HIBAG--HLA genotype imputation with attribute bagging

Female-mediated selective sperm activation may remodel major histocompatibility complex-based mate choice decisions in humans

Major histocompatibility complex (MHC) genes are known to mediate mate choice both at the individual and gamete level. However, it has remained unclear how different episodes of MHC-associated mate choice interact and contribute to the total selection on MHC genes. Here, we clarified this interaction in humans by performing a full-factorial experiment where 10 females first ranked the attractiveness and intensity of the body odours of 11 males. Then we studied whether female odour preferences in these same 110 male-female combinations predicted sperm performance in the presence of follicular fluid (sperm-stimulating female reproductive fluid). When analyzing the total MHC similarity (including classical and non-classical MHC genes) of the male-female combinations, we found that females preferred the body odours of MHC-similar males, but that sperm motility was positively affected by the MHC dissimilarity of the male-female combinations. No associations were found for classical MHC genes only. Furthermore, odour preferences were negatively associated with sperm motility at the end of the follicular fluid treatment. Together, our results indicate that individual and gamete-level mate choice processes may act in opposing directions and that the most attractive males are not necessarily the most optimal partners at the post-copulatory level. Finally, our findings suggest that gamete-mediated mate choice may have a definitive role in disfavouring genetically incompatible partners from fertilizing oocytes.

MultiCook: A Tool That Improves Accuracy of HLA Imputation by Combining Probabilities From Multiple Reference Panels and Methods

HLA molecules are produced by genes within the Major Histocompatibility Complex (MHC). Although the identification of HLA genotype is costly, fortunately, recent computational methods have made it possible to impute HLA genotypes using inexpensive single nucleotide polymorphism (SNP) markers. These imputation methods perform well if ethnicity-matched reference panels are provided. However, the availability of large-sized panels specific to each ethnicity remains limited. As a result, to achieve better imputation for each population, we need to utilise available multiple reference panels together. In this study, we introduce MultiCook, which enables users to simultaneously utilise existing multiple HLA imputation methods with multiple reference panels. MultiCook is versatile in that panels typed with different SNP genotyping platforms can seamlessly be merged, outputs from multiple imputation methods can be combined and the output from the Michigan imputation server with a multiethnic reference panel can also be incorporated. We compared MultiCook to the existing single-reference-panel approaches and the Michigan HLA imputation server. In evaluation with a cohort of 413 Koreans, MultiCook reduced the imputation error rate by about one third, from 4.70% to 3.31%, by combining 1KG EAS (N = 504) and HAN Chinese (N = 9773) reference panels compared to the single-panel approach. Moreover, MultiCook achieved better accuracy for imputing low-frequency alleles in evaluation benchmarks.

HLA haplotype diversity, islet autoantibody status and discriminative ability of type 1 diabetes genetic risk score in Indians

AIMS

We have reported that a 9SNPs type 1 diabetes (T1D) Genetic Risk Score (GRS) developed from European data had a lower power in Indians to distinguish T1D from type 2 diabetes (T2D). We explore the performance of an improved (67SNPs) T1DGRS and also the potential reasons for lower discriminative ability to classify types of diabetes in Indians.

METHODS

We studied the discriminative ability of a 67SNPs European T1DGRS in 611 clinically diagnosed T1D and 1153 T2D patients, and 321 non-diabetic controls, using receiver operating characteristic (ROC) area under the curve (AUC). We also compared the frequency and effect sizes of HLA risk haplotypes between Indians and Europeans.

RESULTS

The T1DGRS was discriminative of T1D from T2D and controls. However, the ability is lower in Indians than Europeans (AUC, Europeans 0.92, Indians all T1D 0.83, AA-positive 0.86). The T1DGRS was higher in AA-positive than in AA-negative persons [13.01 (12.79-13.23) vs. 12.09 (11.64-12.56)], p < 0.0001. The association of common HLA-DQA1 ~ HLA-DQB1 haplotypes was broadly similar; however, important differences were noted in the frequency, direction and magnitude of effect for some haplotypes between Indians and Europeans.

CONCLUSIONS

We confirm broad applicability of European 67SNPs T1DGRS to Indian T1D persons. However, differences in HLA allele frequencies, magnitude and directional differences reduced the predictive value. Our results stress the need to generate ancestry-specific GRS, which we plan to do in the near future.

HLA-DQA1*05 Allele Carriage and Anti-TNF Therapy Persistence in Inflammatory Bowel Disease

INTRODUCTION

Carriage of the HLA-DQA1*05 allele is associated with development of antidrug antibodies (ADAs) to antitumor necrosis factor (anti-TNF) therapy in patients with Crohn's disease. However, ADA is not uniformly associated with treatment failure. We aimed to determine the impact of carriage of HLA-DQA1*05 allele on outcome of biologic therapy evaluated by drug persistence.

METHODS

A multicenter, retrospective study of 877 patients with inflammatory bowel disease (IBD) treated with anti-TNF therapy with HLA-DQA1*05 genotypes were generated by imputation from whole genome sequence using the HIBAG package, in R. Primary end point was anti-TNF therapy persistence, (time to therapy failure), segregated by HLA-DQA1*05 allele genotype and development of a risk score to predict anti-TNF therapy failure, incorporating HLA-DQA1*05 allele genotype status (LORisk score).

RESULTS

In all, 877 patients receiving anti-TNF therapy were included in our study; 543 (62%) had no copy, 281 (32%) one copy, and 53 (6%) 2 copies of HLA-DQA1*05 allele. Mean time to anti-TNF therapy failure in patients with 2 copies of HLA-DQA1*05 allele was significantly shorter compared with patients with 0 or 1 copy at 700 days' follow-up: 418 vs 541 vs 513 days, respectively (P = .012). Factors independently associated with time to anti-TNF therapy failure included carriage of HLA-DQA1*05 allele (hazard ratio [HR], 1.2, P = .02; female gender HR, 1.6, P < .001; UC phenotype HR, 1.4, P = .009; and anti-TNF therapy type [infliximab], HR, 1.5, P = .002). The LORisk score was significantly associated with shorter time to anti-TNF therapy failure (P < .001).

CONCLUSIONS

Carriage of 2 HLA-DQA1*05 alleles is associated with less favorable outcomes for patients receiving anti-TNF therapy with shorter time to therapy failure. HLA-DQA1*05 genotype status in conjunction with clinical factors may aid in therapy selection in patients with IBD.

LILR genotype imputation with attribute bagging (LIBAG): leukocyte immunoglobulin-like receptor copy number imputation system

There are ten leukocyte immunoglobulin (Ig)-like receptor (LILR) genes, i.e., five genes encoding activating receptors (LILRA1, LILRA2, LILRA4, LILRA5, and LILRA6) characterized by their truncated cytoplasmic tails, and five genes encoding inhibitory receptors (LILRB1, LILRB2, LILRB3, LILRB4, and LILRB5) characterized by their extended cytoplasmic tails containing immunoreceptor tyrosine-based inhibitory motifs (ITIMs). Among these, LILRB3, LILRA6, and LILRA3 are known for harboring high frequencies of copy number variations (CNVs). However, the presence of CNVs in the leukocyte receptor complex (LRC) region complicates single nucleotide polymorphism (SNP) association analysis within commercially available SNP microarray datasets. This study introduces LILR Genotype Imputation with Attribute Bagging (LIBAG), a novel method for determining CNVs in LILRB3, LILRA6, and LILRA3 from commercially available SNP genotyping array datasets. LILRA6 CNV imputation accuracy peaked at 98.0% for the Infinium Japanese Screening Array, followed by 97.4% for Axiom Japonica V2, 97.3% for Axiom Japonica Array NEO, and 94.3% for Axiom Japonica V1, with the lowest recorded accuracy of 93.6% for the Axiom Genome-wide ASI1 array. For the 1000 Genomes Project (1kGP) dataset, LILRA6 CNV imputation achieved peak accuracies of 94.5% for 1kGP-EAS (East Asian), 86.6% for 1kGP-AMR (Admixed American), 83.8% for 1kGP-EUR European), and 75.0% for 1kGP-AFR (African), particularly after the 20 kb flanking region. Similarly, imputation accuracy for LILRA3 CNV progressively increased, peaking at the 80 kb flanking region. Accuracy reached 1kGP-AMR, reaching 99.2% and 98.9% for 1kGP-AFR, 98.7% for 1kGP-EUR, and 97.5% for 1kGP-EAS. Investigating the LILR copy number (CN) in diseases associated with HLA class I molecules will provide further insights into disease pathogenesis.

HLA-Region Genetic Association Analysis of Breast Cancer Patients With and Without Persistent Postsurgical Neuropathic Pain

BACKGROUND

Surgical nerve injuries lead to persistent neuropathic pain (NP) in up to 30% of patients. Among many other factors, polymorphisms in the human leukocyte antigen (HLA) genes have been suggested to contribute to the development of neuropathic pain.

