PyIR: a scalable wrapper for processing billions of immunoglobulin and T cell receptor sequences using IgBLAST

An explainable language model for antibody specificity prediction using curated influenza hemagglutinin antibodies

Despite decades of antibody research, it remains challenging to predict the specificity of an antibody solely based on its sequence. Two major obstacles are the lack of appropriate models and the inaccessibility of datasets for model training. In this study, we curated >5,000 influenza hemagglutinin (HA) antibodies by mining research publications and patents, which revealed many distinct sequence features between antibodies to HA head and stem domains. We then leveraged this dataset to develop a lightweight memory B cell language model (mBLM) for sequence-based antibody specificity prediction. Model explainability analysis showed that mBLM could identify key sequence features of HA stem antibodies. Additionally, by applying mBLM to HA antibodies with unknown epitopes, we discovered and experimentally validated many HA stem antibodies. Overall, this study not only advances our molecular understanding of the antibody response to the influenza virus but also provides a valuable resource for applying deep learning to antibody research.

Isolation of human antibodies against influenza B neuraminidase and mechanisms of protection at the airway interface

Influenza B viruses (IBVs) comprise a substantial portion of the circulating seasonal human influenza viruses. Here, we describe the isolation of human monoclonal antibodies (mAbs) that recognized the IBV neuraminidase (NA) glycoprotein from an individual following seasonal vaccination. Competition-binding experiments suggested the antibodies recognized two major antigenic sites. One group, which included mAb FluB-393, broadly inhibited IBV NA sialidase activity, protected prophylactically in vivo, and bound to the lateral corner of NA. The second group contained an active site mAb, FluB-400, that broadly inhibited IBV NA sialidase activity and virus replication in vitro in primary human respiratory epithelial cell cultures and protected against IBV in vivo when administered systemically or intranasally. Overall, the findings described here shape our mechanistic understanding of the human immune response to the IBV NA glycoprotein through the demonstration of two mAb delivery routes for protection against IBV and the identification of potential IBV therapeutic candidates.

Evaluation of the Ability of AlphaFold to Predict the Three-Dimensional Structures of Antibodies and Epitopes

Being able to accurately predict the three-dimensional structure of an Ab can facilitate Ab characterization and epitope prediction, with important diagnostic and clinical implications. In this study, we evaluated the ability of AlphaFold to predict the structures of 222 recently published, high-resolution Fab H and L chain structures of Abs from different species directed against different Ags. We show that although the overall Ab prediction quality is in line with the results of CASP14, regions such as the complementarity-determining regions (CDRs) of the H chain, which are prone to higher variation, are predicted less accurately. Moreover, we discovered that AlphaFold mispredicts the bending angles between the variable and constant domains. To evaluate the ability of AlphaFold to model Ab-Ag interactions based only on sequence, we used AlphaFold-Multimer in combination with ZDOCK to predict the structures of 26 known Ab-Ag complexes. ZDOCK, which was applied on bound components of both the Ab and the Ag, succeeded in assembling 11 complexes, whereas AlphaFold succeeded in predicting only 2 of 26 models, with significant deviations in the docking contacts predicted in the rest of the molecules. Within the 11 complexes that were successfully predicted by ZDOCK, 9 involved short-peptide Ags (18-mer or less), whereas only 2 were complexes of Ab with a full-length protein. Docking of modeled unbound Ab and Ag was unsuccessful. In summary, our study provides important information about the abilities and limitations of using AlphaFold to predict Ab-Ag interactions and suggests areas for possible improvement.

Widespread impact of immunoglobulin V-gene allelic polymorphisms on antibody reactivity

The ability of the human immune system to generate antibodies to any given antigen can be strongly influenced by immunoglobulin V-gene allelic polymorphisms. However, previous studies have provided only limited examples. Therefore, the prevalence of this phenomenon has been unclear. By analyzing >1,000 publicly available antibody-antigen structures, we show that many V-gene allelic polymorphisms in antibody paratopes are determinants for antibody binding activity. Biolayer interferometry experiments further demonstrate that paratope allelic polymorphisms on both heavy and light chains often abolish antibody binding. We also illustrate the importance of minor V-gene allelic polymorphisms with low frequency in several broadly neutralizing antibodies to severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and influenza virus. Overall, this study not only highlights the pervasive impact of V-gene allelic polymorphisms on antibody binding but also provides mechanistic insights into the variability of antibody repertoires across individuals, which in turn have important implications for vaccine development and antibody discovery.

