HLA Imputation Research Articles

Profiling performance of high-resolution HLA imputation with complementary metrics and visualizations using an open-source validation framework

Profiling performance of high-resolution HLA imputation with complementary metrics and visualizations using an open-source validation framework

HLA-DQA1*01:03 and DQB1*06:01 are risk factors for severe COVID-19 pneumonia.

The clinical spectrum of COVID-19 includes a wide range of manifestations, from mild symptoms to severe pneumonia. HLA system plays a pivotal role in immune responses to infectious diseases. The purpose of our study was to investigate the association between HLA and COVID-19 severity in a Japanese population. The study included 209 Japanese COVID-19 patients aged ≥20 years. Saliva samples were collected and used to determine the HLA genotype by HLA imputation through genome-wide association analyses. The association between HLA genotype and COVID-19 severity was then evaluated. The allele frequency was compared between patients with respiratory failure (severe group: 91 cases) and those without respiratory failure (non-severe group: 118 cases), categorising the data into three time periods: pre-Omicron epidemic period, Omicron epidemic period, and total period of this study (from January 2021 to May 2023). In comparing the severe and non-severe groups, the frequencies of the HLA-DQA1*01:03 (35.1% vs. 10.5%, odds ratio [OR] = 4.57, corrected p [pc] = 0.041) and -DQB1*06:01 (32.4% vs. 7.9%, OR = 5.54, pc = 0.030) alleles were significantly higher in the severe group during the pre-Omicron epidemic period. During the Omicron epidemic period, HLA-DQB1*06 (32.4% vs. 7.9%, OR = 5.54, pc = 0.030) was significantly higher in the severe group. During total period of this study, HLA-DQA1*01:03 (30.2% vs. 14.4%, OR = 2.57, corrected pc = 0.0013) and -DQB1*06:01 (44.5% vs. 26.7%, OR = 2.20, pc = 0.013) alleles were significantly higher in the severe group. HLA-DQB1*06:01 and -DQA1*01:03 were in strong linkage disequilibrium with each other (r2 = 0.91) during total period of this study, indicating that these two alleles form a haplotype. The frequency of the HLA-DQA1*01:03-DQB1*06:01 in the severe group was significantly higher than in the non-severe group during pre-Omicron epidemic period (32.4% vs. 7.9%, OR = 5.59, pc = 0.00072), and total period of this study (28.6% vs. 13.1%, OR = 2.63, pc = 0.0013). During Omicron epidemic period, the haplotype did not demonstrate statistical significance, although the odds ratio indicated a value greater 1. Frequencies of the HLA-DQA1*01:03 and -DQB1*06:01 alleles were significantly higher in severe COVID-19 patients, suggesting that these alleles are risk factors for severe COVID-19 pneumonia in the Japanese population.

HLA associations with immunotherapy related endocrine toxicity.

2516 Background: Endocrine immune related adverse events (irAEs), including thyroid dysfunction (ICI-T), diabetes mellitus (ICI-DM) and hypophysitis (ICI-HP), caused by treatment with immune checkpoint inhibitors (ICIs) are largely irreversible and pose a burden to cancer patients. Variation in HLA predisposes to many autoimmune conditions, but data associating endocrine irAEs to HLA types is limited. Methods: We included 6300 patients from multiple centers across the United States, Canada, Europe and Australia who were treated with ICIs for multiple cancer types. The most common endocrine irAE in these patients was ICI-T (806 cases), followed by ICI-HP (163 cases) and ICI-DM (75 cases). Cases of ICI induced adrenal insufficiency were considered to be ICI-HP unless the patient had an elevated ACTH. Patients without the specific irAE were considered controls. Genotyping and HLA imputation were completed at each institution independently. Associations between HLA types and ICI-T, ICI-DM, and ICI-HP were tested in patients of European ancestry after adjustment for type of immunotherapy (PD-1/PD-L1 monotherapy, anti-CTLA-4 monotherapy, and combination), cancer type, sex, age and 5 principal components, at each center. The results were then meta-analyzed using fixed-effects inverse-variance weighted approach. False discovery rate (FDR) adjusted p-value of 0.05 was considered significant whereas those between 0.05 and 0.1 were considered nominally significant. Results: In the ICI-DM group, we identified 1 significant association with HLA DRB1*04:01 (OR=2.45, FDR=0.002) which is known to increase risk for type 1 DM (T1DM) in European ancestry. There were 5 additional nominal associations including DRB1*0301 (OR=2.16, FDR=0.07), a known HLA for T1DM in European ancestry. For ICI-HP, there were 7 significantly associated HLA types. Of particular interest are DRB1*14:01 (OR=3.98, FDR=0.02), DRB1*07:01 (OR=1.86, FDR=0.02) and C*07:02 haplotype (OR=1.79, FDR=0.02). While DR7 and DR14 are novel associations, C*07:02 is in linkage disequilibrium with DR15 and DQB1*06:02, both of which have previously been associated with ICI-HP. For ICI-T, there were no associated HLA types. Conclusions: In our study, we report HLA types associated with endocrine irAEs. In particular, we see HLA associations for ICI-DM that are known to be associated with T1DM. This suggests a potential shared mechanism between these forms of autoimmune DM. Additionally we found novel HLA associations with ICI-HP. These findings may have an impact on the clinical care of patients treated with ICI. However, further work is warranted to determine if HLA typing prior to ICI initiation should be considered for irAE risk prediction, irAE surveillance, irAE prevention, and possibly cancer treatment decisions.

