade III-IV aGVHD. We investigated the association of GB SNP (GBS) mismatches with outcomes after 10/10 and 9/10 URD HCT (n = 714). The primary outcome was acute GVHD. Overall survival, disease-free survival, transplantation-related mortality, relapse, chronic GVHD, and engraftment were also analyzed. DNA samples were GBS genotyped by identifying 338 SNPs across 20 kb using the Illumina NGS platform. The overall 100-day incidence of aGVHD grade II-IV and II-IV were 41% and 17%, respectively. The overall incidence of matching at all GBSs tested and at the C4 SNPs were 23% and 81%, respectively. Neither being matched across all GB SNPs tested (versus mismatched) nor having a higher number of GBS mismatches was associated with transplantation outcomes. There was no association between C4 SNP mismatches and outcomes except for an unexpected significant association between having 2 C4 SNP mismatches and a higher hazard ratio (HR) for relapse (association seen in 15 patients only; HR, 3.38, 95% confidence interval, 1.75 to 6.53; P = .0003). These data do not support the hypothesis that mismatching at GB is associated with outcomes after HCT.

Despite its demonstrated importance in hematopoietic cell transplantation, the HLA-DPB1 locus is only typed in one in five unrelated donors in the United States. Addressing this issue, we developed a DPB1 Prediction Service that leverages seven-locus haplotype frequencies (HLA-A ∼ C ∼ B ∼ DRB3/4/5 ∼ DRB1 ∼ DQB1 ∼ DPB1) to extend the imputation of six-locus HLA typing (HLA-A ∼ C ∼ B ∼ DRB3/4/5 ∼ DRB1 ∼ DQB1) to the HLA-DPB1 locus, including the novel prediction of HLA-DPB1 TCE groups to calculate donor-recipient TCE permissive match probabilities. Simulations of current-day patient searches reveal the service can fill in missing gaps for another four in five donors that appears on lists. To validate its performance, samples of 206,328 registered donors and 5,218 donor-recipient pairs with known high-resolution HLA-DPB1 typing were used for predicted-versus-observed comparisons. These comparisons demonstrated that the predictions were correct for 11.9–19.7% of HLA-DPB1 genotypes, 64.9–70.0% of TCE groups, and 61.0% of permissive match categories. Although HLA-DPB1 match predictions must be confirmed by additional typing, knowledge of TCE match probabilities facilitates rapid and improved identification of best donor options, especially for populations of color. Thus, we developed the TCE Prediction Tool user interface for a pilot program with several transplant centers to preview the accuracy and utility of this prediction framework, which provides valuable upfront optimization of donor selection.

There is long-standing tension regarding whether and how to use race or geographic ancestry in biomedical research. We examined multiple self-reported measures of race and ancestry from a cohort of over 100,000 U.S. residents alongside genetic data. We found that these measures are often non-overlapping, and that no single self-reported measure alone provides a better fit to genetic ancestry than a combination including both race and geographic ancestry. We also found that patterns of reporting for race and ancestry appear to be influenced by participation in direct-to-consumer genetic ancestry testing. Our results demonstrate that there is a place for the language of both race and geographic ancestry as we seek to empower individuals to fully describe their family history in research and medicine.One Sentence SummarySelf-identification in the United States according to both racial and geographic terms best reflects genetic ancestry in individuals.

To optimize strategies that mitigate the risk of graft loss associated with HLA incompatibility, we evaluated whether sequence defined HLA targets (eplets) that result in donor-specific antibodies are associated with transplant outcomes. To define this, we fit multivariable Cox proportional hazard models in a cohort of 118 382 United States first kidney transplant recipients to assess risk of death-censored graft failure by increments of ten antibody-verified eplet mismatches. To verify robustness of our findings, we conducted sensitivity analysis in this United States cohort and assessed the role of antibody-verified eplet mismatches as autonomous predictors of transplant glomerulopathy in an independent Canadian cohort. Antibody-verified eplet mismatches were found to be independent predictors of death-censored graft failure with hazard ratios of 1.231 [95% confidence interval 1.195, 1. 268], 1.268 [1.231, 1.305] and 1.411 [1.331, 1.495] for Class I (HLA-A, B, and C), -DRB1 and -DQB1 loci, respectively. To address linkage disequilibrium between HLA-DRB1 and -DQB1, we fit models in a subcohort without HLA-DQB1 eplet mismatches and found hazard ratios for death-censored graft failure of 1.384 [1.293, 1.480] for each additional antibody-verified HLA-DRB1 eplet mismatch. In a subcohort without HLA-DRB1 mismatches, the hazard ratio was 1.384 [1.072, 1.791] for each additional HLA-DQB1 mismatch. In the Canadian cohort, antibody-verified eplet mismatches were independent predictors of transplant glomerulopathy with hazard ratios of 5.511 [1.442, 21.080] for HLA-DRB1 and 3.640 [1.574, 8.416] for -DRB1/3/4/5. Thus, donor-recipient matching for specific HLA eplets appears to be a feasible and clinically justifiable strategy to mitigate risk of graft loss.

