How repertoire data are changing antibody science

Addressing the antibody germline bias and its effect on language models for improved antibody design

MOTIVATION

The versatile binding properties of antibodies have made them an extremely important class of biotherapeutics. However, therapeutic antibody development is a complex, expensive, and time-consuming task, with the final antibody needing to not only have strong and specific binding but also be minimally impacted by developability issues. The success of transformer-based language models in protein sequence space and the availability of vast amounts of antibody sequences, has led to the development of many antibody-specific language models to help guide antibody design. Antibody diversity primarily arises from V(D)J recombination, mutations within the CDRs, and/or from a few nongermline mutations outside the CDRs. Consequently, a significant portion of the variable domain of all natural antibody sequences remains germline. This affects the pre-training of antibody-specific language models, where this facet of the sequence data introduces a prevailing bias toward germline residues. This poses a challenge, as mutations away from the germline are often vital for generating specific and potent binding to a target, meaning that language models need be able to suggest key mutations away from germline.

RESULTS

In this study, we explore the implications of the germline bias, examining its impact on both general-protein and antibody-specific language models. We develop and train a series of new antibody-specific language models optimized for predicting nongermline residues. We then compare our final model, AbLang-2, with current models and show how it suggests a diverse set of valid mutations with high cumulative probability.

AVAILABILITY AND IMPLEMENTATION

AbLang-2 is trained on both unpaired and paired data, and is freely available at https://github.com/oxpig/AbLang2.git.

Pre-training with a rational approach for antibody sequence representation

Introduction

Antibodies represent a specific class of proteins produced by the adaptive immune system in response to pathogens. Mining the information embedded in antibody amino acid sequences can benefit both antibody property prediction and novel therapeutic development. However, antibodies possess unique features that should be incorporated using specifically designed training methods, leaving room for improvement in pre-training models for antibody sequences.

Methods

In this study, we present a Pre-trained model of Antibody sequences trained with a Rational Approach for antibodies (PARA). PARA employs a strategy conforming to antibody sequence patterns and an advanced natural language processing self-encoding model structure. This approach addresses the limitations of existing protein pre-training models, which primarily utilize language models without fully considering the differences between protein sequences and language sequences.

Results

We demonstrate PARA's performance on several tasks by comparing it to various published pre-training models of antibodies. The results show that PARA significantly outperforms existing models on these tasks, suggesting that PARA has an advantage in capturing antibody sequence information.

Discussion

The antibody latent representation provided by PARA can substantially facilitate studies in relevant areas. We believe that PARA's superior performance in capturing antibody sequence information offers significant potential for both antibody property prediction and the development of novel therapeutics. PARA is available at https://github.com/xtalpi-xic.

The Observed T Cell Receptor Space database enables paired-chain repertoire mining, coherence analysis, and language modeling

T cell activation is governed through T cell receptors (TCRs), heterodimers of two sequence-variable chains (often an α and β chain) that synergistically recognize antigen fragments presented on cell surfaces. Despite this, there only exist repositories dedicated to collecting single-chain, not paired-chain, TCR sequence data. We addressed this gap by creating the Observed TCR Space (OTS) database, a source of consistently processed and annotated, full-length, paired-chain TCR sequences. Currently, OTS contains 5.35 million redundant (1.63 million non-redundant), predominantly human sequences from across 50 studies and at least 75 individuals. Using OTS, we identify pairing biases, public TCRs, and distinct chain coherence patterns relative to antibodies. We also release a paired-chain TCR language model, providing paired embedding representations and a method for residue in-filling conditional on the partner chain. OTS will be updated as a central community resource and is freely downloadable and available as a web application.

VCAb: a web-tool for structure-guided exploration of antibodies

Motivation

Effective responses against immune challenges require antibodies of different isotypes performing specific effector functions. Structural information on these isotypes is essential to engineer antibodies with desired physico-chemical features of their antigen-binding properties, and optimal developability as potential therapeutics. In silico mutational scanning profiles on antibody structures would further pinpoint candidate mutations for enhancing antibody stability and function. Current antibody structure databases lack consistent annotations of isotypes and structural coverage of 3D antibody structures, as well as computed deep mutation profiles.