METHODS

We performed a genetic association analysis of HLA class I and class II alleles in women who had been operated on for breast cancer. Patients had a surgeon-confirmed perioperative nerve injury and were examined 4-9 years after their surgery. Patients with painful (cases, n = 27) and painless (controls, n = 30) intercostobrachial nerve resection were studied. Cases included patients with definite NP with worst pain intensity in the past week ≥ 4/10 on a numerical rating scale (NRS) and controls had the same nerve injury with no NP or other pains. Whole-genome single nucleotide polymorphism data were produced, and HLA class I (HLA-A, -B, -C) and class II (HLA-DRB1, -DQA1, -DQB1 and -DPB1) alleles were determined by imputation.

RESULTS

HLA-DRB1*03:01, DQA1*05:01 and DQB1*02:01 alleles appeared to be associated with painful nerve injury after breast cancer surgery (nominal p = 0.007 for all, carriership OR = 12.0, 95% CI 1.38-104; FDR corrected p > 0.07). These alleles comprise the DR3-DQ2 haplotype, which is part of the ancestral haplotype AH8.1.

CONCLUSIONS

Our results provide further support for the role of HLA genetic variation in the development of persistent post-surgical neuropathic pain, which indirectly implies a mechanism involving immunological memory in this process.

SIGNIFICANCE STATEMENT

We report a novel association between the HLA-DR3-DQ2 haplotype and the development of persistent neuropathic pain after breast cancer surgery. Our results provide further evidence for the role of HLA polymorphism in persistent neuropathic pain.

Machine Learning Methods for Classifying Multiple Sclerosis and Alzheimer's Disease Using Genomic Data

Complex diseases pose challenges in prediction due to their multifactorial and polygenic nature. This study employed machine learning (ML) to analyze genomic data from the UK Biobank, aiming to predict the genomic predisposition to complex diseases like multiple sclerosis (MS) and Alzheimer's disease (AD). We tested logistic regression (LR), ensemble tree methods, and deep learning models for this purpose. LR displayed remarkable stability across various subsets of data, outshining deep learning approaches, which showed greater variability in performance. Additionally, ML methods demonstrated an ability to maintain optimal performance despite correlated genomic features due to linkage disequilibrium. When comparing the performance of polygenic risk score (PRS) with ML methods, PRS consistently performed at an average level. By employing explainability tools in the ML models of MS, we found that the results confirmed the polygenicity of this disease. The highest-prioritized genomic variants in MS were identified as expression or splicing quantitative trait loci located in non-coding regions within or near genes associated with the immune response, with a prevalence of human leukocyte antigen (HLA) gene annotations. Our findings shed light on both the potential and the challenges of employing ML to capture complex genomic patterns, paving the way for improved predictive models.

HLA-DQA1*05 Associates With Anti-Tumor Necrosis Factor Immunogenicity and Low Adalimumab Trough Concentrations in Inflammatory Bowel Disease Patients From the SERENE Ulcerative Colitis and Crohn's Disease Studies

BACKGROUND AND AIMS

Anti-tumor necrosis factor (anti-TNF) therapies are commonly prescribed treatments for Crohn's disease (CD) and ulcerative colitis (UC). Many patients treated with anti-TNF therapy eventually develop anti-drug antibodies (ADAs). Understanding the factors associated with immunogenicity in anti-TNF-treated patients can help guide treatment. The Humira SERENE studies were Phase 3 trials investigating adalimumab induction regimens in CD and UC patients.

METHODS

We imputed alleles for 7 HLA genes in 1100 patients from the SERENE CD and SERENE UC trials. We then tested these alleles for association with time to immunogenicity. Subsequently, we tested loci significantly associated with immunogenicity for their association with patients who had consistently low drug serum concentrations.

RESULTS

This study replicated the association of HLA-DQA1*05 with time to immunogenicity (hazard ratio [HR] 1.42, p = 2.22E-06). Specifically, HLA-DQA1*05:05 was strongly associated (HR 1.76, p = 2.02E-10) and we detected a novel association represented by HLA-DRB1*01:02 (HR 3.16, p = 2.92E-07). Carriage of HLA-DQA1*05:05 and HLA-DRB1*01:02 was associated with patients who experienced consistently low adalimumab trough concentrations (HLA-DQA1*05:05: odds ratio [OR] 1.98, p = 0.0049; HLA DRB1*01:02: OR 7.06, p = 7.44E-05).

CONCLUSIONS

We found a significant association between alleles at genes in the human HLA locus and the formation of adalimumab immunogenicity and low adalimumab drug serum concentrations in large clinical studies of CD and UC patients. This work extends previous findings in CD to UC and directly shows a genetic association in patients with low drug concentrations. This work builds on existing literature to suggest that genetic screening could be a useful tool for clinicians concerned with patient anti-TNF immunogenicity.

CLINICAL TRIAL REGISTRATION NUMBERS

SERENE CD (NCT02065570), SERENE UC (NCT02065622).

Blood immunophenotyping of multiple sclerosis patients at diagnosis identifies a classical monocyte subset associated to disease evolution

Introduction

Myeloid cells trafficking from the periphery to the central nervous system are key players in multiple sclerosis (MS) through antigen presentation, cytokine secretion and repair processes.

Methods

Combination of mass cytometry on blood cells from 60 MS patients at diagnosis and 29 healthy controls, along with single cell RNA sequencing on paired blood and cerebrospinal fluid (CSF) samples from 5 MS patients were used for myeloid cells detailing.

Results

Myeloid compartment study demonstrated an enrichment of a peculiar classical monocyte population in 22% of MS patients at the time of diagnosis. Notably, this patients' subgroup exhibited a more aggressive disease phenotype two years post-diagnosis. This monocytic population, detected in both the CSF and blood, was characterized by CD206, CD209, CCR5 and CCR2 expression, and was found to be more frequent in MS patients carrying the HLA-DRB1*15:01 allele. Furthermore, pathways analysis predicted that these cells had antigen presentation capabilities coupled with pro-inflammatory phenotype.

Discussion

Altogether, these results point toward the amplification of a specific and pathogenic myeloid cell subset in MS patients with genetic susceptibilities.

Deep learning captures the effect of epistasis in multifactorial diseases

Background

Polygenic risk score (PRS) prediction is widely used to assess the risk of diagnosis and progression of many diseases. Routinely, the weights of individual SNPs are estimated by the linear regression model that assumes independent and linear contribution of each SNP to the phenotype. However, for complex multifactorial diseases such as Alzheimer's disease, diabetes, cardiovascular disease, cancer, and others, association between individual SNPs and disease could be non-linear due to epistatic interactions. The aim of the presented study is to explore the power of non-linear machine learning algorithms and deep learning models to predict the risk of multifactorial diseases with epistasis.

Methods

Simulated data with 2- and 3-loci interactions and tested three different models of epistasis: additive, multiplicative and threshold, were generated using the GAMETES. Penetrance tables were generated using PyTOXO package. For machine learning methods we used multilayer perceptron (MLP), convolutional neural network (CNN) and recurrent neural network (RNN), Lasso regression, random forest and gradient boosting models. Performance of machine learning models were assessed using accuracy, AUC-ROC, AUC-PR, recall, precision, and F1 score.

Results

First, we tested ensemble tree methods and deep learning neural networks against LASSO linear regression model on simulated data with different types and strength of epistasis. The results showed that with the increase of strength of epistasis effect, non-linear models significantly outperform linear. Then the higher performance of non-linear models over linear was confirmed on real genetic data for multifactorial phenotypes such as obesity, type 1 diabetes, and psoriasis. From non-linear models, gradient boosting appeared to be the best model in obesity and psoriasis while deep learning methods significantly outperform linear approaches in type 1 diabetes.

Conclusion

Overall, our study underscores the efficacy of non-linear models and deep learning approaches in more accurately accounting for the effects of epistasis in simulations with specific configurations and in the context of certain diseases.

Genome-Wide Association Study Reveals a Causal Relationship Between Allergic Rhinitis and Hazelnut Allergy

BACKGROUND

Little is known about the genetics of food allergy (FA) to various allergens and the heterogeneity of FA in adults.

OBJECTIVE

We aimed to investigate genetic susceptibility to FA in an adult population and to assess the association between secondary FA and allergic rhinitis (AR).

METHODS

FA and allergen-specific FA were defined based on in-depth questionnaires and a previously published FA algorithm in the Lifelines. We performed a series of genome-wide association studies (GWAS) on FA and nine allergen-specific (e.g., hazelnut) FA in 21,353 adults in Lifelines. Single nucleotide polymorphisms (SNPs) (p < 1E-5) were replicated in a second independent set of 15,518 adults participating in the Lifelines followed by meta-analysis of the results of the two datasets. We subsequently investigated the causal relationship of AR to FA using Mendelian randomization (MR) analysis.

RESULTS

We observed co-occurrence of tree nuts and apple FA, with over 80% of this group also reporting AR. After meta-analysis, we identified one genome-wide significant locus near HLA-DPA1 associated with self-reported hazelnut allergy (hazelnutFA), of which the top SNP is rs5025825 (p = 2.51E-9, OR = 1.43). Two-sample MR indicated that AR is a significant causal risk factor for hazelnutFA (p-IVW = 5.27E-10, β = 5.90, p-pleiotropy = 0.46).

