Single-residue linear and conformational B cell epitopes prediction using random and ESM-2 based projections

B cell epitope prediction methods are separated into linear sequence-based predictors and conformational epitope predictions that typically use the measured or predicted protein structure. Most linear predictions rely on the translation of the sequence to biologically based representations and the applications of machine learning on these representations. We here present CALIBER 'Conformational And LInear B cell Epitopes pRediction', and show that a bidirectional long short-term memory with random projection produces a more accurate prediction (test set AUC=0.789) than all current linear methods. The same predictor when combined with an Evolutionary Scale Modeling-2 projection also improves on the state of the art in conformational epitopes (AUC = 0.776). The inclusion of the graph of the 3D distances between residues did not increase the prediction accuracy. However, the long-range sequence information was essential for high accuracy. While the same model structure was applicable for linear and conformational epitopes, separate training was required for each. Combining the two slightly increased the linear accuracy (AUC 0.775 versus 0.768) and reduced the conformational accuracy (AUC = 0.769).

Combined imputation of HLA genotype and self-identified race leads to better donor-recipient matching

Allogeneic Hematopoietic Cell Transplantation (HCT) is a curative therapy for hematologic disorders and often requires human leukocyte antigen (HLA)-matched donors. Donor registries have recruited donors utilizing evolving technologies of HLA genotyping methods. This necessitates in-silico ambiguity resolution and statistical imputation based on haplotype frequencies estimated from donor data stratified by self-identified race and ethnicity (SIRE). However, SIRE has limited genetic validity and presents a challenge for individuals with unknown or mixed SIRE. We present MR-GRIMM "Multi-Race Graph IMputation and Matching" that simultaneously imputes the race/ethnic category and HLA genotype using a SIRE based prior. Additionally, we propose a novel method to impute HLA typing inconsistent with current haplotype frequencies. The performance of MR-GRIMM was validated using a dataset of 170,000 donor-recipient pairs. MR-GRIMM has an average 20 % lower matching error (1-AUC) than single-race imputation. The recall metric (sensitivity) of the race/ethnic category imputation from HLA was measured by comparing the imputed donor race with the donor-provided SIRE. Accuracies of 0.74 and 0.55 were obtained for the prediction of 5 broad and 21 detailed US population groups respectively. The operational implementation of this algorithm in a registry search could help improve match predictions and access to HLA-matched donors.

GRAMM: A new method for analysis of HLA in families

Recently, haplo-identical transplantation with multiple HLA mismatches has become a viable option for stem cell transplants. Haplotype sharing detection requires the imputation of donor and recipient. We show that even in high-resolution typing when all alleles are known, there is a 15% error rate in haplotype phasing, and even more in low-resolution typings. Similarly, in related donors, the parents' haplotypes should be imputed to determine what haplotype each child inherited. We propose graph-based family imputation (GRAMM) to phase alleles in family pedigree HLA typing data, and in mother-cord blood unit pairs. We show that GRAMM has practically no phasing errors when pedigree data are available. We apply GRAMM to simulations with different typing resolutions as well as paired cord-mother typings, and show very high phasing accuracy, and improved allele imputation accuracy. We use GRAMM to detect recombination events and show that the rate of falsely detected recombination events (false-positive rate) in simulations is very low. We then apply recombination detection to typed families to estimate the recombination rate in Israeli and Australian population datasets. The estimated recombination rate has an upper bound of 10%-20% per family (1%-4% per individual).

Trans-population graph-based coverage optimization of allogeneic cellular therapy

Background

Pre-clinical development and in-human trials of 'off-the-shelf' immune effector cell therapy (IECT) are burgeoning. IECT offers many potential advantages over autologous products. The relevant HLA matching criteria vary from product to product and depend on the strategies employed to reduce the risk of GvHD or to improve allo-IEC persistence, as warranted by different clinical indications, disease kinetics, on-target/off-tumor effects, and therapeutic cell type (T cell subtype, NK, etc.).

Objective

The optimal choice of candidate donors to maximize target patient population coverage and minimize cost and redundant effort in creating off-the-shelf IECT product banks is still an open problem. We propose here a solution to this problem, and test whether it would be more expensive to recruit additional donors or to prevent class I or class II HLA expression through gene editing.

Study design

We developed an optimal coverage problem, combined with a graph-based algorithm to solve the donor selection problem under different, clinically plausible scenarios (having different HLA matching priorities). We then compared the efficiency of different optimization algorithms - a greedy solution, a linear programming (LP) solution, and integer linear programming (ILP) -- as well as random donor selection (average of 5 random trials) to show that an optimization can be performed at the entire population level.