Results

The V and C region bearing antibody (VCAb) web-tool is established to clarify these annotations and provides an accessible resource to facilitate antibody engineering and design. VCAb currently provides data on 7,166 experimentally determined antibody structures including both V and C regions from different species. Additionally, VCAb provides annotations of species and isotypes with numbering schemes applied. These information can be interactively queried or downloaded in batch.

Availability and implementation

VCAb is implemented as a R shiny application to enable interactive data interrogation. The online application is freely accessible https://fraternalilab.cs.ucl.ac.uk/VCAb/. The source code to generate the database and the online application is available open-source at https://github.com/Fraternalilab/VCAb.

Accelerating AI for science: open data science for science

Aspirations for artificial intelligence (AI) as a catalyst for scientific discovery are growing. High-profile successes deploying AI in domains such as protein folding have highlighted AI's potential to unlock new frontiers of scientific knowledge. However, the pathway from AI innovation to deployment in research is not linear. Those seeking to drive a new wave of scientific progress through the application of AI require a diffusion engine that can enhance AI adoption across disciplines. Lessons from previous waves of technology change, experiences of deploying AI in real-world contexts and an emerging research agenda from the AI for science community suggest a framework for accelerating AI adoption. This framework requires action to build supply chains of ideas between disciplines; rapidly transfer technological capabilities through open research; create AI tools that empower researchers; and embed effective data stewardship. Together, these interventions can cultivate an environment of open data science that deliver the benefits of AI across the sciences.

Biophysical cartography of the native and human-engineered antibody landscapes quantifies the plasticity of antibody developability

Designing effective monoclonal antibody (mAb) therapeutics faces a multi-parameter optimization challenge known as "developability", which reflects an antibody's ability to progress through development stages based on its physicochemical properties. While natural antibodies may provide valuable guidance for mAb selection, we lack a comprehensive understanding of natural developability parameter (DP) plasticity (redundancy, predictability, sensitivity) and how the DP landscapes of human-engineered and natural antibodies relate to one another. These gaps hinder fundamental developability profile cartography. To chart natural and engineered DP landscapes, we computed 40 sequence- and 46 structure-based DPs of over two million native and human-engineered single-chain antibody sequences. We find lower redundancy among structure-based compared to sequence-based DPs. Sequence DP sensitivity to single amino acid substitutions varied by antibody region and DP, and structure DP values varied across the conformational ensemble of antibody structures. We show that sequence DPs are more predictable than structure-based ones across different machine-learning tasks and embeddings, indicating a constrained sequence-based design space. Human-engineered antibodies localize within the developability and sequence landscapes of natural antibodies, suggesting that human-engineered antibodies explore mere subspaces of the natural one. Our work quantifies the plasticity of antibody developability, providing a fundamental resource for multi-parameter therapeutic mAb design.

RAIN: machine learning-based identification for HIV-1 bNAbs

Broadly neutralizing antibodies (bNAbs) are promising candidates for the treatment and prevention of HIV-1 infections. Despite their critical importance, automatic detection of HIV-1 bNAbs from immune repertoires is still lacking. Here, we develop a straightforward computational method for the Rapid Automatic Identification of bNAbs (RAIN) based on machine learning methods. In contrast to other approaches, which use one-hot encoding amino acid sequences or structural alignment for prediction, RAIN uses a combination of selected sequence-based features for the accurate prediction of HIV-1 bNAbs. We demonstrate the performance of our approach on non-biased, experimentally obtained and sequenced BCR repertoires from HIV-1 immune donors. RAIN processing leads to the successful identification of distinct HIV-1 bNAbs targeting the CD4-binding site of the envelope glycoprotein. In addition, we validate the identified bNAbs using an in vitro neutralization assay and we solve the structure of one of them in complex with the soluble native-like heterotrimeric envelope glycoprotein by single-particle cryo-electron microscopy (cryo-EM). Overall, we propose a method to facilitate and accelerate HIV-1 bNAbs discovery from non-selected immune repertoires.