CONCLUSION

Our questionnaire enabled a large GWAS on self-reported FA in Dutch adults. We report one novel locus in the human leukocyte antigens (HLA) region associated with hazelnutFA, implying an association with antigen recognition. Our findings genetically link secondary FA to AR in adults.

No association between migraine and HLA alleles in a cohort of 13,210 individuals with migraine from the Danish Blood Donor Study

OBJECTIVE

To determine the association between human leukocyte antigen (HLA) alleles and migraine, migraine subtypes, and sex-specific factors.

BACKGROUND

It has long been hypothesized that inflammation contributes to migraine pathophysiology. This study examined the association between migraine and alleles in the HLA system, a key player in immune response and genetic diversity.

METHODS

We performed a case-control study and included 13,210 individuals with migraine and 86,738 controls. All participants were part of the Danish Blood Donor Study Genomic Cohort. Participants were genotyped and 111 HLA alleles on 15 HLA genes were imputed. We examined the association between HLA alleles and migraine subtypes, considering sex-specific differences.

RESULTS

We found no association between HLA alleles and migraine, neither overall, nor in the sex-specific analysis. In the migraine subtype analysis, three HLA alleles were associated with migraine without aura; however, these associations could not be replicated in an independent Icelandic cohort (2191 individuals with migraine without aura and 278,858 controls). Furthermore, we found no association between HLA alleles and migraine with aura or chronic migraine.

CONCLUSION

We found no evidence of an association between the HLA system and migraine, suggesting that genetic factors related to the HLA system do not play a significant role in migraine susceptibility.

Genome-wide association analyses identify distinct genetic architectures for age-related macular degeneration across ancestries

To effectively reduce vision loss due to age-related macular generation (AMD) on a global scale, knowledge of its genetic architecture in diverse populations is necessary. A critical element, AMD risk profiles in African and Hispanic/Latino ancestries, remains largely unknown. We combined data in the Million Veteran Program with five other cohorts to conduct the first multi-ancestry genome-wide association study of AMD and discovered 63 loci (30 novel). We observe marked cross-ancestry heterogeneity at major risk loci, especially in African-ancestry populations which demonstrate a primary signal in a major histocompatibility complex class II haplotype and reduced risk at the established CFH and ARMS2/HTRA1 loci. Dissecting local ancestry in admixed individuals, we find significantly smaller marginal effect sizes for CFH risk alleles in African ancestry haplotypes. Broadening efforts to include ancestrally distinct populations helped uncover genes and pathways that boost risk in an ancestry-dependent manner and are potential targets for corrective therapies.

Utility of polygenic risk scores to aid in the diagnosis of rheumatic diseases

Rheumatic diseases (RDs) are characterized by autoimmunity and autoinflammation and are recognized as complex due to the interplay of multiple genetic, environmental, and lifestyle factors in their pathogenesis. The rapid advancement of genome-wide association studies (GWASs) has enabled the identification of numerous single nucleotide polymorphisms (SNPs) associated with RD susceptibility. Based on these SNPs, polygenic risk scores (PRSs) have emerged as promising tools for quantifying genetic risk in this disease group. This chapter reviews the current status of PRSs in assessing the risk of RDs and discusses their potential to improve the accuracy of the diagnosis of these complex diseases through their ability to discriminate among different RDs. PRSs demonstrate a high discriminatory capacity for various RDs and show potential clinical utility. As GWASs continue to evolve, PRSs are expected to enable more precise risk stratification by integrating genetic, environmental, and lifestyle factors, thereby refining individual risk predictions and advancing disease management strategies.

An archaic HLA class I receptor allele diversifies natural killer cell-driven immunity in First Nations peoples of Oceania

Genetic variation in host immunity impacts the disproportionate burden of infectious diseases that can be experienced by First Nations peoples. Polymorphic human leukocyte antigen (HLA) class I and killer cell immunoglobulin-like receptors (KIRs) are key regulators of natural killer (NK) cells, which mediate early infection control. How this variation impacts their responses across populations is unclear. We show that HLA-A∗24:02 became the dominant ligand for inhibitory KIR3DL1 in First Nations peoples across Oceania, through positive natural selection. We identify KIR3DL1∗114, widespread across and unique to Oceania, as an allele lineage derived from archaic humans. KIR3DL1∗114+NK cells from First Nations Australian donors are inhibited through binding HLA-A∗24:02. The KIR3DL1∗114 lineage is defined by phenylalanine at residue 166. Structural and binding studies show phenylalanine 166 forms multiple unique contacts with HLA-peptide complexes, increasing both affinity and specificity. Accordingly, assessing immunogenetic variation and the functional implications for immunity are fundamental toward understanding population-based disease associations.

Association of Polygenic Risk Score With Lifetime Risk of Developing Multiple Sclerosis in a Population-Based Birth-Year Cohort

BACKGROUND AND OBJECTIVES

More than 200 genetic variants have been associated with multiple sclerosis (MS) susceptibility. However, it is unclear to what extent genetic factors influence lifetime risk of MS. Using a population-based birth-year cohort, we investigate the effect of genetics on lifetime risk of MS.

METHODS

In the Project Y study, we tracked down almost all persons with MS (pwMS) from birth year 1966 in the Netherlands. As control participants, we included non-MS participants from the Project Y cohort (born 1965-1967 in the Netherlands) and non-MS participants from the Amsterdam Dementia Cohort born between 1963 and 1969. Genetic variants associated with MS were determined in pwMS and control participants using genotyping or imputation methods. Polygenic risk scores (PRSs) based on variants and weights from the largest genetic study in MS were calculated for each participant and assigned into deciles based on the PRS distribution in the control participants. We examined the lifetime risk for each decile and the association between PRS and MS disease variables, including age at onset and time to secondary progression.

RESULTS

MS-PRS was calculated for 285 pwMS (mean age 53.0 ± 0.9 years, 72.3% female) and 267 control participants (mean age 51.8 ± 3.2 years, 58.1% female). Based on the lifetime risk estimation, we observed that 1:2,739 of the women with the lowest 30% genetic risk developed MS, whereas 1:92 of the women with the top 10% highest risk developed MS. For men, only 1:7,900 developed MS in the lowest 30% genetic risk group, compared with 1:293 men with the top 10% genetic risk. The PRS was not significantly associated with age at onset and time to secondary progression in both sexes.

DISCUSSION

Our results show that the lifetime risk of MS is strongly influenced by genetic factors. Our findings have the potential to support diagnostic certainty in individuals with suspected MS: a high PRS could strengthen a diagnosis, but especially a PRS from the lowest tail of the PRS distribution should be considered a red flag and could prevent misdiagnosing conditions that mimic MS.

Polymorphisms of HLA genes and hypersensitivity to penicillin among patients in a Taiwanese population

Penicillin allergy is a potentially life-threatening condition that is common among patients. However, the genetic associations with penicillin allergy are not yet recognized for prevention or diagnosis, particularly in East Asian populations. We conducted a retrospective case-control study using data from the Taiwan Precision Medicine Initiative and analysing DNA samples to identify eight major MHC Class I and Class II loci. We employed imputation methods for accurate HLA typing and enrolled 17,827 individuals who received penicillin. Logistic regression analyses were utilized to explore associations between HLA genotypes, comorbidities and allergy risk, while simultaneously conducting a subgroup analysis to explore the association between HLA genotypes, comorbidities and the severity of allergic reactions. Our study assigned 496 cases to the penicillin allergy group and 4960 controls to a matched group. The risk of penicillin allergy was significantly higher with HLA-DPB1*05:01 (OR = 1.36, p = .004) and HLA-DQB1*05:01 (OR = 1.54, p = .03), with adjusted p-values of .032 and .24, respectively. Urticaria was identified as a separate risk factor (OR = 1.73, p < .001). However, neither the HLA alleles nor the comorbidities had a significant relationship with the risk of severe penicillin-induced allergy. HLA-DPB1*05:01 was found to be significantly associated with penicillin allergy reactions among the Taiwanese population.

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.

Two-stage strategy using denoising autoencoders for robust reference-free genotype imputation with missing input genotypes

Widely used genotype imputation methods are based on the Li and Stephens model, which assumes that new haplotypes can be represented by modifying existing haplotypes in a reference panel through mutations and recombinations. These methods use genotypes from SNP arrays as inputs to estimate haplotypes that align with the input genotypes by analyzing recombination patterns within a reference panel, and then infer unobserved variants. While these methods require reference panels in an identifiable form, their public use is limited due to privacy and consent concerns. One strategy to overcome these limitations is to use de-identified haplotype information, such as summary statistics or model parameters. Advances in deep learning (DL) offer the potential to develop imputation methods that use haplotype information in a reference-free manner by handling it as model parameters, while maintaining comparable imputation accuracy to methods based on the Li and Stephens model. Here, we provide a brief introduction to DL-based reference-free genotype imputation methods, including RNN-IMP, developed by our research group. We then evaluate the performance of RNN-IMP against widely-used Li and Stephens model-based imputation methods in terms of accuracy (R2), using the 1000 Genomes Project Phase 3 dataset and corresponding simulated Omni2.5 SNP genotype data. Although RNN-IMP is sensitive to missing values in input genotypes, we propose a two-stage imputation strategy: missing genotypes are first imputed using denoising autoencoders; RNN-IMP then processes these imputed genotypes. This approach restores the imputation accuracy that is degraded by missing values, enhancing the practical use of RNN-IMP.