Since their inception, the International HLAImmunogenetics Workshops (IHIW) served as a collaborative platform for exchange of specimens, reference materials, experiences and best practices. In this report we present a subset of the results of human leukocyte antigen (HLA) haplotypes in families tested by next generation sequencing (NGS) under the 17th IHIW. We characterized 961 haplotypes in 921 subjects belonging to 250 families from 8 countries (Argentina, Austria, Egypt, Jamaica, Germany, Greece, Kuwait, and Switzerland). These samples were tested in a single core laboratory in a high throughput fashion using 6 different reagents/software platforms. Families tested included patients evaluated clinically as transplant recipients (kidney and hematopoietic cell transplant) and their respective family members. We identified 486 HLA alleles at the following loci HLA-A, -B, -C, -DRB1, -DRB3, -DRB4, -DRB5, -DQA1, -DQB1, -DPA1, -DPB1 (77, 115, 68, 69, 10, 6, 4, 44, 31, 20 and 42 alleles, respectively). We also identified nine novel alleles with polymorphisms in coding regions. This approach of testing samples from multiple laboratories across the world in different stages of technology implementation in a single core laboratory may be useful for future international workshops. Although data presented may not be reflective of allele and haplotype frequencies in the countries to which the families belong, they represent an extensive collection of 3rd and 4th field resolution level 11-locus haplotype associations of 486 alleles identified in families from 8 countries.

We have investigated HLA population alleles and haplotype frequencies for the ethnicities that comprise the contemporary population of Israel, using a large data set from the Ezer Mizion Bone Barrow Donor Registry. We genotyped 275,699 individuals at the HLA-A, -B and -DRB1 loci using HLA genotyping methods. HLA A∼B∼DRB1 haplotype frequencies were estimated from 19 sub-ethnic Jewish populations and other non-Jewish minorities using the maximum likelihood model, which accommodates typing ambiguities. We present overall and sub-ethnicity specific HLA diversity results of the registry, which will help guide a data-driven strategy for future registry expansion.

The search for a suitable human leukocyte antigen (HLA)–matched unrelated adult stem cell donor (URD) or umbilical cord blood unit (UCB) is a complex process. The National Marrow Donor Program (NMDP) developed a search algorithm known as HapLogic, which is currently provided within the NMDP Traxis application. The HapLogic algorithm has been in use since 2006 and has advanced URD/UCB HLA-matching technology. The algorithm has been shown to have high predictive accuracy, which can streamline URD/UCB selection and drive efficiencies in the search process to the benefit of the stem cell transplantation community. Here, we describe the fundamental components of the NMDP matching algorithm, output, validation, and future directions.

Histocompatibility testing, and HLA antibody screening in particular, varies in practice among laboratories. Currently, standards are lacking regarding the reporting of testing methods in publications. It is essential that scientific methods are rigorously and transparently described upon publication, so that results can be accurately interpreted and independently corroborated. Additionally, this would allow work groups to compile diverse data to achieve clinically significant conclusions from meta-analyses. These efforts are hindered when there is a paucity of method descriptions and where variability in serum treatment, protocol modifications, and assay thresholds affecting assay interpretation are known to exist. Thus, the ASHI Science and Technology Initiatives Committee (ASHI STIC) undertook the task of formulating recommendations for reporting HLA antibody testing by solid phase assays, and the associated HLA typing required for interpretation, in scientific publications. Herein we put forth standards for minimum information about HLA antibody testing methods reported in histocompatibility publications.

Five locus allele-level HLA-A, -B, -C, -DRB1, -DQB1 allele and haplotype frequencies have been calculated for almost 29,000 people from three Arab populations that live in Israel and were recruited as donors to the Hadassah bone marrow donor registry. These groups are of Muslim, Christian and Bedouin Arab descent which represent more than 90% of the Arabs that live in Israel. The goal of the study was to describe the HLA genetic profiles of the Hadassah Arab registry donors and investigate the utility of these donors for the local and international hematopoietic stem-cell transplant community. The results demonstrate that the analyzed Arab populations share at least seven of the top ten most frequent alleles. Comparison with other populations confirmed the proximity of the three Arab populations to each other and to the Be The Match® Middle Eastern population. Despite these similarities, some alleles are private to each of the three groups, possibly because of historical, environmental or societal events. Clinical data showed that Arab donors were HLA matched with Arab and international patients. This analysis indicates the value added by the Hadassah Arab donors to the local and global transplant community.