Prediction of polyspecificity from antibody sequence data by machine learning

Antibodies are generated with great diversity in nature resulting in a set of molecules, each optimized to bind a specific target. Taking advantage of their diversity and specificity, antibodies make up for a large part of recently developed biologic drugs. For therapeutic use antibodies need to fulfill several criteria to be safe and efficient. Polyspecific antibodies can bind structurally unrelated molecules in addition to their main target, which can lead to side effects and decreased efficacy in a therapeutic setting, for example via reduction of effective drug levels. Therefore, we created a neural-network-based model to predict polyspecificity of antibodies using the heavy chain variable region sequence as input. We devised a strategy for enriching antibodies from an immunization campaign either for antigen-specific or polyspecific binding properties, followed by generation of a large sequencing data set for training and cross-validation of the model. We identified important physico-chemical features influencing polyspecificity by investigating the behaviour of this model. This work is a machine-learning-based approach to polyspecificity prediction and, besides increasing our understanding of polyspecificity, it might contribute to therapeutic antibody development.

Disease diagnostics using machine learning of immune receptors

Clinical diagnosis typically incorporates physical examination, patient history, and various laboratory tests and imaging studies, but makes limited use of the human system's own record of antigen exposures encoded by receptors on B cells and T cells. We analyzed immune receptor datasets from 593 individuals to develop MAchine Learning for Immunological Diagnosis (Mal-ID) , an interpretive framework to screen for multiple illnesses simultaneously or precisely test for one condition. This approach detects specific infections, autoimmune disorders, vaccine responses, and disease severity differences. Human-interpretable features of the model recapitulate known immune responses to SARS-CoV-2, Influenza, and HIV, highlight antigen-specific receptors, and reveal distinct characteristics of Systemic Lupus Erythematosus and Type-1 Diabetes autoreactivity. This analysis framework has broad potential for scientific and clinical interpretation of human immune responses.

Innovations and trends in antibody repertoire analysis

Monoclonal antibodies have revolutionized the treatment of human diseases, which has made them the fastest-growing class of therapeutics, with global sales expected to reach $346.6 billion USD by 2028. Advances in antibody engineering and development have led to the creation of increasingly sophisticated antibody-based therapeutics (e.g. bispecific antibodies and chimeric antigen receptor T cells). However, approaches for antibody discovery have remained comparatively grounded in conventional yet reliable in vitro assays. Breakthrough developments in high-throughput single B-cell sequencing and immunoglobulin proteomic serology, however, have enabled the identification of high-affinity antibodies directly from endogenous B cells or circulating immunoglobulin produced in vivo. Moreover, advances in artificial intelligence offer vast potential for antibody discovery and design with large-scale repertoire datasets positioned as the optimal source of training data for such applications. We highlight advances and recent trends in how these technologies are being applied to antibody repertoire analysis.

RAIN: a Machine Learning-based identification for HIV-1 bNAbs

Broadly neutralizing antibodies (bNAbs) are promising candidates for the treatment and prevention of HIV-1 infection. Despite their critical importance, automatic detection of HIV-1 bNAbs from immune repertoire is still lacking. Here, we developed a straightforward computational method for Rapid Automatic Identification of bNAbs (RAIN) based on Machine Learning methods. In contrast to other approaches using one-hot encoding amino acid sequences or structural alignment for prediction, RAIN uses a combination of selected sequence-based features for accurate prediction of HIV-1 bNAbs. We demonstrate the performance of our approach on non-biased, experimentally obtained sequenced BCR repertoires from HIV-1 immune donors. RAIN processing leads to the successful identification of novel HIV-1 bNAbs targeting the CD4-binding site of the envelope glycoprotein. In addition, we validate the identified bNAbs using in vitro neutralization assay and we solve the structure of one of them in complex with the soluble native-like heterotrimeric envelope glycoprotein by single-particle cryo-electron microscopy (cryo-EM). Overall, we propose a method to facilitate and accelerate HIV-1 bNAbs discovery from non-selected immune repertoires.