Efficient HLA imputation from sequential SNPs data by transformer

Human leukocyte antigen (HLA) genes are associated with a variety of diseases, yet the direct typing of HLA alleles is both time-consuming and costly. Consequently, various imputation methods leveraging sequential single nucleotide polymorphisms (SNPs) data have been proposed, employing either statistical or deep learning models, such as the convolutional neural network (CNN)-based model, DEEP*HLA. However, these methods exhibit limited imputation efficiency for infrequent alleles and necessitate a large size of reference dataset. In this context, we have developed a Transformer-based model to HLA allele imputation, named "HLA Reliable IMpuatioN by Transformer (HLARIMNT)" designed to exploit the sequential nature of SNPs data. We evaluated HLARIMNT's performance using two distinct reference panels; Pan-Asian reference panel (n = 530) and Type 1 Diabetes genetics Consortium (T1DGC) reference panel (n = 5225), alongside a combined panel (n = 1060). HLARIMNT demonstrated superior accuracy to DEEP*HLA across several indices, particularly for infrequent alleles. Furthermore, we explored the impact of varying training data sizes on imputation accuracy, finding that HLARIMNT consistently outperformed across all data size. These findings suggest that Transformer-based models can efficiently impute not only HLA types but potentially other gene types from sequential SNPs data.

HLA-DR3 ~ DQ2 associates with sensory neuropathy in paraneoplastic neurological syndromes with Hu antibodies

OBJECTIVES

To investigate the association between human leukocyte antigen (HLA) and paraneoplastic neurological syndromes (PNS) with Hu antibodies, and potential specificities according to clinical presentation and cancer status.

METHODS

HLA genotypes at four-digit resolution were imputed from available genome-wide association data. Allele carrier frequencies were compared between patients (whole cohort, n = 100, and according to clinical presentation and cancer status) and matched healthy controls (n = 508) using logistic regression controlled by the three main principal components.

RESULTS

The clinical presentation of 100 anti-Hu patients involved the central nervous system (28, 28%), the peripheral nervous system (36, 36%) or both combined (36, 36%). Cancer diagnosis was certain in 75 (75%). HLA association analyses revealed that anti-Hu PNS patients were more frequently carriers of DQA1*05:01 (39% vs. 19%, OR = 2.8 [1.74-4.49]), DQB1*02:01 (39% vs. 18%, OR = 2.88 [1.79-4.64]) and DRB1*03:01 (41% vs. 19%, OR = 2.92 [1.80-4.73]) than healthy controls. Remarkably, such DR3 ~ DQ2 association was absent in patients with pure central involvement, but more specific to those manifesting with peripheral involvement: DQA1*05:01 (OR = 3.12 [1.48-6.60]), DQB1*02:01 (OR = 3.35 [1.57-7.15]) and DRB1*03:01 (OR = 3.62 [1.64-7.97]); being even stronger in cases with sensory neuropathy, DQA1*05:01 (OR = 4.41 [1.89-10.33]), DQB1*02:01 (OR = 4.85 [2.04-11.53]) and DRB1*03:01 (OR = 5.79 [2.28-14.74]). Similarly, DR3 ~ DQ2 association was only observed in patients with cancer.

DISCUSSION

Patients with anti-Hu PNS show different HLA profiles according to clinical presentation and, probably, cancer status, suggesting pathophysiological differences.

Accurate multi-population imputation of MICA, MICB, HLA-E, HLA-F and HLA-G alleles from genome SNP data

In addition to the classical HLA genes, the major histocompatibility complex (MHC) harbors a high number of other polymorphic genes with less established roles in disease associations and transplantation matching. To facilitate studies of the non-classical and non-HLA genes in large patient and biobank cohorts, we trained imputation models for MICA, MICB, HLA-E, HLA-F and HLA-G alleles on genome SNP array data. We show, using both population-specific and multi-population 1000 Genomes references, that the alleles of these genes can be accurately imputed for screening and research purposes. The best imputation model for MICA, MICB, HLA-E, -F and -G achieved a mean accuracy of 99.3% (min, max: 98.6, 99.9). Furthermore, validation of the 1000 Genomes exome short-read sequencing-based allele calling against a clinical-grade reference data showed an average accuracy of 99.8%, testifying for the quality of the 1000 Genomes data as an imputation reference. We also fitted the models for Infinium Global Screening Array (GSA, Illumina, Inc.) and Axiom Precision Medicine Research Array (PMRA, Thermo Fisher Scientific Inc.) SNP content, with mean accuracies of 99.1% (97.2, 100) and 98.9% (97.4, 100), respectively.

Association of HLA alleles with cephalosporin allergy in the Taiwanese population

Cephalosporin antibiotics are widely used in clinical settings, but they can cause hypersensitivity reactions, which may be influenced by genetic factors such as the expression of Human leukocyte antigen (HLA) molecules. This study aimed to investigate whether specific HLA alleles were associated with an increased risk of adverse reactions to cephalosporins among individuals in the Taiwanese population. This retrospective case-control study analyzed data from the Taiwan Precision Medicine Initiative (TPMI) on 27,933 individuals who received cephalosporin exposure and had HLA allele genotyping information available. Using logistic regression analyses, we examined the associations between HLA genotypes, comorbidities, allergy risk, and severity. Among the study population, 278 individuals had cephalosporin allergy and 2780 were in the control group. Our results indicated that certain HLA alleles, including HLA-B*55:02 (OR = 1.76, 95% CI 1.18-2.61, p = 0.005), HLA-C*01:02 (OR = 1.36, 95% CI 1.05-1.77, p = 0.018), and HLA-DQB1*06:09 (OR = 2.58, 95% CI 1.62-4.12, p < 0.001), were significantly associated with an increased risk of cephalosporin allergy reactions. Additionally, the HLA-C*01:02 allele genotype was significantly associated with a higher risk of severe allergy (OR = 2.33, 95% CI 1.05-5.15, p = 0.04). This study identified significant associations between HLA alleles and an increased risk of cephalosporin allergy, which can aid in early detection and prediction of adverse drug reactions to cephalosporins. Furthermore, our study highlights the importance of HLA typing in drug safety and expanding our knowledge of drug hypersensitivity syndromes.

Lack of association between classical HLA genes and asymptomatic SARS-CoV-2 infection

Human genetic studies of critical COVID-19 pneumonia have revealed the essential role of type I interferon-dependent innate immunity to SARS-CoV-2 infection. Conversely, an association between the HLA-B∗15:01 allele and asymptomatic SARS-CoV-2 infection in unvaccinated individuals was recently reported, suggesting a contribution of pre-existing T cell-dependent adaptive immunity. We report a lack of association of classical HLA alleles, including HLA-B∗15:01, with pre-omicron asymptomatic SARS-CoV-2 infection in unvaccinated participants in a prospective population-based study in the United States (191 asymptomatic vs. 945 symptomatic COVID-19 cases). Moreover, we found no such association in the international COVID Human Genetic Effort cohort (206 asymptomatic vs. 574 mild or moderate COVID-19 cases and 1,625 severe or critical COVID-19 cases). Finally, in the Human Challenge Characterisation study, the three HLA-B∗15:01 individuals infected with SARS-CoV-2 developed symptoms. As with other acute primary infections studied, no classical HLA alleles favoring an asymptomatic course of SARS-CoV-2 infection were identified.

eQTLs identify regulatory networks and drivers of variation in the individual response to sepsis

Sepsis is a clinical syndrome of life-threatening organ dysfunction caused by a dysregulated response to infection, for which disease heterogeneity is a major obstacle to developing targeted treatments. We have previously identified gene-expression-based patient subgroups (sepsis response signatures [SRS]) informative for outcome and underlying pathophysiology. Here, we aimed to investigate the role of genetic variation in determining the host transcriptomic response and to delineate regulatory networks underlying SRS. Using genotyping and RNA-sequencing data on 638 adult sepsis patients, we report 16,049 independent expression (eQTLs) and 32 co-expression module (modQTLs) quantitative trait loci in this disease context. We identified significant interactions between SRS and genotype for 1,578 SNP-gene pairs and combined transcription factor (TF) binding site information (SNP2TFBS) and predicted regulon activity (DoRothEA) to identify candidate upstream regulators. Overall, these approaches identified putative mechanistic links between host genetic variation, cell subtypes, and the individual transcriptomic response to infection.