A multi-ethnic reference panel to impute HLA classical and non-classical class I alleles in admixed samples: Testing imputation accuracy in an admixed sample from Brazil.

The MHC class I region contains crucial genes for the innate and adaptive immune response, playing a key role in susceptibility to many autoimmune and infectious diseases. Genome-wide association studies have identified numerous disease-associated SNPs within this region. However, these associations do not fully capture the immune-biological relevance of specific HLA alleles. HLA imputation techniques may leverage available SNP arrays by predicting allele genotypes based on the linkage disequilibrium between SNPs and specific HLA alleles. Successful imputation requires diverse and large reference panels, especially for admixed populations. This study employed a bioinformatics approach to call SNPs and HLA alleles in multi-ethnic samples from the 1000 genomes (1KG) dataset and admixed individuals from Brazil (SABE), utilising 30X whole-genome sequencing data. Using HIBAG, we created three reference panels: 1KG (n = 2504), SABE (n = 1171), and the full model (n = 3675) encompassing all samples. In extensive cross-validation of these reference panels, the multi-ethnic 1KG reference exhibited overall superior performance than the reference with only Brazilian samples. However, the best results were achieved with the full model. Additionally, we expanded the scope of imputation by developing reference panels for non-classical, MICA, MICB and HLA-H genes, previously unavailable for multi-ethnic populations. Validation in an independent Brazilian dataset showcased the superiority of our reference panels over the Michigan Imputation Server, particularly in predicting HLA-B alleles among Brazilians. Our investigations underscored the need to enhance or adapt reference panels to encompass the target population's genetic diversity, emphasising the significance of multiethnic references for accurate imputation across different populations.

Genotype imputation methods for whole and complex genomic regions utilizing deep learning technology.

The imputation of unmeasured genotypes is essential in human genetic research, particularly in enhancing the power of genome-wide association studies and conducting subsequent fine-mapping. Recently, several deep learning-based genotype imputation methods for genome-wide variants with the capability of learning complex linkage disequilibrium patterns have been developed. Additionally, deep learning-based imputation has been applied to a distinct genomic region known as the major histocompatibility complex, referred to as HLA imputation. Despite their various advantages, the current deep learning-based genotype imputation methods do have certain limitations and have not yet become standard. These limitations include the modest accuracy improvement over statistical and conventional machine learning-based methods. However, their benefits include other aspects, such as their "reference-free" nature, which ensures complete privacy protection, and their higher computational efficiency. Furthermore, the continuing evolution of deep learning technologies is expected to contribute to further improvements in prediction accuracy and usability in the future.

Optimal population-specific HLA imputation with dimension reduction.

Human genomics has quickly evolved, powering genome-wide association studies (GWASs). SNP-based GWASs cannot capture the intense polymorphism of HLA genes, highly associated with disease susceptibility. There are methods to statistically impute HLA genotypes from SNP-genotypes data, but lack of diversity in reference panels hinders their performance. We evaluated the accuracy of the 1000 Genomes data as a reference panel for imputing HLA from admixed individuals of African and European ancestries, focusing on (a) the full dataset, (b) 10 replications from 6 populations, and (c) 19 conditions for the custom reference panels. The full dataset outperformed smaller models, with a good F1-score of 0.66 for HLA-B. However, custom models outperformed the multiethnic or population models of similar size (F1-scores up to 0.53, against up to 0.42). We demonstrated the importance of using genetically specific models for imputing populations, which are currently underrepresented in public datasets, opening the door to HLA imputation for every genetic population.