The Public Cord Blood Bank of Jeevan Stem Cell Foundation was established in 2008 to harvest cord blood units and make them available to treat multiple blood disorders through Hematopoietic Stem Cell Transplants. We studied Human Leucocyte Antigen (HLA)-A, -B, -C, -DRB1 and -DQB1 allele and haplotype diversity in a sample of 2491 unrelated cord-blood units from Jeevan’s Public Cord Blood Bank (part of Be The Cure Registry) in the Tamil Nadu state in the Indian Peninsula.

The Human Leukocyte Antigen (HLA) is the most polymorphic region in humans. Anthropologists use HLA to trace populations’ migration and evolution. However, recent admixture between populations masks the ancestral haplotype frequency distribution.We present an HLA-based method based on high-resolution HLA haplotype frequencies to resolve population admixture using a non-negative matrix factorization formalism and validated using haplotype frequencies from 56 populations. The result is a minimal set of original populations decoding roughly 90% of the total variance in the studied admixtures. These original populations agree with the geographical distribution, phylogenies and recent admixture events of the studied groups.With the growing population of multi-ethnic individuals, the matching process for stem-cell and solid organ transplants is becoming more challenging. The presented algorithm provides a framework that facilitates the breakdown of highly admixed populations into original groups, which can be used to better match the rapidly growing population of multi-ethnic individuals worldwide.Author SummaryHuman Leukocyte Antigen (HLA) is known to be the most polymorphic region in the human genome. Anthropologists frequently use HLA to trace migration and evolution of different populations. This is due to the high linkage among HLA genes leading to the transmission of intact haplotypes from parents to offspring, hence preserving key population ancestral features.We developed a new HLA-based method to identify admixture models in mixed populations using high-resolution HLA haplotype frequencies. Our results highlight that a single highly polymorphic locus can contain enough information to map clearly human admixture and the population genetics of the different human populations, and reproduces results based on SNP arrays.The presented algorithm is validated using haplotype frequencies sampled from 56 worldwide populations. Under such factorization we demonstrate that 90% of the variance in these populations can be explained using a much-reduced set of 8 ethnic groups. We demonstrate that the estimated ethnic groups and admixture models agree with the geographical distribution, population phylogenies and recent historic admixture events of the studied populations.

Disparities in survival after allogeneic hematopoietic cell transplantation have been reported for some race and ethnic groups, despite comparable HLA matching. Individuals' ethnic and race groups, as reported through self-identification, can change over time because of multiple sociological factors. We studied the effect of 2 measures of genetic similarity in 1378 recipients who underwent myeloablative first allogeneic hematopoietic cell transplantation between 1995 and 2011 and their unrelated 10 of 10 HLA-A, -B, -C, -DRB1, and-DQB1- matched donors. The studied factors were as follows (1) donor and recipient genetic ancestral admixture and (2) pairwise donor/recipient genetic distance. Increased African genetic admixture for either transplant recipients or donors was associated with increased risk of overall mortality (hazard ratio [HR], 2.26; P = .005 and HR, 3.09; P = .0002, respectively) and transplant-related mortality (HR, 3.3; P = .0003 and HR, 3.86; P = .0001, respectively) and decreased disease-free survival (HR, 1.9; P = .02 and HR, 2.46; P = .002 respectively). The observed effect, albeit statistically significant, was relevant to a small subset of the studied population and was notably correlated with self-reported African-American race. We were not able to control for other nongenetic factors, such as access to health care or other socioeconomic factors; however, the results suggest the influence of a genetic driver. Our findings confirm what has been previously reported for African-American recipients and show similar results for donors. No significant association was found with donor/recipient genetic distance.