An explainable language model for antibody specificity prediction using curated influenza hemagglutinin antibodies

Despite decades of antibody research, it remains challenging to predict the specificity of an antibody solely based on its sequence. Two major obstacles are the lack of appropriate models and inaccessibility of datasets for model training. In this study, we curated a dataset of >5,000 influenza hemagglutinin (HA) antibodies by mining research publications and patents, which revealed many distinct sequence features between antibodies to HA head and stem domains. We then leveraged this dataset to develop a lightweight memory B cell language model (mBLM) for sequence-based antibody specificity prediction. Model explainability analysis showed that mBLM captured key sequence motifs of HA stem antibodies. Additionally, by applying mBLM to HA antibodies with unknown epitopes, we discovered and experimentally validated many HA stem antibodies. Overall, this study not only advances our molecular understanding of antibody response to influenza virus, but also provides an invaluable resource for applying deep learning to antibody research.

Widespread impact of immunoglobulin V gene allelic polymorphisms on antibody reactivity

The ability of human immune system to generate antibodies to any given antigen can be strongly influenced by immunoglobulin V gene (IGV) allelic polymorphisms. However, previous studies have provided only a limited number of examples. Therefore, the prevalence of this phenomenon has been unclear. By analyzing >1,000 publicly available antibody-antigen structures, we show that many IGV allelic polymorphisms in antibody paratopes are determinants for antibody binding activity. Biolayer interferometry experiment further demonstrates that paratope allelic mutations on both heavy and light chain often abolish antibody binding. We also illustrate the importance of minor IGV allelic variants with low frequency in several broadly neutralizing antibodies to SARS-CoV-2 and influenza virus. Overall, this study not only highlights the pervasive impact of IGV allelic polymorphisms on antibody binding, but also provides mechanistic insights into the variability of antibody repertoires across individuals, which in turn have important implications for vaccine development and antibody discovery.

Systematic analysis of human antibody response to ebolavirus glycoprotein shows high prevalence of neutralizing public clonotypes

Understanding the human antibody response to emerging viral pathogens is key to epidemic preparedness. As the size of the B cell response to a pathogenic-virus-protective antigen is poorly defined, we perform deep paired heavy- and light-chain sequencing in Ebola virus glycoprotein (EBOV-GP)-specific memory B cells, allowing analysis of the ebolavirus-specific antibody repertoire both genetically and functionally. This approach facilitates investigation of the molecular and genetic basis for the evolution of cross-reactive antibodies by elucidating germline-encoded properties of antibodies to EBOV and identification of the overlap between antibodies in the memory B cell and serum repertoire. We identify 73 public clonotypes of EBOV, 20% of which encode antibodies with neutralization activity and capacity to protect mice in vivo. This comprehensive analysis of the public and private antibody repertoire provides insight into the molecular basis of the humoral immune response to EBOV GP, which informs the design of vaccines and improved therapeutics.

Quantifying changes in the T cell receptor repertoire during thymic development

One of the feats of adaptive immunity is its ability to recognize foreign pathogens while sparing the self. During maturation in the thymus, T cells are selected through the binding properties of their antigen-specific T-cell receptor (TCR), through the elimination of both weakly (positive selection) and strongly (negative selection) self-reactive receptors. However, the impact of thymic selection on the TCR repertoire is poorly understood. Here, we use transgenic Nur77-mice expressing a T-cell activation reporter to study the repertoires of thymic T cells at various stages of their development, including cells that do not pass selection. We combine high-throughput repertoire sequencing with statistical inference techniques to characterize the selection of the TCR in these distinct subsets. We find small but significant differences in the TCR repertoire parameters between the maturation stages, which recapitulate known differentiation pathways leading to the CD4⁺ and CD8⁺ subtypes. These differences can be simulated by simple models of selection acting linearly on the sequence features. We find no evidence of specific sequences or sequence motifs or features that are suppressed by negative selection. These results favour a collective or statistical model for T-cell self non-self discrimination, where negative selection biases the repertoire away from self recognition, rather than ensuring lack of self-reactivity at the single-cell level.

Human antibodies to SARS-CoV-2 with a recurring YYDRxG motif retain binding and neutralization to variants of concern including Omicron

Studying the antibody response to SARS-CoV-2 informs on how the human immune system can respond to antigenic variants as well as other SARS-related viruses. Here, we structurally identified a YYDRxG motif encoded by IGHD3-22 in CDR H3 that facilitates antibody targeting to a functionally conserved epitope on the SARS-CoV-2 receptor binding domain. A computational search for a YYDRxG pattern in publicly available sequences uncovered 100 such antibodies, many of which can neutralize SARS-CoV-2 variants and SARS-CoV. Thus, the YYDRxG motif represents a common convergent solution for the human humoral immune system to target sarbecoviruses including the Omicron variant. These findings suggest an epitope-targeting strategy to identify potent and broadly neutralizing antibodies for design of pan-sarbecovirus vaccines and antibody therapeutics.