Identification of Distinct Genetic Profiles of Palindromic Rheumatism Using Whole-Exome Sequencing.

Previous studies have underlined the genetic susceptibility in the pathogenesis of palindromic rheumatism (PR), but the known PR loci only partially explain the disease's genetic background. We aimed to genetically identify PR by whole-exome sequencing (WES). This multicenter prospective study was conducted in 10 Chinese specialized rheumatology centers between September 2015 and January 2020. WES was performed in 185 patients with PR and in 272 healthy controls. PR patients were divided into PR subgroups who were negative for anti-citrullinated protein antibody (ACPA-) and positive for ACPA (ACPA+) according to ACPA titer (cutoff value 20 IU/liter). We conducted whole-exome association analysis for the WES data. We used HLA imputation to type HLA genes. In addition, we used the polygenic risk score to measure the genetic correlations between PR and rheumatoid arthritis (RA) and the genetic correlations between ACPA- PR and ACPA+ PR. Among 185 patients with PR enrolled in our study, 50 patients (27.02%) were ACPA+ and 135 PR patients (72.98%) were ACPA-. We identified 8 novel loci (in the ACPA- PR group: ZNF503, RPS6KL1, HOMER3, HLA-DRA; in the ACPA+ PR group: RPS6KL1, TNPO2, WASH2P, FANK1) and 3 HLA alleles (in the ACPA- PR group: HLA-DRB1*0803 and HLA-DQB1; in the ACPA+ PR group: HLA-DPA1*0401) that were associated with PR and that surpassed genome-wide significance (P < 5 × 10-8 ). Furthermore, polygenic risk score analysis showed that PR and RA were not similar (R2 < 0.025), whereas ACPA+ PR and ACPA- PR showed a moderate genetic correlation (0.38 < R2 < 0.8). This study demonstrated the distinct genetic background between ACPA- and ACPA+ PR patients. Additionally, our findings strengthened that PR and RA were not genetically similar.

Immunogenetics of HLA-B: SNP, allele, and haplotype diversity in populations from different continents and ancestry backgrounds.

HLA-B is among the most variable gene in the human genome. This gene encodes a key molecule for antigen presentation to CD8+ T lymphocytes and NK cell modulation. Despite the myriad of studies evaluating its coding region (with an emphasis on exons 2 and 3), few studies evaluated introns and regulatory sequences in real population samples. Thus, HLA-B variability is probably underestimated. We applied a bioinformatics pipeline tailored for HLA genes on 5347 samples from 80 different populations, which includes more than 1000 admixed Brazilians, to evaluate the HLA-B variability (SNPs, indels, MNPs, alleles, and haplotypes) in exons, introns, and regulatory regions. We observed 610 variable sites throughout HLA-B; the most frequent variants are shared worldwide. However, the haplotype distribution is geographically structured. We detected 920 full-length haplotypes (exons, introns, and untranslated regions) encoding 239 different protein sequences. HLA-B gene diversity is higher in admixed populations and Europeans while lower in African ancestry individuals. Each HLA-B allele group is associated with specific promoter sequences. This HLA-B variation resource may improve HLA imputation accuracy and disease-association studies and provide evolutionary insights regarding HLA-B genetic diversity in human populations.

Genome-Wide Association Study in Acute Tubulointerstitial Nephritis.