HLA and KIR genetic association and NK cells in anti-NMDAR encephalitis

Introduction

Genetic predisposition to autoimmune encephalitis with antibodies against N-methyl-D-aspartate receptor (NMDAR) is poorly understood. Given the diversity of associated environmental factors (tumors, infections), we hypothesized that human leukocyte antigen (HLA) and killer-cell immunoglobulin-like receptors (KIR), two extremely polymorphic gene complexes key to the immune system, might be relevant for the genetic predisposition to anti-NMDAR encephalitis. Notably, KIR are chiefly expressed by Natural Killer (NK) cells, recognize distinct HLA class I allotypes and play a major role in anti-tumor and anti-infection responses.

Methods

We conducted a Genome Wide Association Study (GWAS) with subsequent control-matching using Principal Component Analysis (PCA) and HLA imputation, in a multi-ethnic cohort of anti-NMDAR encephalitis (n=479); KIR and HLA were further sequenced in a large subsample (n=323). PCA-controlled logistic regression was then conducted for carrier frequencies (HLA and KIR) and copy number variation (KIR). HLA-KIR interaction associations were also modeled. Additionally, single cell sequencing was conducted in peripheral blood mononuclear cells from 16 cases and 16 controls, NK cells were sorted and phenotyped.

Results

Anti-NMDAR encephalitis showed a weak HLA association with DRB1*01:01~DQA1*01:01~DQB1*05:01 (OR=1.57, 1.51, 1.45; respectively), and DRB1*11:01 (OR=1.60); these effects were stronger in European descendants and in patients without an underlying ovarian teratoma. More interestingly, we found increased copy number variation of KIR2DL5B (OR=1.72), principally due to an overrepresentation of KIR2DL5B*00201. Further, we identified two allele associations in framework genes, KIR2DL4*00103 (25.4% vs. 12.5% in controls, OR=1.98) and KIR3DL3*00302 (5.3% vs. 1.3%, OR=4.44). Notably, the ligands of these KIR2DL4 and KIR3DL3, respectively, HLA-G and HHLA2, are known to act as immune checkpoint with immunosuppressive functions. However, we did not find differences in specific KIR-HLA ligand interactions or HLA-G polymorphisms between cases and controls. Similarly, gene expression of CD56dim or CD56bright NK cells did not differ between cases and controls.

Discussion

Our observations for the first time suggest that the HLA-KIR axis might be involved in anti-NMDAR encephalitis. While the genetic risk conferred by the identified polymorphisms appears small, a role of this axis in the pathophysiology of this disease appears highly plausible and should be analyzed in future studies.

HLA-DQB1*05 subtypes and not DRB1*10:01 mediates risk in anti-IgLON5 disease

Anti-IgLON5 disease is a rare and likely underdiagnosed subtype of autoimmune encephalitis. The disease displays a heterogeneous phenotype that includes sleep, movement and bulbar-associated dysfunction. The presence of IgLON5-antibodies in CSF/serum, together with a strong association with HLA-DRB1*10:01∼DQB1*05:01, supports an autoimmune basis. In this study, a multicentric human leukocyte antigen (HLA) study of 87 anti-IgLON5 patients revealed a stronger association with HLA-DQ than HLA-DR. Specifically, we identified a predisposing rank-wise association with HLA-DQA1*01:05∼DQB1*05:01, HLA-DQA1*01:01∼DQB1*05:01 and HLA-DQA1*01:04∼DQB1*05:03 in 85% of patients. HLA sequences and binding cores for these three DQ heterodimers were similar, unlike those of linked DRB1 alleles, supporting a causal link to HLA-DQ. This association was further reflected in an increasingly later age of onset across each genotype group, with a delay of up to 11 years, while HLA-DQ-dosage dependent effects were also suggested by reduced risk in the presence of non-predisposing DQ1 alleles. The functional relevance of the observed HLA-DQ molecules was studied with competition binding assays. These proof-of-concept experiments revealed preferential binding of IgLON5 in a post-translationally modified, but not native, state to all three risk-associated HLA-DQ receptors. Further, a deamidated peptide from the Ig2-domain of IgLON5 activated T cells in two patients, compared with one control carrying HLA-DQA1*01:05∼DQB1*05:01. Taken together, these data support a HLA-DQ-mediated T-cell response to IgLON5 as a potentially key step in the initiation of autoimmunity in this disease.

Epistatic interaction between ERAP2 and HLA modulates HIV-1 adaptation and disease outcome in an Australian population

A strong genetic predictor of outcome following untreated HIV-1 infection is the carriage of specific alleles of human leukocyte antigens (HLAs) that present viral epitopes to T cells. Residual variation in outcome measures may be attributed, in part, to viral adaptation to HLA-restricted T cell responses. Variants of the endoplasmic reticulum aminopeptidases (ERAPs) influence the repertoire of T cell epitopes presented by HLA alleles as they trim pathogen-derived peptide precursors to optimal lengths for antigen presentation, along with other functions unrelated to antigen presentation. We investigated whether ERAP variants influence HLA-associated HIV-1 adaptation with demonstrable effects on overall HIV-1 disease outcome. Utilizing host and viral data of 249 West Australian individuals with HIV-1 subtype B infection, we identified a novel association between two linked ERAP2 single nucleotide polymorphisms (SNPs; rs2248374 and rs2549782) with plasma HIV RNA concentration (viral load) (P adjusted = 0.0024 for both SNPs). Greater HLA-associated HIV-1 adaptation in the HIV-1 Gag gene correlated significantly with higher viral load, lower CD4+ T cell count and proportion; P = 0.0103, P = 0.0061, P = 0.0061, respectively). When considered together, there was a significant interaction between the two ERAP2 SNPs and HLA-associated HIV-1 adaptation on viral load (P = 0.0111). In a comprehensive multivariate model, addition of ERAP2 haplotypes and HLA associated adaptation as an interaction term to known HLA and CCR5 determinants and demographic factors, increased the explanatory variance of population viral load from 17.67% to 45.1% in this dataset. These effects were not replicated in publicly available datasets with comparably sized cohorts, suggesting that any true global epistasis may be dependent on specific HLA-ERAP allelic combinations. Our data raises the possibility that ERAP2 variants may shape peptide repertoires presented to HLA class I-restricted T cells to modulate the degree of viral adaptation within individuals, in turn contributing to disease variability at the population level. Analyses of other populations and experimental studies, ideally with locally derived ERAP genotyping and HLA-specific viral adaptations are needed to elucidate this further.

Unveiling new genetic insights in rheumatoid arthritis for drug discovery through Taxonomy3 analysis

Genetic support for a drug target has been shown to increase the probability of success in drug development, with the potential to reduce attrition in the pharmaceutical industry alongside discovering novel therapeutic targets. It is therefore important to maximise the detection of genetic associations that affect disease susceptibility. Conventional statistical methods such as genome-wide association studies (GWAS) only identify some of the genetic contribution to disease, so novel analytical approaches are required to extract additional insights. C4X Discovery has developed Taxonomy3, a unique method for analysing genetic datasets based on mathematics that is novel in drug discovery. When applied to a previously published rheumatoid arthritis GWAS dataset, Taxonomy3 identified many additional novel genetic signals associated with this autoimmune disease. Follow-up studies using tool compounds support the utility of the method in identifying novel biology and tractable drug targets with genetic support for further investigation.

Oral and non-oral lichen planus show genetic heterogeneity and differential risk for autoimmune disease and oral cancer

Lichen planus (LP) is a T-cell-mediated inflammatory disease affecting squamous epithelia in many parts of the body, most often the skin and oral mucosa. Cutaneous LP is usually transient and oral LP (OLP) is most often chronic, so we performed a large-scale genetic and epidemiological study of LP to address whether the oral and non-oral subgroups have shared or distinct underlying pathologies and their overlap with autoimmune disease. Using lifelong records covering diagnoses, procedures, and clinic identity from 473,580 individuals in the FinnGen study, genome-wide association analyses were conducted on carefully constructed subcategories of OLP (n = 3,323) and non-oral LP (n = 4,356) and on the combined group. We identified 15 genome-wide significant associations in FinnGen and an additional 12 when meta-analyzed with UKBB (27 independent associations at 25 distinct genomic locations), most of which are shared between oral and non-oral LP. Many associations coincide with known autoimmune disease loci, consistent with the epidemiologic enrichment of LP with hypothyroidism and other autoimmune diseases. Notably, a third of the FinnGen associations demonstrate significant differences between OLP and non-OLP. We also observed a 13.6-fold risk for tongue cancer and an elevated risk for other oral cancers in OLP, in agreement with earlier reports that connect LP with higher cancer incidence. In addition to a large-scale dissection of LP genetics and comorbidities, our study demonstrates the use of comprehensive, multidimensional health registry data to address outstanding clinical questions and reveal underlying biological mechanisms in common but understudied diseases.

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.

Genome-wide association analysis reveals the associations of NPHP4, TYW1-AUTS2 and SEMA6D for Behçet's disease and HLA-B*46:01 for its intestinal involvement

BACKGROUND

Intestinal involvement in Behçet's disease (BD) is associated with poor prognosis and is more prevalent in East Asian than in Mediterranean populations. Identifying the genetic causes of intestinal BD is important for understanding the pathogenesis and for appropriate treatment of BD patients.