Computational Analysis of B-Cell Receptor (BCR) Immune Repertoires with Abalign

The widespread application of high-throughput sequencing technology has generated massive sequences of B-cell receptor (BCR) immune repertoires. Computational analysis of these data has gained significant attention due to the increasing importance of immunotherapy and precision medicine. It not only reveals the diversity and dynamic changes in immune responses, contributing to the study of associated diseases, but also provides valuable information for immunodiagnostics and drug development. Recently, we introduced a BCR-specific multiple sequence alignment (MSA) method along with a comprehensive platform software called Abalign, which stands out as an excellent choice for analyzing BCR immune repertoires due to its unique high-throughput processing capability. It offers ultra-fast MSA functionality and a wide range of analytical features, including BCR/antibody extraction, clonal grouping, lineage tree construction, mutation profiling, diversity statistics, VJ gene assignment, antibody humanization, and more. Importantly, users can perform these analyses using the graphical user interface without any programming skills or scripts. In this article, we present a series of protocols that integrate Abalign's analysis modules into a cohesive workflow. This step-by-step workflow provides detailed instructions for software installation, data preparation, and comprehensive analysis of BCR immune repertoires. This workflow facilitates the efficient acquisition of comprehensive results in profiling BCR immune repertoires, offering insights into the impacts of infectious diseases, allergies, autoimmune disorders, tumor immunology, and antibody drugs. Abalign is freely available at http://cao.labshare.cn/abalign/. © 2024 Wiley Periodicals LLC. Basic Protocol 1: Resource preparation Basic Protocol 2: Analyzing BCR immune repertoires Support Protocol 1: Aiding antibody humanization Support Protocol 2: Constructing B-cell lineage trees Alternate Protocol: Running with Linux command line Basic Protocol 3: Comparing BCR immune repertoires.

KA-Search, a method for rapid and exhaustive sequence identity search of known antibodies

Antibodies with similar amino acid sequences, especially across their complementarity-determining regions, often share properties. Finding that an antibody of interest has a similar sequence to naturally expressed antibodies in healthy or diseased repertoires is a powerful approach for the prediction of antibody properties, such as immunogenicity or antigen specificity. However, as the number of available antibody sequences is now in the billions and continuing to grow, repertoire mining for similar sequences has become increasingly computationally expensive. Existing approaches are limited by either being low-throughput, non-exhaustive, not antibody specific, or only searching against entire chain sequences. Therefore, there is a need for a specialized tool, optimized for a rapid and exhaustive search of any antibody region against all known antibodies, to better utilize the full breadth of available repertoire sequences. We introduce Known Antibody Search (KA-Search), a tool that allows for the rapid search of billions of antibody variable domains by amino acid sequence identity across either the variable domain, the complementarity-determining regions, or a user defined antibody region. We show KA-Search in operation on the [Formula: see text]2.4 billion antibody sequences available in the OAS database. KA-Search can be used to find the most similar sequences from OAS within 30 minutes and a representative subset of 10 million sequences in less than 9 seconds. We give examples of how KA-Search can be used to obtain new insights about an antibody of interest. KA-Search is freely available at https://github.com/oxpig/kasearch .

Applications of single-cell RNA sequencing in drug discovery and development

Single-cell technologies, particularly single-cell RNA sequencing (scRNA-seq) methods, together with associated computational tools and the growing availability of public data resources, are transforming drug discovery and development. New opportunities are emerging in target identification owing to improved disease understanding through cell subtyping, and highly multiplexed functional genomics screens incorporating scRNA-seq are enhancing target credentialling and prioritization. ScRNA-seq is also aiding the selection of relevant preclinical disease models and providing new insights into drug mechanisms of action. In clinical development, scRNA-seq can inform decision-making via improved biomarker identification for patient stratification and more precise monitoring of drug response and disease progression. Here, we illustrate how scRNA-seq methods are being applied in key steps in drug discovery and development, and discuss ongoing challenges for their implementation in the pharmaceutical industry.