Polymorphisms of HLA genes may confer susceptibility to acute tubulointerstitial nephritis (ATIN), but small sample sizes and candidate gene design have hindered their investigation. The first genome-wide association study of ATIN identified two significant loci, risk haplotype DRB1*14-DQA1*0101-DQB1*0503 (DR14 serotype) and protective haplotype DRB1*1501-DQA1*0102-DQB1*0602 (DR15 serotype), with amino acid position 60 in the peptide-binding groove P10 of HLA-DR β 1 key. Risk alleles were shared among different causes of ATIN and HLA genotypes associated with kidney injury and immune therapy response. HLA alleles showed the strongest association. The findings suggest that a genetically conferred risk of immune dysregulation is part of the pathogenesis of ATIN. Acute tubulointerstitial nephritis (ATIN) is a rare immune-related disease, accounting for approximately 10% of patients with unexplained AKI. Previous elucidation of the relationship between genetic factors that contribute to its pathogenesis was hampered because of small sample sizes and candidate gene design. We undertook the first two-stage genome-wide association study and meta-analysis involving 544 kidney biopsy-defined patients with ATIN and 2346 controls of Chinese ancestry. We conducted statistical fine-mapping analysis, provided functional annotations of significant variants, estimated single nucleotide polymorphism (SNP)-based heritability, and checked genotype and subphenotype correlations. Two genome-wide significant loci, rs35087390 of HLA-DQA1 ( P =3.01×10 -39 ) on 6p21.32 and rs2417771 of PLEKHA5 on 12p12.3 ( P =2.14×10 -8 ), emerged from the analysis. HLA imputation using two reference panels suggested that HLA-DRB1*14 mainly drives the HLA risk association . HLA-DRB1 residue 60 belonging to pocket P10 was the key amino acid position. The SNP-based heritability estimates with and without the HLA locus were 20.43% and 10.35%, respectively. Different clinical subphenotypes (drug-related or tubulointerstitial nephritis and uveitis syndrome) seemed to share the same risk alleles. However, the HLA risk genotype was associated with disease severity and response rate to immunosuppressive therapy. We identified two candidate genome regions associated with susceptibility to ATIN. The findings suggest that a genetically conferred risk of immune dysregulation is involved in the pathogenesis of ATIN.

Estimation of Antibody-Verified Eplet Mismatch Load, 2-Field HLA Resolution vs Imputation in a Large Cohort of European Donors

The assessment of class II eplet mismatch load is useful to determine the risk of chronic rejection in solid organ transplantation. However, high-resolution (2-field) HLA typing is mandatory to accurately define eplet mismatches. The imputation of the most frequent allele has been used in retrospective studies. Here, we sought to investigate the concordant of antibody-verified (AbV) eplet load in different class II alelles between real 2-field HLA typing and HLA imputed by most frequent allele in a large White cohort. The allelic frequency of the different HLA class II loci was calculated using a database of high-resolution typing of 23,628 voluntary Spanish bone marrow donors obtained from the Spanish Registry of Bone Marrow Donors, managed by the Josep Carreras Foundation. The AbV eplet count in the different class II alleles was performed using the HLA-Matchmaker v3.1 algorithm. The probability of imputing the correct allele compared to the most frequent for DRB1 and DQB1 loci was 69.3% and 53.0%, respectively. However, studying the less frequent alleles, the same AbV eplet load was observed in 82.22% and 76.15%, respectively. Our data show that despite the potential error in the imputation of class II HLA typing, the number of AbV eplets is not significantly over- or underestimated in our population. Until high-resolution typing is widely established for solid organ transplantation, retrospective studies of AbV eplet mismatch load with imputed high-resolution types provide an acceptable outcome in White people.

Blood donor biobank and HLA imputation as a resource for HLA homozygous cells for therapeutic and research use

BackgroundAllogeneic therapeutic cells may be rejected if they express HLA alleles not found in the recipient. As finding cell donors with a full HLA match to a recipient requires vast donor pools, the use of HLA homozygous cells has been suggested as an alternative. HLA homozygous cells should be well tolerated by those who carry at least one copy of donor HLA alleles. HLA-A-B homozygotes could be valuable for HLA-matched thrombocyte products. We evaluated the feasibility of blood donor biobank and HLA imputation for the identification of potential cell donors homozygous for HLA alleles.MethodsWe imputed HLA-A, -B, -C, -DRB1, -DQA1, -DQB1 and -DPB1 alleles from genotypes of 20,737 Finnish blood donors in the Blood Service Biobank. We confirmed homozygosity by sequencing HLA alleles in 30 samples and by examining 36,161 MHC-located polymorphic DNA markers.ResultsThree hundred and seventeen individuals (1.5%), representing 41 different haplotypes, were found to be homozygous for HLA-A, -B, -C, -DRB1, -DQA1 and -DQB1 alleles. Ten most frequent haplotypes homozygous for HLA-A to -DQB1 were HLA-compatible with 49.5%, and three most frequent homozygotes to 30.4% of the Finnish population. Ten most frequent HLA-A-B homozygotes were compatible with 75.3%, and three most frequent haplotypes to 42.6% of the Finnish population. HLA homozygotes had a low level of heterozygosity in MHC-located DNA markers, in particular in HLA haplotypes enriched in Finland.ConclusionsThe present study shows that HLA imputation in a blood donor biobank of reasonable size can be used to identify HLA homozygous blood donors suitable for cell therapy, HLA-typed thrombocytes and research. The homozygotes were HLA-compatible with a large fraction of the Finnish population. Regular blood donors reported to have positive attitude to research donation appear a good option for these purposes. Differences in population frequencies of HLA haplotypes emphasize the need for population-specific collections of HLA homozygous samples.