METHODS

We performed genome-wide association studies (GWAS) and imputation/replication genotyping of human leukocyte antigen (HLA) alleles for 1,689 Korean and Turkish patients with BD (including 379 patients with intestinal BD) and 2,327 healthy controls, followed by replication using 593 Japanese patients with BD (101 patients with intestinal BD) and 737 healthy controls. Stratified cross-phenotype analyses were performed for 1) overall BD, 2) intestinal BD, and 3) intestinal BD without association of overall BD.

RESULTS

We identified three novel genome-wide significant susceptibility loci including NPHP4 (rs74566205; P=1.36 × 10-8), TYW1-AUTS2 (rs60021986; P=1.14 × 10-9), and SEMA6D (rs4143322; P=5.54 × 10-9) for overall BD, and a new association with HLA-B*46:01 for intestinal BD (P=1.67 × 10-8) but not for BD without intestinal involvement. HLA peptide binding analysis revealed that Mycobacterial peptides, have a stronger binding affinity to HLA-B*46:01 compared to the known risk allele HLA-B*51:01.

CONCLUSIONS

HLA-B*46:01 is associated with the development of intestinal BD; NPHP4, TYW1-AUTS2, and SEMA6D are susceptibility loci for overall BD.

Drug-induced liver injury (DILI) ascribed to non-steroidal anti-inflammatory drugs (NSAIDs) in the USA-Update with genetic correlations

OBJECTIVE

To describe patients with NSAID-DILI, including genetic factors associated with idiosyncratic DILI.

METHODS

In DILIN, subjects with presumed DILI are enrolled and followed for at least 6 months. Causality is adjudicated by a Delphic approach. HLA sequencing of multiethnic NSAID-DILI patients and HLA allele imputation of matching population controls were performed following overall, class and drug-based association analysis. Significant results were tested in a non-Hispanic White (NHW) case-control replication cohort.

RESULTS

Between September 2004 and March 2022, causality was adjudicated in 2498, and 55 (41 [75%] women) were assessed as likely due to NSAIDs. Median age at onset was 55 y (range 22-83 y). Diclofenac was the causative drug in 29, celecoxib in 7, ibuprofen in 5, etodolac and meloxicam each in 4. Except for meloxicam and oxaprozin (n = 2), the liver injury was hepatocellular with median R 15-25. HLA-DRB1*04:03 and HLA-B*35:03 were significantly more frequent in NSAID-DILI patients than in non-NSAID DILI controls. Interestingly, 85% of the HLA-DRB1*04:03 carriers developed DILI due to the use of acetic acid derivative NSAIDs, supporting the hypothesis that HLA-DRB1*04:03 could be a drug and/or class risk factor. HLA-B*35:03 but not HLA-DRB1*04:03 association was confirmed in the independent NHW replication cohort, which was largely driven by diclofenac.

CONCLUSIONS

Despite prevalent use, NSAID-DILI is infrequent in the United States. Diclofenac is the most commonly implicated, and adherence to warnings of risk and close observation are recommended. The increased frequency of HLA-B*35:03 and DRB1*04:03, driven by diclofenac, suggests the importance of immune-mediated responses.

Polygenic risk score for acute rejection based on donor-recipient non-HLA genotype mismatch

BACKGROUND

Acute rejection (AR) after kidney transplantation is an important allograft complication. To reduce the risk of post-transplant AR, determination of kidney transplant donor-recipient mismatching focuses on blood type and human leukocyte antigens (HLA), while it remains unclear whether non-HLA genetic mismatching is related to post-transplant complications.

METHODS

We carried out a genome-wide scan (HLA and non-HLA regions) on AR with a large kidney transplant cohort of 784 living donor-recipient pairs of European ancestry. An AR polygenic risk score (PRS) was constructed with the non-HLA single nucleotide polymorphisms (SNPs) filtered by independence (r2 < 0.2) and P-value (< 1×10-3) criteria. The PRS was validated in an independent cohort of 352 living donor-recipient pairs.

RESULTS

By the genome-wide scan, we identified one significant SNP rs6749137 with HR = 2.49 and P-value = 2.15×10-8. 1,307 non-HLA PRS SNPs passed the clumping plus thresholding and the PRS exhibited significant association with the AR in the validation cohort (HR = 1.54, 95% CI = (1.07, 2.22), p = 0.019). Further pathway analysis attributed the PRS genes into 13 categories, and the over-representation test identified 42 significant biological processes, the most significant of which is the cell morphogenesis (GO:0000902), with 4.08 fold of the percentage from homo species reference and FDR-adjusted P-value = 8.6×10-4.

CONCLUSIONS

Our results show the importance of donor-recipient mismatching in non-HLA regions. Additional work will be needed to understand the role of SNPs included in the PRS and to further improve donor-recipient genetic matching algorithms. Trial registry: Deterioration of Kidney Allograft Function Genomics (NCT00270712) and Genomics of Kidney Transplantation (NCT01714440) are registered on ClinicalTrials.gov.

Multiple polygenic risk scores can improve the prediction of systemic lupus erythematosus in Taiwan

OBJECTIVE

To identify new genetic variants associated with SLE in Taiwan and establish polygenic risk score (PRS) models to improve the early diagnostic accuracy of SLE.

METHODS

The study enrolled 2429 patients with SLE and 48 580 controls from China Medical University Hospital in Taiwan. A genome-wide association study (GWAS) and PRS analyses of SLE and other three SLE markers, namely ANA, anti-double-stranded DNA antibody (dsDNA) and anti-Smith antibody (Sm), were conducted.

RESULTS

Genetic variants associated with SLE were identified through GWAS. Some novel genes, which have been previously reported, such as RCC1L and EGLN3, were revealed to be associated with SLE in Taiwan. Multiple PRS models were established, and optimal cut-off points for each PRS were determined using the Youden Index. Combining the PRSs for SLE, ANA, dsDNA and Sm yielded an area under the curve of 0.64 for the optimal cut-off points. An analysis of human leucocyte antigen (HLA) haplotypes in SLE indicated that individuals with HLA-DQA1*01:01 and HLA-DQB1*05:01 were at a higher risk of being classified into the SLE group.

CONCLUSIONS

The use of PRSs to predict SLE enables the identification of high-risk patients before abnormal laboratory data were obtained or symptoms were manifested. Our findings underscore the potential of using PRSs and GWAS in identifying SLE markers, offering promise for early diagnosis and prediction of SLE.

Phenomic landscape and pharmacogenomic implications for HLA region in a Taiwan Han Chinese population

BACKGROUND

The human leukocyte antigen (HLA) genes, exhibiting significant genetic diversity, are associated with susceptibility to various clinical diseases and diverse in drug responses. High costs of HLA sequencing and the population-specific architecture of this genetic region necessitate the establishment of a population-specific HLA imputation reference panel. Moreover, there is a lack of understanding about the genetic and phenotypic landscape of HLA variations within the Taiwanese population.

METHODS

We created models for a Taiwanese-specific HLA imputation reference panel. These models were trained with the array genotype data and HLA sequencing data from 845 Taiwanese subjects. HLA imputation was applied for 59,448 Taiwanese subjects to characterize the HLA allele and haplotype frequencies. Additionally, a phenome-wide association study (PheWAS) was conducted to identify the phenotypes associated with HLA variations. The association of the biallelic HLA variants with the binary and quantitative traits were evaluated with additive logistic and linear regression models, respectively. Furthermore, an omnibus test with likelihood-ratio test was applied for each HLA amino acid position in the multiallelic HLA amino acid polymorphisms to compare the difference between a fitted model and a null model following a χ2 distribution of n-1 degree of freedom at a position with n residues. Finally, we estimated the prevalence of adverse drug reactions (ADR)-related HLA alleles in the Taiwanese population.

RESULTS

In this study, the reference panel models displayed remarkable accuracy, with averages of 99.3%, 98.9%, and 99.1% for 2-, 4-, 6-digit alleles of the eight classical HLA genes, respectively. For PheWAS, a total of 18,136 significant associations with HLA variants across 26 phenotypes are identified (p < 5×10-8), highlighting the pleiotropy feature of the HLA region. Among the independent signals, 15 are novel, including the association of HLA-B pos 138 variation with ankylosing spondylitis (AS), and rs9266290 and rs9266292 with allergy. Through an analysis spanning the entire HLA region, we identified clusters of phenotype correlations. Finally, the carriers of pharmacogenomic related HLA alleles, including HLA-C*01:02 (35.86%), HLA-B*58:01 (20.9%), and HLA-B*15:02 (8.38%), were characterized in the Taiwanese general population.

CONCLUSIONS

We successfully delivered the HLA imputation for 59,448 Taiwanese subjects and characterized the genetic and phenotypic landscapes of the HLA variations. In addition, we quantified the estimated prevalence of the ADR-related HLA alleles in the Taiwanese population. The developed HLA imputation reference panel could be used for estimation of population HLA allele frequencies, which can facilitate further studies in the role of HLA variants in a wider range of phenotypes in the population.