We have calculated six-locus high resolution HLA A∼C∼B∼DRB3/4/5∼DRB1∼DQB1 haplotype frequencies using all Be The Match® Registry volunteer donors typed by DNA methods at recruitment. Mixed resolution HLA typing data was inputted to a modified expectation–maximization (EM) algorithm in the form of genotype lists generated by interpretation of primary genomic typing data to the IMGT/HLA v3.4.0 allele list. The full cohort consists of 6.59 million subjects categorized at a broad race level. Overall 25.8% of the individuals were typed at the C locus, and 5.2% typed at the DQB1 locus, while all individuals were typed for A, B, DRB1. We also present a subset of 2.90 million subjects with detailed race/ethnic information mapped to 21 population subgroups, 64.1% of which have primary DNA typing data across at least A, B, and DRB1 loci. Sample sizes at the detailed race level range from 1,242,890 for European Caucasian to 1,376 Alaskan Native or Aleut. Genetic distance measurements show high levels of HLA genetic divergence among the 21 detailed race categories, especially among the eight Asian–American populations. These haplotype frequencies will be used to improve match predictions for donor selection algorithms for hematopoietic stem cell transplantation and improve the accuracy in modeling registry match rates.

Estimation of human leukocyte antigen (HLA) haplotype frequencies from unrelatedstem cell donor registries presents a challenge because of large sample sizes and het-erogeneity of HLA typing data. For the 14th International HLA and ImmunogeneticsWorkshop, five bioinformatics groups initiated the ‘Registry Diversity Component’aiming to cross-validate and improve current haplotype estimation tools. Five datasetswere derived from different donor registries and then used as input for five differentcomputer programs for haplotype frequency estimation. Because of issues related toheterogeneity and complexity of HLA typing data identified in the initial phase, thesame five implementations, and two new ones, were used on simulated datasets in acontrolled experiment where the correct results were known apriori . These datasetscontained various fractions of missing HLA-DR modeled after European haplotypefrequencies. We measured the contribution of sampling fluctuation and estimationerror to the deviation of the frequencies from their true values, finding equivalentcontributions of each for the chosen samples. Because of patient-directed activities,selective prospective typing strategies and the variety and evolution of typing technol-ogy, some donors have more complete and better HLA data. In this setting, we showthat restricting estimation to fully typed individuals introduces biases that could beovercome by including all donors in frequency estimation. Our study underlines theimportance of critical review and validation of tools in registry-related activity andprovides a sustainable framework for validating the computational tools used. Accu-rate frequencies are essential for match prediction to improve registry operations andto help more patients identify suitably matched donors.IntroductionBecause of their large ethnic and geographic variation, humanleukocyte antigen (HLA) haplotype frequencies (HF) areof particular importance in the procurement of unrelatedhematopoietic stem cell donors. Haplotype frequency estima-tion (HFE) algorithms have beenavailable for decades (1–10);however, the data provided by unrelated donor registries

HLA haplotype frequencies in a volunteer bone marrow donor registry should reflect the frequencies of potential transplant recipients served by that registry, a challenge in a country with diverse subethnicities of immigrants from Eastern and Western cultures, such as Israel. We evaluated the likelihood of finding suitable donors for hypothetical patients drawn from defined subethnicities in the Ezer Mizion Bone Marrow Donor Registry (EM BMDR) from donors both within and outside the registry now and during the coming decade. On average, bioinformatics modeling predicts that, given current donor recruitment trends, 6/6 high-resolution HLA match rates for Israelis, which currently stand at 40% to 55% for most subethnicities, will rise by up to 1% per year over the next decade. Subethnicities with historically lower rates of interethnic admixture are less likely to find matches outside of their designated group but will benefit from expansion of the registry, whereas ethnically directed drives will enhance matching rates for currently underrepresented subethnicities. Donor searches for the same cohort using a large extramural registry was of only slight benefit for most of the 19 EM BMDR subethnicities evaluated, confirming that local donor registries that reflect the ethnic diversity of the community being served are best equipped to serve the needs of their respective communities. Contemporary trends of an increasingly multiethnic admixture in Israel may impact the effect of ethnic profiling in assessing future match rates for EM BMDR.

This project has the goal to validate bioinformatics methods and tools for HLA haplotype frequency analysis specifically addressing unique issues of haematopoietic stem cell registry data sets. In addition to generating new methods and tools for the analysis of registry data sets, the intent is to produce a comprehensive analysis of HLA data from 20 million donors from the Bone Marrow Donors Worldwide (BMDW) database. This report summarizes the activity on this project as of the 16IHIW meeting in Liverpool.