The ClusPro AbEMap web server for the prediction of antibody epitopes

Antibodies play an important role in the immune system by binding to molecules called antigens at their respective epitopes. These interfaces or epitopes are structural entities determined by the interactions between an antibody and an antigen, making them ideal systems to analyze by using docking programs. Since the advent of high-throughput antibody sequencing, the ability to perform epitope mapping using only the sequence of the antibody has become a high priority. ClusPro, a leading protein-protein docking server, together with its template-based modeling version, ClusPro-TBM, have been re-purposed to map epitopes for specific antibody-antigen interactions by using the Antibody Epitope Mapping server (AbEMap). ClusPro-AbEMap offers three different modes for users depending on the information available on the antibody as follows: (i) X-ray structure, (ii) computational/predicted model of the structure or (iii) only the amino acid sequence. The AbEMap server presents a likelihood score for each antigen residue of being part of the epitope. We provide detailed information on the server's capabilities for the three options and discuss how to obtain the best results. In light of the recent introduction of AlphaFold2 (AF2), we also show how one of the modes allows users to use their AF2-generated antibody models as input. The protocol describes the relative advantages of the server compared to other epitope-mapping tools, its limitations and potential areas of improvement. The server may take 45-90 min depending on the size of the proteins.

Abalign: a comprehensive multiple sequence alignment platform for B-cell receptor immune repertoires

The utilization of high-throughput sequencing (HTS) for B-cell receptor (BCR) immune repertoire analysis has become widespread in the fields of adaptive immunity and antibody drug development. However, the sheer volume of sequences generated by these experiments presents a challenge in data processing. Specifically, multiple sequence alignment (MSA), a critical aspect of BCR analysis, remains inadequate for handling massive BCR sequencing data and lacks the ability to provide immunoglobulin-specific information. To address this gap, we introduce Abalign, a standalone program specifically designed for ultrafast MSA of BCR/antibody sequences. Benchmark tests demonstrate that Abalign achieves comparable or even better accuracy than state-of-the-art MSA tools, and shows remarkable advantages in terms of speed and memory consumption, reducing the time required for high-throughput analysis from weeks to hours. In addition to its alignment capabilities, Abalign offers a broad range of BCR analysis features, including extracting BCRs, constructing lineage trees, assigning VJ genes, analyzing clonotypes, profiling mutations, and comparing BCR immune repertoires. With its user-friendly graphic interface, Abalign can be easily run on personal computers instead of computing clusters. Overall, Abalign is an easy-to-use and effective tool that enables researchers to analyze massive BCR/antibody sequences, leading to new discoveries in the field of immunoinformatics. The software is freely available at http://cao.labshare.cn/abalign/.

Complete variable domain sequences of monoclonal antibody light chains identified from untargeted RNA sequencing data

Introduction

Monoclonal antibody light chain proteins secreted by clonal plasma cells cause tissue damage due to amyloid deposition and other mechanisms. The unique protein sequence associated with each case contributes to the diversity of clinical features observed in patients. Extensive work has characterized many light chains associated with multiple myeloma, light chain amyloidosis and other disorders, which we have collected in the publicly accessible database, AL-Base. However, light chain sequence diversity makes it difficult to determine the contribution of specific amino acid changes to pathology. Sequences of light chains associated with multiple myeloma provide a useful comparison to study mechanisms of light chain aggregation, but relatively few monoclonal sequences have been determined. Therefore, we sought to identify complete light chain sequences from existing high throughput sequencing data.

Methods

We developed a computational approach using the MiXCR suite of tools to extract complete rearranged IGVL-IGJL sequences from untargeted RNA sequencing data. This method was applied to whole-transcriptome RNA sequencing data from 766 newly diagnosed patients in the Multiple Myeloma Research Foundation CoMMpass study.

Results

Monoclonal IGVL-IGJL sequences were defined as those where >50% of assigned IGK or IGL reads from each sample mapped to a unique sequence. Clonal light chain sequences were identified in 705/766 samples from the CoMMpass study. Of these, 685 sequences covered the complete IGVL-IGJL region. The identity of the assigned sequences is consistent with their associated clinical data and with partial sequences previously determined from the same cohort of samples. Sequences have been deposited in AL-Base.

Discussion

Our method allows routine identification of clonal antibody sequences from RNA sequencing data collected for gene expression studies. The sequences identified represent, to our knowledge, the largest collection of multiple myeloma-associated light chains reported to date. This work substantially increases the number of monoclonal light chains known to be associated with non-amyloid plasma cell disorders and will facilitate studies of light chain pathology.