A large-scale systematic survey reveals recurring molecular features of public antibody responses to SARS-CoV-2

Global research to combat the COVID-19 pandemic has led to the isolation and characterization of thousands of human antibodies to the SARS-CoV-2 spike protein, providing an unprecedented opportunity to study the antibody response to a single antigen. Using the information derived from 88 research publications and 13 patents, we assembled a dataset of ∼8,000 human antibodies to the SARS-CoV-2 spike protein from >200 donors. By analyzing immunoglobulin V and D gene usages, complementarity-determining region H3 sequences, and somatic hypermutations, we demonstrated that the common (public) responses to different domains of the spike protein were quite different. We further used these sequences to train a deep-learning model to accurately distinguish between the human antibodies to SARS-CoV-2 spike protein and those to influenza hemagglutinin protein. Overall, this study provides an informative resource for antibody research and enhances our molecular understanding of public antibody responses.

Computational identification of HCV neutralizing antibodies with a common HCDR3 disulfide bond motif in the antibody repertoires of infected individuals

Despite recent success in hepatitis C virus (HCV) treatment using antivirals, an HCV vaccine is still needed to prevent reinfections in treated patients, to avert the emergence of drug-resistant strains, and to provide protection for people with no access to the antiviral therapeutics. The early production of broadly neutralizing antibodies (bNAbs) associates with HCV clearance. Several potent bNAbs bind a conserved HCV glycoprotein E2 epitope using an unusual heavy chain complementarity determining region 3 (HCDR3) containing an intra-loop disulfide bond. Isolation of additional structurally-homologous bNAbs would facilitate the recognition of key determinants of such bNAbs and guide rational vaccine design. Here we report the identification of new antibodies containing an HCDR3 disulfide bond motif using computational screening with the Rosetta software. Using the newly-discovered and already-known members of this antibody family, we review the required HCDR3 amino acid composition and propose determinants for the bent versus straight HCDR3 loop conformation observed in these antibodies.

A large-scale systematic survey of SARS-CoV-2 antibodies reveals recurring molecular features

In the past two years, the global research in combating COVID-19 pandemic has led to isolation and characterization of numerous human antibodies to the SARS-CoV-2 spike. This enormous collection of antibodies provides an unprecedented opportunity to study the antibody response to a single antigen. From mining information derived from 88 research publications and 13 patents, we have assembled a dataset of ∼8,000 human antibodies to the SARS-CoV-2 spike from >200 donors. Analysis of antibody targeting of different domains of the spike protein reveals a number of common (public) responses to SARS-CoV-2, exemplified via recurring IGHV/IGK(L)V pairs, CDR H3 sequences, IGHD usage, and somatic hypermutation. We further present a proof-of-concept for prediction of antigen specificity using deep learning to differentiate sequences of antibodies to SARS-CoV-2 spike and to influenza hemagglutinin. Overall, this study not only provides an informative resource for antibody and vaccine research, but fundamentally advances our molecular understanding of public antibody responses to a viral pathogen.

Convergent antibody responses to the SARS-CoV-2 spike protein in convalescent and vaccinated individuals

Unrelated individuals can produce genetically similar clones of antibodies, known as public clonotypes, which have been seen in responses to different infectious diseases, as well as healthy individuals. Here we identify 37 public clonotypes in memory B cells from convalescent survivors of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection or in plasmablasts from an individual after vaccination with mRNA-encoded spike protein. We identify 29 public clonotypes, including clones recognizing the receptor-binding domain (RBD) in the spike protein S1 subunit (including a neutralizing, angiotensin-converting enzyme 2 [ACE2]-blocking clone that protects in vivo) and others recognizing non-RBD epitopes that bind the S2 domain. Germline-revertant forms of some public clonotypes bind efficiently to spike protein, suggesting these common germline-encoded antibodies are preconfigured for avid recognition. Identification of large numbers of public clonotypes provides insight into the molecular basis of efficacy of SARS-CoV-2 vaccines and sheds light on the immune pressures driving the selection of common viral escape mutants.

Therapeutic alphavirus cross-reactive E1 human antibodies inhibit viral egress

Alphaviruses cause severe arthritogenic or encephalitic disease. The E1 structural glycoprotein is highly conserved in these viruses and mediates viral fusion with host cells. However, the role of antibody responses to the E1 protein in immunity is poorly understood. We isolated E1-specific human monoclonal antibodies (mAbs) with diverse patterns of recognition for alphaviruses (ranging from Eastern equine encephalitis virus [EEEV]-specific to alphavirus cross-reactive) from survivors of natural EEEV infection. Antibody binding patterns and epitope mapping experiments identified differences in E1 reactivity based on exposure of epitopes on the glycoprotein through pH-dependent mechanisms or presentation on the cell surface prior to virus egress. Therapeutic efficacy in vivo of these mAbs corresponded with potency of virus egress inhibition in vitro and did not require Fc-mediated effector functions for treatment against subcutaneous EEEV challenge. These studies reveal the molecular basis for broad and protective antibody responses to alphavirus E1 proteins.