Aim Genetic recombination occurs at specific hotspots within the MHC, leading to widely dispersed block-like structures; the alpha block containing HLA-A, beta block (HLA-C and HLA-B) and delta block (HLA-DR and HLA-DQ). Linkage disequilibrium (LD) is much stronger within than across these blocks. The HLA-A locus (most telomeric) has weaker LD with extended HLA haplotypes. Additionally, a recombination hotspot exists between HLA-DP genes and other class II loci, which leads to an HLA-DPB1 match rate of only 10–15% in 10/10 HLA-matched donor/recipient pairs. This report describes an extended 9 locus HLA haplotype found in 2 HCT and 1 kidney transplant recipients and their potential donors (8, 2 and 1 donors, respectively). Methods 1. Case review of 2 unrelated HCT recipients, both European Caucasian (CAU) and their related and unrelated potential donors. 2. Family study of HCT recipient and a sibling to identify the haplotype by segregation. 3. Determine frequency of observed haplotype in different populations through Haplostats.org (6 locus at 2 field resolution could yield higher estimate than the 9 locus 3 field resolution haplotype described here). 4. Searching lab database for incidence of this haplotype. Results All identified cases shared the following haplotype HLA-A*31:01:02∼C*07:02:01∼B*07:02:01∼DRB5*01:01:01∼DRB1*15:01:01∼DQA1*01:02:01∼DQB1*06:02:01∼DPA1*01:03:01∼DPB1*04:01:01. Although most alleles in this haplotype are fairly common, with HLA-A*31:01 least common (∼5%), Haplostats indicated that the haplotype HLA-A*31:01∼C*07:02∼B*07:02∼DRB5*01:01∼DRB1*15:01∼DQB1*06:02 ranks 61898th & 170477th in CAU & African Americans (AFA), respectively and is not seen in other populations. Searching our laboratory database for subjects A31 positive, 382 subjects out of 8340 typed between 2015 & 2017 were identified. Of those, 40 were positive for the haplotype HLA-A*31:01∼C*07:02∼B*07:02∼DRB5*01:01∼DRB1*15:01∼DQB1*06:02 and of these 17 were typed for HLA-DPB1. All were positive for the entire extended 9 locus haplotype and were either CAU (n = 6) or of unknown ethnicity (n = 11). Conclusions Although the 6 locus haplotype portion of this extended haplotype seems fairly uncommon, the extended 9 locus subset of that haplotype seems to be conserved especially in Caucasians.

Effective communication between clinicians is a crucial component of safe care. High-quality communication may be especially critical during care transitions between clinicians (handoffs). In a two-year quasi-experimental Quality Improvement project, we are using simulation learning to evaluate and improve communication between anesthesia providers and nurses as care is transitioned from the operating room to the post-anesthesia care unit (PACU). Using a multiple baseline, staggered entry, prospective cohort design with repeated measures, we are introducing a training intervention and evaluating its effects on dyadic communication. The primary hypothesis is that simulation-based communication skills training of PACU personnel will significantly improve the quality of handoffs. Clinicians' performance is being compared before and after their training on four dimensions of handoff effectiveness: information transfer; use of best evidence handoff strategies; interpersonal skills; and team behaviors. Clinicians also rate their own handoffs. Lessons learned and results to date will be presented.

Aim Knowledge of allele and haplotype frequencies in human populations guides the search for an HLA- matched unrelated hematopoietic stem cell donor. A new algorithm developed by the National Marrow Donor program uses this information to predict the likelihood that a volunteer of a particular ethnicity will carry specific HLA alleles when typed at high resolution. In the present study, we determined the HLA-A,-B and -DRB1 alleles and HLA-A-B haplotype frequencies in Egyptians and compared them to various populations. Methods Specificities of HLA-A, -B and -DRB1 loci were determined in 286 matched related donors of allogeneic hematopoeitic stem cell transplantation (AHSCT). Typing was performed using serologic and low resolution PCR- reverse sequence-specific oligonucleotide (rSSO) techniques. HLA-A-B haplotypes were defined by segregation analysis. Results The most frequent alleles for HLA class I HLA-A were: HLA-A2 (19.2%), A1 (15.8%), A30 (9.2%) and A24 (8.8%), for HLA-B, B41 (10.9%), B35 (10.2%), B44 (6%) and B50 (5.4%) and for DRB1∗13 (18.7%), ∗04 (15%) ,∗11(14.7%) and ∗03 (13.2%). The most frequent HLA-A-B haplotypes were: A24-B35 (3.8%), A2-B41 (3.4%), A1- B35 (3.4%) and A33-B14 (3.4%). Conclusions This study report HLA allele and HLA-A-B haplotype patterns in Egyptians. These patterns were close to Jordanians and Lebanese and different from Saudi Arabians, Europeans, Americans and Asians. These data explain the difficulty of finding donors in world registries for Egyptian patients in need of a stem-cell transplant and highlight the need to for planning an Egyptian national unrelated donor registry to meet the demands of these patients.