Characterisation of the immune repertoire of a humanised transgenic mouse through immunophenotyping and high-throughput sequencing

Immunoglobulin loci-transgenic animals are widely used in antibody discovery and increasingly in vaccine response modelling. In this study, we phenotypically characterised B-cell populations from the Intelliselect Transgenic mouse (Kymouse) demonstrating full B-cell development competence. Comparison of the naïve B-cell receptor (BCR) repertoires of Kymice BCRs, naïve human, and murine BCR repertoires revealed key differences in germline gene usage and junctional diversification. These differences result in Kymice having CDRH3 length and diversity intermediate between mice and humans. To compare the structural space explored by CDRH3s in each species' repertoire, we used computational structure prediction to show that Kymouse naïve BCR repertoires are more human-like than mouse-like in their predicted distribution of CDRH3 shape. Our combined sequence and structural analysis indicates that the naïve Kymouse BCR repertoire is diverse with key similarities to human repertoires, while immunophenotyping confirms that selected naïve B cells are able to go through complete development.

Understanding repertoire sequencing data through a multiscale computational model of the germinal center

Sequencing of B-cell and T-cell immune receptor repertoires helps us to understand the adaptive immune response, although it only provides information about the clonotypes (lineages) and their frequencies and not about, for example, their affinity or antigen (Ag) specificity. To further characterize the identified clones, usually with special attention to the particularly abundant ones (dominant), additional time-consuming or expensive experiments are generally required. Here, we present an extension of a multiscale model of the germinal center (GC) that we previously developed to gain more insight in B-cell repertoires. We compare the extent that these simulated repertoires deviate from experimental repertoires established from single GCs, blood, or tissue. Our simulations show that there is a limited correlation between clonal abundance and affinity and that there is large affinity variability among same-ancestor (same-clone) subclones. Our simulations suggest that low-abundance clones and subclones, might also be of interest since they may have high affinity for the Ag. We show that the fraction of plasma cells (PCs) with high B-cell receptor (BcR) mRNA content in the GC does not significantly affect the number of dominant clones derived from single GCs by sequencing BcR mRNAs. Results from these simulations guide data interpretation and the design of follow-up experiments.

Assessing developability early in the discovery process for novel biologics

Beyond potency, a good developability profile is a key attribute of a biological drug. Selecting and screening for such attributes early in the drug development process can save resources and avoid costly late-stage failures. Here, we review some of the most important developability properties that can be assessed early on for biologics. These include the influence of the source of the biologic, its biophysical and pharmacokinetic properties, and how well it can be expressed recombinantly. We furthermore present in silico, in vitro, and in vivo methods and techniques that can be exploited at different stages of the discovery process to identify molecules with liabilities and thereby facilitate the selection of the most optimal drug leads. Finally, we reflect on the most relevant developability parameters for injectable versus orally delivered biologics and provide an outlook toward what general trends are expected to rise in the development of biologics.

Magnetic Nanoparticle-Based Semi-automated Panning for High-Throughput Antibody Selection

Bio-panning is a common process involved in recombinant antibody selection against defined targets. The biopanning process aims to isolate specific antibodies against an antigen via affinity selection from a phage display library. In general, antigens are immobilized on solid surfaces such as polystyrene plastic, magnetic beads, and nitrocellulose. For high-throughput selection, semi-automated panning selection allows simultaneous panning against multiple target antigens adapting automated particle processing systems such as the KingFisher Flex. The system setup allows for minimal human intervention for pre- and post-panning steps such as antigen immobilization, phage rescue, and amplification. In addition, the platform is also adaptable to perform polyclonal and monoclonal ELISA for the evaluation process. This chapter will detail the protocols involved from the selection stage until the monoclonal ELISA evaluation with important notes attached at the end of this chapter for optimization and troubleshooting purposes.

Designing antibodies as therapeutics

Antibody therapeutics are a large and rapidly expanding drug class providing major health benefits. We provide a snapshot of current antibody therapeutics including their formats, common targets, therapeutic areas, and routes of administration. Our focus is on selected emerging directions in antibody design where progress may provide a broad benefit. These topics include enhancing antibodies for cancer, antibody delivery to organs such as the brain, gastrointestinal tract, and lungs, plus antibody developability challenges including immunogenicity risk assessment and mitigation and subcutaneous delivery. Machine learning has the potential, albeit as yet largely unrealized, for a transformative future impact on antibody discovery and engineering.