Relevance of the MHC region for breast cancer susceptibility in Asians

BackgroundHuman leukocyte antigen (HLA) genes play critical roles in immune surveillance, an important defence against tumors. Imputing HLA genotypes from existing single-nucleotide polymorphism datasets is low-cost and efficient. We investigate the relevance of the major histocompatibility complex region in breast cancer susceptibility, using imputed class I and II HLA alleles, in 25,484 women of Asian ancestry.MethodsA total of 12,901 breast cancer cases and 12,583 controls from 12 case–control studies were included in our pooled analysis. HLA imputation was performed using SNP2HLA on 10,886 quality-controlled variants within the 15–55 Mb region on chromosome 6. HLA alleles (n = 175) with info scores greater than 0.8 and frequencies greater than 0.01 were included (resolution at two-digit level: 71; four-digit level: 104). We studied the associations between HLA alleles and breast cancer risk using logistic regression, adjusting for population structure and age. Associations between HLA alleles and the risk of subtypes of breast cancer (ER-positive, ER-negative, HER2-positive, HER2-negative, early-stage, and late-stage) were examined.ResultsWe did not observe associations between any HLA allele and breast cancer risk at P < 5e−8; the smallest p value was observed for HLA-C*12:03 (OR = 1.29, P = 1.08e−3). Ninety-five percent of the effect sizes (OR) observed were between 0.90 and 1.23. Similar results were observed when different subtypes of breast cancer were studied (95% of ORs were between 0.85 and 1.18).ConclusionsNo imputed HLA allele was associated with breast cancer risk in our large Asian study. Direct measurement of HLA gene expressions may be required to further explore the associations between HLA genes and breast cancer risk.

Protective association of HLA-DPB1*04:01:01 with acute encephalopathy with biphasic seizures and late reduced diffusion identified by HLA imputation.

Acute encephalopathy with biphasic seizures and late reduced diffusion (AESD) is a severe syndrome of acute encephalopathy that affects infants and young children. AESD is a polygenic disorder preceded by common viral infections with high fever. We conducted an association study of human leukocyte antigen (HLA) regions with AESD using HLA imputation. SNP genotyping was performed on 254 Japanese patients with AESD and 799 healthy controls. We conducted 3-field HLA imputation for 14 HLA genes based on Japanese-specific references using data from our previous genome-wide association study. After quality control, 208 patients and 737 controls were included in the analysis of HLA alleles. We then compared the carrier frequencies of HLA alleles and haplotypes between the patients and controls. HLA-DPB1*04:01:01 showed a significant association with AESD, exerting a protective effect against the disease (p = 0.0053, pcorrected = 0.042, odds ratio = 0.43, 95% confidence interval = 0.21-0.80). The allele frequency of HLA-DPB1*04:01:01 was lower in East Asians than in Caucasians, which may partially account for the higher incidence of AESD in the Japanese population. The present results demonstrate the importance of fine-mapping of the HLA region to investigate disease susceptibilities and elucidate the pathogenesis of AESD.

Whole-genome sequencing of 1,171 elderly admixed individuals from Brazil

As whole-genome sequencing (WGS) becomes the gold standard tool for studying population genomics and medical applications, data on diverse non-European and admixed individuals are still scarce. Here, we present a high-coverage WGS dataset of 1,171 highly admixed elderly Brazilians from a census-based cohort, providing over 76 million variants, of which ~2 million are absent from large public databases. WGS enables identification of ~2,000 previously undescribed mobile element insertions without previous description, nearly 5 Mb of genomic segments absent from the human genome reference, and over 140 alleles from HLA genes absent from public resources. We reclassify and curate pathogenicity assertions for nearly four hundred variants in genes associated with dominantly-inherited Mendelian disorders and calculate the incidence for selected recessive disorders, demonstrating the clinical usefulness of the present study. Finally, we observe that whole-genome and HLA imputation could be significantly improved compared to available datasets since rare variation represents the largest proportion of input from WGS. These results demonstrate that even smaller sample sizes of underrepresented populations bring relevant data for genomic studies, especially when exploring analyses allowed only by WGS.