Genetic matching for loci in the human leukocyte antigen (HLA) region between a donor and a patient in hematopoietic stem cell transplantation (HSCT) is critical to outcome; however, methods for HLA genotyping of donors in unrelated stem cell registries often yield results with allelic and phase ambiguity and/or do not query all clinically relevant loci. We present and evaluate a statistical method for in silico imputation of HLA alleles and haplotypes in large ambiguous population data from the Be The Match(®) Registry. Our method builds on haplotype frequencies estimated from registry populations and exploits patterns of linkage disequilibrium (LD) across HLA haplotypes to infer high resolution HLA assignments. We performed validation on simulated and real population data from the Registry with non-trivial ambiguity content. While real population datasets caused some predictions to deviate from expectation, validations still showed high percent recall for imputed results with average recall >76% when imputing HLA alleles from registry data. We simulated ambiguity generated by several HLA genotyping methods to evaluate the imputation performance on several levels of typing resolution. On average, imputation percent recall of allele-level HLA haplotypes was >95% for allele-level typing, >92% for intermediate resolution typing and >58% for serology (low-resolution) typing. Thus, allele-level HLA assignments can be imputed through the application of a set of statistical and population genetics inferences and with knowledge of haplotype frequencies and self-identified race and ethnicities.

In hematopoietic stem cell transplantation, donor selection is based primarily on matching donor and patient HLA genes. These genes are highly polymorphic and their typing can result in exact allele assignment at each gene (the resolution at which patients and donors are matched), but it can also result in a set of ambiguous assignments, depending on the typing methodology used. To facilitate rapid identification of matched donors, registries employ statistical algorithms to infer HLA alleles from ambiguous genotypes. Linkage disequilibrium information encapsulated in haplotype frequencies is used to facilitate prediction of the most likely haplotype assignment. An HLA typing with less ambiguity produces fewer high-probability haplotypes and a more reliable prediction. We estimated ambiguity for several HLA typing methods across four continental populations using an information theory-based measure, Shannon's entropy. We used allele and haplotype frequencies to calculate entropy for different sets of 1,000 subjects with simulated HLA typing. Using allele frequencies we calculated an average entropy in Caucasians of 1.65 for serology, 1.06 for allele family level, 0.49 for a 2002-era SSO kit, and 0.076 for single-pass SBT. When using haplotype frequencies in entropy calculations, we found average entropies of 0.72 for serology, 0.73 for allele family level, 0.05 for SSO, and 0.002 for single-pass SBT. Application of haplotype frequencies further reduces HLA typing ambiguity. We also estimated expected confirmatory typing mismatch rates for simulated subjects. In a hypothetical registry with all donors typed using the same method, the entropy values based on haplotype frequencies correspond to confirmatory typing mismatch rates of 1.31% for SSO versus only 0.08% for SBT. Intermediate-resolution single-pass SBT contains the least ambiguity of the methods we evaluated and therefore the most certainty in allele prediction. The presented measure objectively evaluates HLA typing methods and can help define acceptable HLA typing for donor recruitment.

Aim Recent data show that DP matching plays an important role in hematopoietic cell transplantation. Our previous work demonstrated that in European Americans haplotypic associations for the DP heterodimer are determined by near complete linkage disequilibrium between amino acid motifs on DPA1 (pos 31) and DPB1 (pos 85–87). We aimed to generate motif and allele-level haplotype frequencies (HF) for four race groups to extend these insights to minorities. Methods We calculated DPA1∼DPB1 HF at allele and motif level from 16,802 Be the Match Registry® members typed by DNA methods, resolving phase, allelic and motif ambiguity using expectation maximization. Motif frequencies were calculated for position 31 on DPA1 and positions 85–87 on DPB1, which play a critical role in the P1 pocket of the peptide binding region. Normalized linkage disequilibrium (LD) was calculated on each haplotype (D′) and globally (Wn) for each population. Results The top three allele-level haplotypes in all populations account for over 66% of the HF, with DPA1∗01:03 ∼ DPB1∗04:01 most common in the Caucasian (CAU), Asian Pacific Islander (API) and Hispanic (HIS) populations and DPA1∗02:02 ∼ DPB1∗01:01 in the African American (AFA) population. Wn for DPA1∼DPB1 allele associations for CAU, HIS, API, and AFA populations was .55, .45, .49, and .52 respectively. At the motif level there were only four possible haplotypes. M ∼ GPM and Q ∼ EAV were the two most common in every population. M ∼ GPM was the most common in CAU, HIS and API with frequencies 69%, 68% and 56% respectively, with D′ = 0.97 for CAU (complete LD). The most common haplotype for AFA was Q ∼ EAV, at 48% and D′ = .99. Conclusions Low global LD was observed at the allele level for all populations. However, at the motif level haplotypes were in near complete LD suggesting that for the DPA1 ∼ DPB1 heterodimer the unit of selection for all populations is the combined amino acid motif rather than the allele, as previously observed in European Americans.