High-resolution African HLA resource uncovers HLA-DRB1 expression effects underlying vaccine response

How human genetic variation contributes to vaccine effectiveness in infants is unclear, and data are limited on these relationships in populations with African ancestries. We undertook genetic analyses of vaccine antibody responses in infants from Uganda (n = 1391), Burkina Faso (n = 353) and South Africa (n = 755), identifying associations between human leukocyte antigen (HLA) and antibody response for five of eight tested antigens spanning pertussis, diphtheria and hepatitis B vaccines. In addition, through HLA typing 1,702 individuals from 11 populations of African ancestry derived predominantly from the 1000 Genomes Project, we constructed an imputation resource, fine-mapping class II HLA-DR and DQ associations explaining up to 10% of antibody response variance in our infant cohorts. We observed differences in the genetic architecture of pertussis antibody response between the cohorts with African ancestries and an independent cohort with European ancestry, but found no in silico evidence of differences in HLA peptide binding affinity or breadth. Using immune cell expression quantitative trait loci datasets derived from African-ancestry samples from the 1000 Genomes Project, we found evidence of differential HLA-DRB1 expression correlating with inferred protection from pertussis following vaccination. This work suggests that HLA-DRB1 expression may play a role in vaccine response and should be considered alongside peptide selection to improve vaccine design.

Narrative Review Explaining the Role of HLA-A, -B, and -C Molecules in COVID-19 Disease in and around Africa

The coronavirus disease 2019 (COVID-19) has left a devasting effect on various regions globally. Africa has exceptionally high rates of other infectious diseases, such as tuberculosis (TB), human immunodeficiency virus (HIV), and malaria, and was not impacted by COVID-19 to the extent of other continents Globally, COVID-19 has caused approximately 7 million deaths and 700 million infections thus far. COVID-19 disease severity and susceptibility vary among individuals and populations, which could be attributed to various factors, including the viral strain, host genetics, environment, lifespan, and co-existing conditions. Host genetics play a substantial part in COVID-19 disease severity among individuals. Human leukocyte antigen (HLA) was previously been shown to be very important across host immune responses against viruses. HLA has been a widely studied gene region for various disease associations that have been identified. HLA proteins present peptides to the cytotoxic lymphocytes, which causes an immune response to kill infected cells. The HLA molecule serves as the central region for infectious disease association; therefore, we expect HLA disease association with COVID-19. Therefore, in this narrative review, we look at the HLA gene region, particularly, HLA class I, to understand its role in COVID-19 disease.

Local Ancestry at the Major Histocompatibility Complex Region is Not a Major Contributor to Disease Heterogeneity in a Multiethnic Lupus Cohort

OBJECTIVE

Systemic lupus erythematosus (SLE) is an autoimmune disease resulting in debilitating clinical manifestations that vary in severity by race and ethnicity with a disproportionate burden in African American, Mestizo, and Asian populations compared with populations of European descent. Differences in global and local genetic ancestry may shed light on the underlying mechanisms contributing to these disparities, including increased prevalence of lupus nephritis, younger age of symptom onset, and presence of autoantibodies.

METHODS

A total of 1,139 European, African American, and Mestizos patients with SLE were genotyped using the Affymetrix LAT1 World array. Global ancestry proportions were estimated using ADMIXTURE, and local ancestry was estimated using RFMIXv2.0. We investigated associations between lupus nephritis, age at onset, and autoantibody status with both global and local ancestry proportions within the Major Histocompatibility Complex region.

RESULTS

Our results showed small effect sizes that did not meet the threshold for statistical significance for global or local ancestry proportions in either African American or Mestizo patients with SLE who presented with the clinical manifestations of interest compared with those who did not.

CONCLUSION

These findings suggest that local genetic ancestry within the Major Histocompatibility Complex region is not a major contributor to these SLE manifestations among patients with SLE from admixed populations.

A machine-learning method for biobank-scale genetic prediction of blood group antigens

A key element for successful blood transfusion is compatibility of the patient and donor red blood cell (RBC) antigens. Precise antigen matching reduces the risk for immunization and other adverse transfusion outcomes. RBC antigens are encoded by specific genes, which allows developing computational methods for determining antigens from genomic data. We describe here a classification method for determining RBC antigens from genotyping array data. Random forest models for 39 RBC antigens in 14 blood group systems and for human platelet antigen (HPA)-1 were trained and tested using genotype and RBC antigen and HPA-1 typing data available for 1,192 blood donors in the Finnish Blood Service Biobank. The algorithm and models were further evaluated using a validation cohort of 111,667 Danish blood donors. In the Finnish test data set, the median (interquartile range [IQR]) balanced accuracy for 39 models was 99.9 (98.9-100)%. We were able to replicate 34 out of 39 Finnish models in the Danish cohort and the median (IQR) balanced accuracy for classifications was 97.1 (90.1-99.4)%. When applying models trained with the Danish cohort, the median (IQR) balanced accuracy for the 40 Danish models in the Danish test data set was 99.3 (95.1-99.8)%. The RBC antigen and HPA-1 prediction models demonstrated high overall accuracies suitable for probabilistic determination of blood groups and HPA-1 at biobank-scale. Furthermore, population-specific training cohort increased the accuracies of the models. This stand-alone and freely available method is applicable for research and screening for antigen-negative blood donors.

Genome-wide association study identifies human genetic variants associated with fatal outcome from Lassa fever

Infection with Lassa virus (LASV) can cause Lassa fever, a haemorrhagic illness with an estimated fatality rate of 29.7%, but causes no or mild symptoms in many individuals. Here, to investigate whether human genetic variation underlies the heterogeneity of LASV infection, we carried out genome-wide association studies (GWAS) as well as seroprevalence surveys, human leukocyte antigen typing and high-throughput variant functional characterization assays. We analysed Lassa fever susceptibility and fatal outcomes in 533 cases of Lassa fever and 1,986 population controls recruited over a 7 year period in Nigeria and Sierra Leone. We detected genome-wide significant variant associations with Lassa fever fatal outcomes near GRM7 and LIF in the Nigerian cohort. We also show that a haplotype bearing signatures of positive selection and overlapping LARGE1, a required LASV entry factor, is associated with decreased risk of Lassa fever in the Nigerian cohort but not in the Sierra Leone cohort. Overall, we identified variants and genes that may impact the risk of severe Lassa fever, demonstrating how GWAS can provide insight into viral pathogenesis.

Human Leukocyte Antigen Signatures as Pathophysiological Discriminants of Microscopic Colitis Subtypes

BACKGROUND AND AIMS

Microscopic colitis [MC] is currently regarded as an inflammatory bowel disease that manifests as two subtypes: collagenous colitis [CC] and lymphocytic colitis [LC]. Whether these represent a clinical continuum or distinct entities is, however, an open question. Genetic investigations may contribute important insight into their respective pathophysiologies.

METHODS

We conducted a genome-wide association study [GWAS] meta-analysis in 1498 CC, 373 LC patients, and 13 487 controls from Europe and the USA, combined with publicly available MC GWAS data from UK Biobank and FinnGen [2599 MC cases and 552 343 controls in total]. Human leukocyte antigen [HLA] alleles and polymorphic residues were imputed and tested for association, including conditional analyses for the identification of key causative variants and residues. Genetic correlations with other traits and diagnoses were also studied.

RESULTS

We detected strong HLA association with CC, and conditional analyses highlighted the DRB1*03:01 allele and its residues Y26, N77, and R74 as key to this association (best p = 1.4 × 10-23, odds ratio [OR] = 1.96). Nominally significant genetic correlations were detected between CC and pneumonia [rg = 0.77; p = 0.048] and oesophageal diseases [rg = 0.45, p = 0.023]. An additional locus was identified in MC GWAS analyses near the CLEC16A and RMI2 genes on chromosome 16 [rs35099084, p = 2.0 × 10-8, OR = 1.31]. No significant association was detected for LC.

CONCLUSION

Our results suggest CC and LC have distinct pathophysiological underpinnings, characterised by an HLA predisposing role only in CC. This challenges existing classifications, eventually calling for a re-evaluation of the utility of MC umbrella definitions.

Genetic risk for hospitalization of African American patients with severe mental illness reveals HLA loci

Background

Mood disorders such as major depressive and bipolar disorders, along with posttraumatic stress disorder (PTSD), schizophrenia (SCZ), and other psychotic disorders, constitute serious mental illnesses (SMI) and often lead to inpatient psychiatric care for adults. Risk factors associated with increased hospitalization rate in SMI (H-SMI) are largely unknown but likely involve a combination of genetic, environmental, and socio-behavioral factors. We performed a genome-wide association study in an African American cohort to identify possible genes associated with hospitalization due to SMI (H-SMI).

Methods

Patients hospitalized for psychiatric disorders (H-SMI; n=690) were compared with demographically matched controls (n=4467). Quality control and imputation of genome-wide data were performed following the Psychiatric Genetic Consortium (PGC)-PTSD guidelines. Imputation of the Human Leukocyte Antigen (HLA) locus was performed using the HIBAG package.