A high-resolution HLA reference panel capturing global population diversity enables multi-ancestry fine-mapping in HIV host response.

Fine-mapping to plausible causal variation may be more effective in multi-ancestry cohorts, particularly in the MHC, which has population-specific structure. To enable such studies, we constructed a large (n = 21,546) HLA reference panel spanning five global populations based on whole-genome sequences. Despite population specific long-range haplotypes, we demonstrated accurate imputation at G-group resolution (94.2%, 93.7%, 97.8% and 93.7% in Admixed African (AA), East Asian (EAS), European (EUR) and Latino (LAT) populations). Applying HLA imputation to genome-wide association study (GWAS) data for HIV-1 viral load in three populations (EUR, AA and LAT), we obviated effects of previously reported associations from population-specific HIV studies and discovered a novel association at position 156 in HLA-B. We pinpointed the MHC association to three amino acid positions (97, 67 and 156) marking three consecutive pockets (C, B and D) within the HLA-B peptide binding groove, explaining 12.9% of trait variance.

048 The anti-drug antibody response is associated with amino acid variation within the HLADRB1 peptide-binding groove

Targeted biologic therapies can elicit an undesirable host immune response characterised by the development of anti-drug antibodies (ADA), an important cause of treatment failure. The most widely used biologic across immune-mediated diseases is the TNF inhibitor adalimumab. This genome-wide association study aimed to identify genetic predictors of developing ADA in psoriasis patients on their first course of adalimumab. Within the BSTOP bioresource, 784 psoriasis patients had ADA data 6-36 months after starting adalimumab (discovery cohort), 232 patients had ADA data <6 months after starting adalimumab (early ADA cohort), and 716 patients had only clinical data available (clinical outcome cohort). We genotyped and imputed 10,917,604 SNPs genome-wide and performed HLA imputation. We tested for association between genetic variation and development of ADA (logistic regression), as well as treatment failure (stopping adalimumab due to ineffectiveness [Cox proportional hazards]). We observed a genome-wide association with ADA within the Major Histocompatibility Complex (rs9268628, OR 3.44, 95% CI 2.21-5.36, p=4.93 x10-8) that mapped to the presence of tryptophan at position 9 (OR 0.34, 95% CI 0.25-0.45, p=1.76x10-12) and lysine at position 71 (OR 0.39, 95% CI 0.28-0.55, p=5.77x10-8) within the HLA-DRB1 peptide-binding groove, explaining 20% of ADA heritability. The protective effect of these residues on early ADA development was also observed in the second cohort of 232 patients. Underscoring their clinical relevance, these HLA-DRB1 residues were protective against treatment failure in the third cohort of 716 patients (HR 0.73, 95% CI 0.57-0.94, p=0.013). We suggest that pre-treatment HLA-DRB1 typing could direct clinical decision-making in terms of biologic selection and immunosuppressant co-therapy. Our findings also provide fundamental insight into the mechanisms by which amino acids within MHC Class II subunits play a critical role in selection of drug antigens displayed to the T-cell receptor, thereby determining the immune response to biologics.

HLA haplotype frequency estimation for heterogeneous populations using a graph-based imputation algorithm

HLA haplotype frequencies are estimated from ambiguous unphased HLA genotyping data using Expectation-Maximization (EM) algorithms. Current population genetics methods require independent EM frequency estimates for each population, and assume that each population is in Hardy-Weinberg Equilibrium (HWE). The HWE assumption of EM has thus far resulted in the exclusion of individuals from mixed or unknown ethnic backgrounds from reference datasets. Multi-region populations are currently poorly served by stem cell donor registry HLA imputation and matching implementations due to the inability of such algorithms to incorporate admixture into their population genetics models. To address this unmet need, we have expanded the imputation component of our GRaph IMputation and Matching (GRIMM) framework, where imputation becomes the expectation step in an iterative EM algorithm. Our novel multi-region EM implementation considers region as a Bayesian prior, enabling integration of HLA information from multiple single-region population groups, and for the first time including individuals with ambiguous or mixed ethnic backgrounds. We show that our multi-region EM produces much higher likelihood values and better haplotype recovery as measured by Kullback-Leibler divergence than all evaluated EM implementations when tested on real datasets of US donor registry HLA typings as well as simulated multi-region datasets of ambiguous HLA typings.