Results

The average additional population coverage per donor decrease with the number of donors, and varies with the scenario. The Greedy, LP and ILP algorithms consistently achieve the optimal coverage with far fewer donors than the random choice. In all cases, the number of randomly-selected donors required to achieve a desired coverage increases with increasing population. However, when optimal donors are selected, the number of donors required may counter-intuitively decrease with increasing population size. When comparing recruiting more donors vs gene editing, the latter was generally more expensive. When choosing donors and patients from different populations, the number of random donors required drastically increases, while the number of optimal donors does not change. Random donors fail to cover populations different from their original populations, while a small number of optimal donors from one population can cover a different population.

Discussion

Graph-based coverage optimization algorithms can flexibly handle various HLA matching criteria and accommodate additional information such as KIR genotype, when such information becomes routinely available. These algorithms offer a more efficient way to develop off-the-shelf IECT product banks compared to random donor selection and offer some possibility of improved transparency and standardization in product design.

New Insights on DR and DQ Human Leukocyte Antigens in Anti-LGI1 Encephalitis

BACKGROUND AND OBJECTIVES

To explore the clinical characteristics and HLA associations of patients with anti-leucine-rich glioma-inactivated 1 encephalitis (LGI1E) from a large single center in Israel. Anti-LGI1E is the most commonly diagnosed antibody-associated encephalitic syndrome in adults. Recent studies of various populations reveal significant associations with specific HLA genes. We examined the clinical characteristics and HLA associations of a cohort of Israeli patients.

METHODS

Seventeen consecutive patients with anti-LGI1E diagnosed at Tel Aviv Medical Center between the years 2011 and 2018 were included. HLA typing was performed using next-generation sequencing at the tissue typing laboratory of Sheba Medical Center and compared with data from the Ezer Mizion Bone Marrow Donor Registry, containing over 1,000,000 samples.

RESULTS

Our cohort displayed a male predominance and median age at onset in the 7th decade, as previously reported. The most common presenting symptom was seizures. Notably, paroxysmal dizziness spells were significantly more common than previously reported (35%), whereas faciobrachial dystonic seizures were found only in 23%. HLA analysis revealed overrepresentation of DRB1*07:01 (OR: 3.18, CI: 20.9 p < 1.e-5) and DRB1*04:02 (OR: 3.8, CI: 20.1 p < 1.e-5), as well as of the DQ allele DQB1*02:02 (OR: 2.8, CI: 14.2 p < 0.0001) as previously reported. A novel overrepresentation observed among our patients was of the DQB1*03:02 allele (OR: 2.3, CI: 6.9 p < 0.008). In addition, we found DR-DQ associations, among patients with anti-LGI1E, that showed complete or near-complete linkage disequilibrium (LD). By applying LD analysis to an unprecedentedly large control cohort, we were able to show that although in the general population, DQB*03:02 is not fully associated with DRB1*04:02, in the patient population, both alleles are always coupled, suggesting the DRB1*04:02 association to be primary to disease predisposition. In silico predictions performed for the overrepresented DQ alleles reveal them to be strong binders of LGI1-derived peptides, similarly to overrepresented DR alleles. These predictions suggest a possible correlation between peptide binding sites of paired DR-DQ alleles.

DISCUSSION

Our cohort presents distinct immune characteristics with substantially higher overrepresentation of DRB1*04:02 and slightly lower overrepresentation of DQB1*07:01 compared with previous reports implying differences between different populations. DQ-DR interactions found in our cohort may shed additional light on the complex role of immunogenetics in the pathogenesis of anti-LGI1E, implying a possible relevance of certain DQ alleles and DR-DQ interactions.

Expanding Our Knowledge of the Immunogenetic Characteristics of Anti-LGI1 Encephalitis—A Study of an Israeli Cohort Suggests Additional Significant HLA Associations With DQ Alleles

Objective

Exploring the clinical characteristics and HLA associations of patients with anti-leucine–rich glioma-inactivated 1 encephalitis (LGI1E) from a large single center in Israel

Background

Anti-LGI1E is one of the most commonly diagnosed antibody-associated encephalitic syndromes in adults. Recent studies of various populations reveal significant associations with specific Human Leukocyte Antigen (HLA) genes. We examined the clinical characteristics and HLA associations of a cohort of Israeli patients.