Resistance of SARS-CoV-2 variants to neutralization by antibodies induced in convalescent patients with COVID-19

Administration of convalescent plasma or neutralizing monoclonal antibodies (mAbs) is a potent therapeutic option for coronavirus disease 2019 (COVID-19) caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection. However, SARS-CoV-2 variants with mutations in the spike protein have emerged in many countries. To evaluate the efficacy of neutralizing antibodies induced in convalescent patients against emerging variants, we isolate anti-spike mAbs from two convalescent COVID-19 patients infected with prototypic SARS-CoV-2 by single-cell sorting of immunoglobulin-G-positive (IgG+) memory B cells. Anti-spike antibody induction is robust in these patients, and five mAbs have potent neutralizing activities. The efficacy of most neutralizing mAbs and convalescent plasma samples is maintained against B.1.1.7 and mink cluster 5 variants but is significantly decreased against variants B.1.351 from South Africa and P.1 from Brazil. However, mAbs with a high affinity for the receptor-binding domain remain effective against these neutralization-resistant variants. Rapid spread of these variants significantly impacts antibody-based therapies and vaccine strategies against SARS-CoV-2.

Convergent antibody responses to the SARS-CoV-2 spike protein in convalescent and vaccinated individuals

Unrelated individuals can produce genetically similar clones of antibodies, known as public clonotypes, which have been seen in responses to different infectious diseases as well as healthy individuals. Here we identify 37 public clonotypes in memory B cells from convalescent survivors of SARS-CoV-2 infection or in plasmablasts from an individual after vaccination with mRNA-encoded spike protein. We identified 29 public clonotypes, including clones recognizing the receptor-binding domain (RBD) in the spike protein S1 subunit (including a neutralizing, ACE2-blocking clone that protects in vivo ), and others recognizing non-RBD epitopes that bound the heptad repeat 1 region of the S2 domain. Germline-revertant forms of some public clonotypes bound efficiently to spike protein, suggesting these common germline-encoded antibodies are preconfigured for avid recognition. Identification of large numbers of public clonotypes provides insight into the molecular basis of efficacy of SARS-CoV-2 vaccines and sheds light on the immune pressures driving the selection of common viral escape mutants.

ClonoMatch: a tool for identifying homologous immunoglobulin and T cell receptor sequences in large databases

SUMMARY

B cell receptor (BCR) and T cell receptor (TCR) repertoires are generated through somatic DNA rearrangements and are responsible for the molecular basis of antigen recognition in the immune system. Next-generation sequencing (NGS) of DNA and the falling cost of sequencing due to continued development of these technologies have made sequencing assays an affordable way to characterize the repertoire of adaptive immune receptors (sometimes termed the 'immunome'). Many new workflows have been developed to take advantage of NGS and have placed the resulting immunome datasets in the public domain. The scale of these NGS datasets has made it challenging to search through the Complementarity-determining region 3 (CDR3), which is responsible for imparting specific antibody-antigen interactions. Thus, there is an increasing demand for sequence analysis tools capable of searching through CDR3s from immunome data collections containing millions of sequences. To address this need, we created a software package called ClonoMatch that facilitates rapid searches in bulk immunome data for BCR or TCR sequences based on their CDR3 sequence or V3J clonotype.

AVAILABILITY AND IMPLEMENTATION

Documentation, software support and the codebase are all available at https://github.com/crowelab/clonomatch. This software is distributed under the GPL v3 license.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Rapid isolation and profiling of a diverse panel of human monoclonal antibodies targeting the SARS-CoV-2 spike protein

Antibodies are a principal determinant of immunity for most RNA viruses and have promise to reduce infection or disease during major epidemics. The novel coronavirus SARS-CoV-2 has caused a global pandemic with millions of infections and hundreds of thousands of deaths to date ^1,2 . In response, we used a rapid antibody discovery platform to isolate hundreds of human monoclonal antibodies (mAbs) against the SARS-CoV-2 spike (S) protein. We stratify these mAbs into five major classes based on their reactivity to subdomains of S protein as well as their cross-reactivity to SARS-CoV. Many of these mAbs inhibit infection of authentic SARS-CoV-2 virus, with most neutralizing mAbs recognizing the receptor-binding domain (RBD) of S. This work defines sites of vulnerability on SARS-CoV-2 S and demonstrates the speed and robustness of new antibody discovery methodologies.