Results

Genome-wide association analysis revealed a genome-wide significant association at 6p22.1 locus in the ubiquitin D (UBD/FAT10) gene (rs362514, p=9.43x10-9) and around the HLA locus. Heritability of H-SMI (14.6%) was comparable to other psychiatric disorders (4% to 45%). We observed a nominally significant association with 2 HLA alleles: HLA-A*23:01 (OR=1.04, p=2.3x10-3) and HLA-C*06:02 (OR=1.04, p=1.5x10-3). Two other genes (VSP13D and TSPAN9), possibly associated with immune response, were found to be associated with H-SMI using gene-based analyses.

Conclusion

We observed a strong association between H-SMI and a locus that has been consistently and strongly associated with SCZ in multiple studies (6p21.32-p22.1), possibly indicating an involvement of the immune system and the immune response in the development of severe transdiagnostic SMI.

The absence of seroconversion after exposition to hepatitis C virus is not related to KIR-HLA genotype combinations (GEHEP-012 study)

BACKGROUND & AIMS

It has been reported that specific killer-cell immunoglobulin-like receptors (KIRs) and HLA genotype combinations, such as KIR2DS4/HLA-C1 with presence of KIRDL2 or KIRDL3, homozygous KIRDL3/HLA-C1 and KIR3DL1/≥2HLA-Bw4, are strongly associated with the lack of active infection and seroconversion after exposition to hepatitis C virus (HCV).

OBJECTIVE

To determine whether these KIR-HLA combinations are relevant factors involved in that phenotype.

PATIENTS AND METHODS

In this retrospective case-control study, genotype data from a genome-wide association study previously performed on low susceptibility to HCV-infection carried out on 27 high-risk HCV-seronegative (HRSN) individuals and 743 chronically infected (CI) subjects were used. HLA alleles were imputed using R package HIBAG v1.2223 and KIR genotypes were imputed using the online resource KIR*IMP v1.2.0.

RESULTS

It was possible to successfully impute at least one KIR-HLA genotype combination previously associated with the lack of infection and seroconversion after exposition to HCV in a total of 23 (85.2%) HRSN individuals and in 650 (87.5%) CI subjects. No KIR-HLA genotype combination analyzed was related to the HRSN condition.

CONCLUSIONS

Our results suggest that those KIR-HLA genotype combinations are not relevant factors involved in the lack of infection and seroconversion after exposition to HCV. More studies will be needed to completely understand this phenotype.

Multi-ancestry meta-analysis of host genetic susceptibility to tuberculosis identifies shared genetic architecture

The heritability of susceptibility to tuberculosis (TB) disease has been well recognized. Over 100 genes have been studied as candidates for TB susceptibility, and several variants were identified by genome-wide association studies (GWAS), but few replicate. We established the International Tuberculosis Host Genetics Consortium to perform a multi-ancestry meta-analysis of GWAS, including 14,153 cases and 19,536 controls of African, Asian, and European ancestry. Our analyses demonstrate a substantial degree of heritability (pooled polygenic h2 = 26.3%, 95% CI 23.7-29.0%) for susceptibility to TB that is shared across ancestries, highlighting an important host genetic influence on disease. We identified one global host genetic correlate for TB at genome-wide significance (p<5 × 10-8) in the human leukocyte antigen (HLA)-II region (rs28383206, p-value=5.2 × 10-9) but failed to replicate variants previously associated with TB susceptibility. These data demonstrate the complex shared genetic architecture of susceptibility to TB and the importance of large-scale GWAS analysis across multiple ancestries experiencing different levels of infection pressure.

Blood donor biobank as a resource in personalised biomedical genetic research

Health questionnaires and donation criteria result in accumulation of highly selected individuals in a blood donor population. To understand better the usefulness of a blood donor-based biobank in personalised disease-associated genetic studies, and for possible personalised blood donation policies, we evaluated the occurrence and distributions of common and rare disease-associated genetic variants in Finnish Blood Service Biobank. We analysed among 31,880 blood donors the occurrence and geographical distribution of (i) 53 rare Finnish-enriched disease-associated variants, (ii) mutations assumed to influence blood donation: four Bernard-Soulier syndrome and two hemochromatosis mutations, (iii) type I diabetes risk genotype HLA-DQ2/DQ8. In addition, we analysed the level of consanguinity in Blood Service Biobank. 80.3% of blood donors carried at least one (range 0-9 per donor) of the rare variants, many in homozygous form, as well. Donors carrying multiple rare variants were enriched in Eastern Finland. Haemochromatosis mutation HFE C282Y homozygosity was 43.8% higher than expected, whereas mutations leading to Bernard-Soulier thrombocytopenia were rare. The frequency of HLA-DQ2/DQ8 genotype was slightly lower than expected. First-degree consanguinity was higher in Blood Service Biobank than in the general population. We demonstrate that despite donor selection, the Blood Service Biobank is a valuable resource for personalised medical research and for genotype-selected samples from unaffected individuals. The geographical genetic substructure of Finland enables efficient recruitment of donors carrying rare variants. Furthermore, we show that blood donor biobank material can be utilised for personalised blood donation policies.

HLA-DPA1*02:01~B1*01:01 is a risk haplotype for primary sclerosing cholangitis mediating activation of NKp44+ NK cells

OBJECTIVE

Primary sclerosing cholangitis (PSC) is characterised by bile duct strictures and progressive liver disease, eventually requiring liver transplantation. Although the pathogenesis of PSC remains incompletely understood, strong associations with HLA-class II haplotypes have been described. As specific HLA-DP molecules can bind the activating NK-cell receptor NKp44, we investigated the role of HLA-DP/NKp44-interactions in PSC.

DESIGN

Liver tissue, intrahepatic and peripheral blood lymphocytes of individuals with PSC and control individuals were characterised using flow cytometry, immunohistochemical and immunofluorescence analyses. HLA-DPA1 and HLA-DPB1 imputation and association analyses were performed in 3408 individuals with PSC and 34 213 controls. NK cell activation on NKp44/HLA-DP interactions was assessed in vitro using plate-bound HLA-DP molecules and HLA-DPB wildtype versus knock-out human cholangiocyte organoids.

RESULTS

NKp44+NK cells were enriched in livers, and intrahepatic bile ducts of individuals with PSC showed higher expression of HLA-DP. HLA-DP haplotype analysis revealed a highly elevated PSC risk for HLA-DPA1*02:01~B1*01:01 (OR 1.99, p=6.7×10-50). Primary NKp44+NK cells exhibited significantly higher degranulation in response to plate-bound HLA-DPA1*02:01-DPB1*01:01 compared with control HLA-DP molecules, which were inhibited by anti-NKp44-blocking. Human cholangiocyte organoids expressing HLA-DPA1*02:01-DPB1*01:01 after IFN-γ-exposure demonstrated significantly increased binding to NKp44-Fc constructs compared with unstimulated controls. Importantly, HLA-DPA1*02:01-DPB1*01:01-expressing organoids increased degranulation of NKp44+NK cells compared with HLA-DPB1-KO organoids.

CONCLUSION

Our studies identify a novel PSC risk haplotype HLA-DP A1*02:01~DPB1*01:01 and provide clinical and functional data implicating NKp44+NK cells that recognise HLA-DPA1*02:01-DPB1*01:01 expressed on cholangiocytes in PSC pathogenesis.

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.

18th International HLA and Immunogenetics Workshop: Report on the SNP-HLA Reference Consortium (SHLARC) component

The SNP-HLA Reference Consortium (SHLARC), a component of the 18th International HLA and Immunogenetics Workshop, is aimed at collecting diverse and extensive human leukocyte antigen (HLA) data to create custom reference panels and enhance HLA imputation techniques. Genome-wide association studies (GWAS) have significantly contributed to identifying genetic associations with various diseases. The HLA genomic region has emerged as the top locus in GWAS, particularly in immune-related disorders. However, the limited information provided by single nucleotide polymorphisms (SNPs), the hallmark of GWAS, poses challenges, especially in the HLA region, where strong linkage disequilibrium (LD) spans several megabases. HLA imputation techniques have been developed using statistical inference in response to these challenges. These techniques enable the prediction of HLA alleles from genotyped GWAS SNPs. Here we present the SHLARC activities, a collaborative effort to create extensive, and multi-ethnic reference panels to enhance HLA imputation accuracy.

Imputation-Based HLA Typing with GWAS SNPs

SNP-based imputation approaches for human leukocyte antigen (HLA) typing take advantage of the haplotype structure within the major histocompatibility complex (MHC) region. These methods predict HLA classical alleles using dense SNP genotypes, commonly found on array-based platforms used in genome-wide association studies (GWAS). The analysis of HLA classical alleles can be conducted on current SNP datasets at no additional cost. Here, we describe the workflow of HIBAG, an imputation method with attribute bagging, to infer a sample's HLA classical alleles using SNP data. Two examples are offered to demonstrate the functionality using public HLA and SNP data from the latest release of the 1000 Genomes project: genotype imputation using pre-built classifiers in a GWAS, and model training to create a new prediction model. The GPU implementation facilitates model building, making it hundreds of times faster compared to the single-threaded implementation.