Design/Methods

Seventeen consecutive anti-LGI1E patients diagnosed at Tel Aviv Sourasky Medical Center between the years 2011-2018 were included. HLA typing was performed using NGS methodology at the tissue typing laboratory of Sheba Medical Center and compared to data from the "Ezer Mizion" Bone Marrow Donor Registry, containing over 1,000,000 samples.

Results

Our cohort displayed a male predominance and median age of onset in the 7th decade, as previously reported. All patients responded to immunotherapy, though residual damage was not uncommon (23% with MRS >1). HLA analysis revealed overexpression of DRB1*07:01 (OR 13, CI 0.6 p < 1.e-10) and DRB1*04:02 (OR 12, CI-0.6 p < 1.e-10), as previously reported, as well as of the DQ alleles DQB1*02:02 (OR 12, CI 0.6 p < 1.e-10), DQB1*03:03 (OR 27, CI 0.9 p < 1.e-10), previously attributed to linkage disequilibrium (LD) with the mentioned DR alleles. An additional allele overexpressed among our patients was the DQB1*03:02 allele (OR 12, CI 0.6 p < 1.e-10), which appeared in complete LD with DRB1*04:02. Linkage disequilibrium analysis performed on patients and controls suggests these DR-DQ associations are unique to anti-LGI1E patients. In silico predictions performed for the overexpressed DQ alleles reveal them to be strong binders of LGI1 derived peptides, and suggest a correlation between peptide binding sites of paired DR-DQ alleles.

Conclusions

Our findings shed additional light on the complex role of immunogenetics in the pathogenesis of anti-LGI1E, implying the possible relevance of certain DQ alleles as well as DR-DQ interactions.

Limited Contribution of Donor Characteristics to One-Year Survival After Hematopoietic Stem Cell Transplantation

A large number of association studies have related donor characteristics to survival after bone marrow transplantation, for leukemia in general and specifically for acute myeloid leukemia (AML) patients. However, population-based differences often do not hold at the single transplant level. We test whether transplantation outcomes can be predicted at the single-patient level and whether such predictions can be used to better choose donors. The analysis was performed on a mixture of different diseases or with AML only, and with either patient and donor information or donor information only. We analyzed 3671 8-of-8 HLA-matched AML donor-recipient pairs and tested whether the outcome, including 1-year total and event-free survival, can be predicted from patient and donor-related factors. We used multiple machine learning and survival analysis methods. The best method is a fully connected neural network. Multiple outcomes can be predicted, with area under the specificity-sensitivity curve (AUC) values between 0.54 and 0.67 for the different outcomes. The patient age has a strong impact on prediction. However, for a given patient, when only donor or transplant information is used, limited prediction accuracy of 0.54 to 0.56 AUC for event-free survival and survival is obtained. Graft-versus-host disease and rejection after 1 year have slightly higher AUC values of around 0.59, whereas the relapse prediction accuracy was random. All donors' characteristics have a limited influence on the quality of hematopoietic stem cell transplantation for fully matched donors. Many factors with a population effect on survival have a very limited effect when combined with all other factors in a single-donor predictive model.

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.

Non-syndromic autosomal recessive congenital ichthyosis in the Israeli population

BACKGROUND

Autosomal recessive congenital ichthyosis (ARCI) is the term given to a complex and heterogeneous group of cornification disorders associated with mutations in at least eight distinct genes. Mutation distribution and prevalence rates are instrumental for the design of diagnostic strategies in ARCI but have not yet been systematically explored in the Israeli population. Previous data suggest that the demographic features specific to Middle Eastern populations, such as a high frequency of consanguineous marriages, may have an effect on the molecular epidemiology of genodermatoses.

METHODS

We systematically assessed all families with ARCI presenting at our clinics over a period of 9 years, using a combination of homozygosity mapping, direct sequencing and PCR-restriction fragment length polymorphism assays.

RESULTS

In total, 20 families with ARCI were assessed, and causative mutations were identified in 7 genes: TGM1 (30% of patients), ALOX12B (20%), ABCA12 (5%), CYP4F22 (10%), ALOXE3 (10%), LIPN (5%) and NIPAL4 (5%) Two families (10%) had mutations mapped to an ARCI-associated locus on 12p11.2-q13, while no mutation was found for one additional kindred. In the subgroup of families of Arab Muslim origin, mutations were identified most frequently in ALOX12B and TGM1 (31%), whereas the other subgroups displayed a subtype distribution very similar to that previously reported in western populations.

CONCLUSIONS

The present data point to the need for population-tailored mutation screening strategies in genetically heterogeneous genodermatoses, based on the relative prevalence of the disease subsets.