Genome-wide association study reveals an association between the HLA-DPB1∗02:01:02 allele and wheat-dependent exercise-induced anaphylaxis

Wheat-dependent exercise-induced anaphylaxis (WDEIA) is a life-threatening food allergy triggered by wheat in combination with the second factor such as exercise. The identification of potential genetic risk factors for this allergy might help high-risk individuals before consuming wheat-containing food. We aimed to identify genetic variants associated with WDEIA. A genome-wide association study was conducted in a discovery set of 77 individuals with WDEIA and 924 control subjects via three genetic models. The associations were confirmed in a replication set of 91 affected individuals and 435 control individuals. Summary statistics from the combined set were analyzed by meta-analysis with a random-effect model. In the discovery set, a locus on chromosome 6, rs9277630, was associated with WDEIA in the dominant model (OR = 3.95 [95% CI, 2.31-6.73], p = 7.87× 10-8). The HLA-DPB1∗02:01:02 allele displayed the most significant association with WDEIA (OR = 4.51 [95% CI, 2.66-7.63], p = 2.28× 10-9), as determined via HLA imputation following targeted sequencing. The association of the allele with WDEIA was confirmed in replication samples (OR = 3.82 [95% CI, 2.33-6.26], p = 3.03× 10-8). A meta-analysis performed in the combined set revealed that the HLA-DPB1∗02:01:02 allele was significantly associated with an increased risk of WDEIA (OR = 4.13 [95% CI, 2.89-5.93], p = 1.06× 10-14). Individuals carrying the HLA-DPB1∗02:01:02 allele have a significantly increased risk of WDEIA. Further validation of these findings in independent multiethnic cohorts is needed.

Association of HLA-DQA2 and HLA-B With Moyamoya Disease in the Chinese Han Population

ObjectiveAn HLA imputation was conducted to explore the relationship between HLA and patients with moyamoya disease (MMD) in the Chinese Han population.MethodsIn this study, we performed an association analysis of the major histocompatibility complex region in 2,786 individuals of Chinese Han ancestry (2,031 controls and 755 patients with MMD), through a widely used HLA imputation method.ResultsWe identified that the variant rs3129731 (odds ratio [OR] = 1.79, p = 3.69 × 10−16) located between the MTCO3P1 and HLA-DQA2 is a major genetic risk factor for MMD. In addition to this variant, found in the conditional association analysis, we also detected another independent signal, rs1071817 (OR = 0.62, p = 1.20 × 10−11), in HLA-B.ConclusionsOur research suggests that the genetic polymorphism of HLA-DQA2 and HLA-B could be a genetic predisposing factor for MMD in Chinese Han. This may provide some evidence for further HLA-related studies of patients with MMD of Chinese Han ethnicity and indicates that MMD is an immune-related disease.

Importance of HBsAg recognition by HLA molecules as revealed by responsiveness to different hepatitis B vaccines

Hepatitis B (HB) vaccines (Heptavax-II and Bimmugen) designed based on HBV genotypes A and C are mainly used for vaccination against HB in Japan. To determine whether there are differences in the genetic background associated with vaccine responsiveness, genome-wide association studies were performed on 555 Heptavax-II and 1193 Bimmugen recipients. Further HLA imputation and detailed analysis of the association with HLA genes showed that two haplotypes, DRB1*13:02-DQB1*06:04 and DRB1*04:05-DQB1*04:01, were significantly associated in comparison with high-responders (HBsAb > 100 mIU/mL) for the two HB vaccines. In particular, HLA-DRB1*13:02-DQB1*06:04 haplotype is of great interest in the sense that it could only be detected by direct analysis of the high-responders in vaccination with Heptavax-II or Bimmugen. Compared with healthy controls, DRB1*13:02-DQB1*06:04 was significantly less frequent in high-responders when vaccinated with Heptavax-II, indicating that high antibody titers were less likely to be obtained with Heptavax-II. As Bimmugen and Heptavax-II tended to have high and low vaccine responses to DRB1*13:02, 15 residues were found in the Heptavax-II-derived antigenic peptide predicted to have the most unstable HLA-peptide binding. Further functional analysis of selected hepatitis B patients with HLA haplotypes identified in this study is expected to lead to an understanding of the mechanisms underlying liver disease.