Aim A Neural network (NN) to predict serologic specificity of HLA alleles was first introduced in 2003 by Maiers et al. The procedure involved several manual steps and prediction errors were estimated from an arbitrarily selected single partition of available labeled data. However, these might not reflect the true predictive capability of the model. Our aim was to provide an automated framework for the existing NN, estimate the true prediction accuracy and capability using a cross-validation model and compare with other predictive machine learning techniques. Methods An NN framework was created and validated in the R software suite to predict serologic specificities for HLA-A, B, C, DQB1, DRB1 and DPB1. We established a method to tune NN parameters using a grid technique with leave one out cross validation and compared the prediction accuracy of the NN with a k-Nearest Neighbors (kNN) technique. Results A concordance between the manual and automated NN of upwards of 96% was obtained for all loci, measured on an independent unlabelled test dataset. With parameter tuning techniques for the NN we have estimated the average prediction accuracy across all loci to be 86% .We also observed that kNN provided better prediction accuracy of 90% (average). These numbers were slightly lower than the previous NN validation based on a single partition of the data, but still provide acceptable performance for both the NN and kNN. Conclusions Human involvement in NN model selection is both extremely time consuming and prone to errors. An automated framework minimizes the risks of over-training by monitoring variations in errors. Cross-validation techniques are known to provide better generalization error estimates. In this context, our reported error estimates maybe closer to the actual error rates one can achieve with similar models. We also demonstrate kNN outperforms the existing NN. This result should support efforts in finding relationships between alleles and associated serologic specificities.

Aim The Arab Bone Marrow Registry in Hadassah Medical Center collected HLA typings for 18,000 Arab adult donors. We used this information to estimate population allele and haplotype frequencies (HFs). Methods We calculated HLA A ∼ B ∼ DRB1 HFs from 1,541 Christian Arab (CA) and 16,129 Muslim Arab (MA) donors typed by SBT, resolving phase and allelic ambiguity using Expectation Maximization. HFs were estimated at the four-digit level. Results The most common CA haplotype was A∗33:01 ∼ B∗14:02 ∼ DRB1∗01:02 at a frequency of 1.6%, which is the fifth most common MA haplotype at 0.9%. This haplotype was the second ranked in the US Middle Eastern/North Coast of Africa (MENAFC) at 0.8% and is also a common Jewish haplotype. The most common MA haplotype was A∗01:01 ∼ B∗52:01 ∼ DRB1∗15:02 at 1.4%, which is ranked 41st in CA at 0.04%. This haplotype was ranked 12th in MENAFC at 0.26%. The fourth ranked haplotype for both CA and MA, A∗02:05 ∼ B∗50:01 ∼ DRB1∗07:01, is the most common in Saudi Arabia, Kuwait and Qatar and the fourth most common among Jewish populations. Average homozygosity for CA and MA for HLA-A, -B and -DRB1 was estimated at 6.7% and 6.2% while the top 30 haplotypes had a cumulative frequency of 25% and 19% respectively indicating higher MA genetic diversity. Interestingly, eight DRB1 alleles: 01:02, 03:01, 04:03, 07:01, 11:01, 11:04, 13:03 and 15:02 constituted over 63% of the allele frequencies for both CA and MA while 97% and 30% of patients served by the Arab registry had one or two of these alleles respectively. Additionally, DRB1∗03:01 occurred without the common Caucasian B∗08:01 association in about 85% of CA and MA donors. Clinical data showed that 40% of transplants were for Arab patients while others were for patients worldwide. Conclusions Comparing CA and MA to other published HFs, it is clear that some haplotypes have migrated across populations while others remain isolated, thus adding these donors to the world registry is vital to increase genetic diversity for the benefits of all patients.

Aim The HLA haplotype frequencies for Middle Eastern populations are not well known relative to the rest of the world. At The King Faisal Specialist Hospital and Research Center in Riyadh we have collected HLA typing information for thousands of donors and cord blood units from years of stem-cell and renal transplants with the goal of establishing a stem cell registry in the Kingdom of Saudi Arabia. We aim in this study to use these HLA typings to estimate Saudi population haplotype frequencies and in follow-on work to project HLA match rates for the Saudi population as the registry expands. Methods We calculated HLA A∼C∼B∼DRB1∼DQB1 haplotype frequencies from 2,723 Saudi donors and cord blood units typed by DNA methods, resolving phase and allelic ambiguity using the Expectation Maximization algorithm. Haplotype frequencies were estimated to the allele family level for class I loci and to high resolution at class II loci. Results The most common Saudi HLA haplotype was A∗02∼C∗06∼B∗50∼DRB1∗07:01∼DQB1∗02:02 at a frequency of 4.5%. This haplotype is also the 32nd most common haplotype in the US Caucasian population at a frequency of 0.38%. The 2nd most common Saudi haplotype was A∗02∼C∗07∼B∗07∼DRB1∗15:01∼DQB1∗06:02 at 1.7% frequency, which is also the 4th most common haplotype among US Caucasians, and within Europe is most common in Scandinavia. The 3rd most common Saudi haplotype, A∗26∼C∗07∼B∗08∼DRB1∗03:01∼DQB1∗02:01 at 1.6% frequency, is the 14th most common haplotype in US Asian Pacific Islanders at 0.64%, but is quite rare in US Caucasians (rank 466; 0.0036%). Conclusions his pilot haplotype frequency study and the match rate projection to follow will help to form a data-driven approach to the establishment of a robust stem cell registry of adult donors and cord blood units that will aid the Saudi population in need of transplant. Collection of more detailed ancestry information on Saudi donors has the potential to improve match estimates and ensure all subpopulations are well-served.

Aim A ‘Silver Standard’ for HLA data collection and reporting has been described at ImmPort (immport.niaid.nih.gov, “Proposal for HLA Data Validation”) to address ambiguity resolution in the recording and reporting of HLA typing results. While standards are critical for HLA data interoperability, they are not meaningful until useful tools are developed and made available for community use. We are developing distributable tools that implement this silver standard. Here we describe the development a web service to create, update, and retrieve HLA typing data in standardized formats without the need for NMDP allele codes and the corresponding inherent introduction of new ambiguities. Methods ReST web services with HTTP negotiation are being developed employing a Java library that manages HLA typing data using standardized formats. These formats include the XML based Histoimmunogenetics Markup Language (HML) and a simple character-delimited string format (GL String) able to encode ambiguity within HLA typing. Resources are identified with a simple Uniform Resource Identifier (URI). Results The services build on a foundation of an open access database schema for IMGT/HLA reference sequence data (updated quarterly), and objects such as alleles, lists of alleles, haplotypes, genotypes, lists of genotypes and multi-locus unphased genotypes. Public services include creating, updating, and retrieving these objects. Content negotiation allows data retrieval in a variety of formats including GL String, HML, HTML, JSON, and QR Code. Conclusions The tools being developed here provide the HLA researcher, clinician and lab technician a common resource for managing HLA data in a standardized way. We envision these tools to augment workflows through creating new instances of HLA typing objects when needed, and retrieval of those objects and their associated metadata when called upon.

Many panels of ancestry informative single nucleotide polymorphisms have been proposed in recent years for various purposes including detecting stratification in biomedical studies and determining an individual's ancestry in a forensic context. All of the panels have limitations in their generality and efficiency for routine forensic work. Some panels have used only a few populations to validate them. Some panels are based on very large numbers of SNPs thereby limiting the ability of others to test different populations. We have been working toward an efficient and globally useful panel of ancestry informative markers that is comprised of a small number of highly informative SNPs. We have developed a panel of 55 SNPs analyzed on 73 populations from around the world. We present the details of the panel and discuss its strengths and limitations.

Here, we present results for DPA1 and DPB1 four-digit allele-level typing in a large (n = 5,944) sample of unrelated European American stem cell donors previously characterized for other class I and class II loci. Examination of genetic data for both chains of the DP heterodimer in the largest cohort to date, at the amino acid epitope, allele, genotype, and haplotype level, allows new insights into the functional units of selection and association for the DP heterodimer. The data in this study suggest that for the DPA1-DPB1 heterodimer, the unit of selection is the combined amino acid epitope contributed by both the DPA1 and DPB1 genes, rather than the allele, and that patterns of LD are driven primarily by dimer stability and conformation of the P1 pocket. This may help explain the differential pattern of allele frequency distribution observed for this locus relative to the other class II loci. These findings further support the notion that allele-level associations in disease and transplantation may not be the most important unit of analysis, and that they should be considered instead in the molecular context. Electronic supplementary material The online version of this article (doi:10.1007/s00251-012-0615-3) contains supplementary material, which is available to authorized users.