Artificial intelligence for antibody reading comprehension: AntiBERTa

Utilizing publicly available antibody big data resources, Leem et al. (2022) developed an antibody-specific language model, AntiBERTa, to understand the "language" of antibodies. Case studies reveal that AntiBERTa can be an extremely useful tool for antibody engineering.

Deciphering the language of antibodies using self-supervised learning

An individual’s B cell receptor (BCR) repertoire encodes information about past immune responses and potential for future disease protection. Deciphering the information stored in BCR sequence datasets will transform our understanding of disease and enable discovery of novel diagnostics and antibody therapeutics. A key challenge of BCR sequence analysis is the prediction of BCR properties from their amino acid sequence alone. Here, we present an antibody-specific language model, Antibody-specific Bidirectional Encoder Representation from Transformers (AntiBERTa), which provides a contextualized representation of BCR sequences. Following pre-training, we show that AntiBERTa embeddings capture biologically relevant information, generalizable to a range of applications. As a case study, we fine-tune AntiBERTa to predict paratope positions from an antibody sequence, outperforming public tools across multiple metrics. To our knowledge, AntiBERTa is the deepest protein-family-specific language model, providing a rich representation of BCRs. AntiBERTa embeddings are primed for multiple downstream tasks and can improve our understanding of the language of antibodies.

•

AntiBERTa is an antibody-specific transformer model for representation learning

•

AntiBERTa embeddings capture aspects of antibody function

•

Attention maps of AntiBERTa correspond to structural contacts and binding sites

•

AntiBERTa can be fine-tuned for state-of-the-art paratope prediction

Understanding antibody function is critical for deciphering the biology of disease and for the discovery of novel therapeutic antibodies. The challenge is the vast diversity of antibody variants compared with the limited labeled data available. We overcome this challenge by using self-supervised learning to train a large antibody-specific language model, followed by transfer learning, to fine-tune the model for predicting information related to antibody function. We initially demonstrate the success of the model by providing leading results in antibody binding site prediction. The model is amenable to further fine-tuning for diverse applications to improve our understanding of antibody function.

Computational epitope binning reveals functional equivalence of sequence-divergent paratopes

•

Epitope binning groups target-specific protein binders recognizing the same binding region.

•

The “Epibin” method utilizes docking models to computationally predict competition and identify bins.

•

Epibin recapitulated binding competition of repebody variants as determined by immunoassays.

•

In addition, Epibin enabled identification of ‘paratope-equivalent’ residues in sequence-dissimilar variants.

•

Computational epitope binning can scale to allow characterization of entire antigen-specific antibody repertoires.

The therapeutic efficacy of a protein binder largely depends on two factors: its binding site and its binding affinity. Advances in in vitro library display screening and next-generation sequencing have enabled accelerated development of strong binders, yet identifying their binding sites still remains a major challenge. The differentiation, or “binning”, of binders into different groups that recognize distinct binding sites on their target is a promising approach that facilitates high-throughput screening of binders that may show different biological activity. Here we study the extent to which the information contained in the amino acid sequences comprising a set of target-specific binders can be leveraged to bin them, inferring functional equivalence of their binding regions, or paratopes, based directly on comparison of the sequences, their modeled structures, or their modeled interactions. Using a leucine-rich repeat binding scaffold known as a “repebody” as the source of diversity in recognition against interleukin-6 (IL-6), we show that the “Epibin” approach introduced here effectively utilized structural modelling and docking to extract specificity information encoded in the repebody amino acid sequences and thereby successfully recapitulate IL-6 binding competition observed in immunoassays. Furthermore, our computational binning provided a basis for designing in vitro mutagenesis experiments to pinpoint specificity-determining residues. Finally, we demonstrate that the Epibin approach can extend to antibodies, retrospectively comparing its predictions to results from antigen-specific antibody competition studies. The study thus demonstrates the utility of modeling structure and binding from the amino acid sequences of different binders against the same target, and paves the way for larger-scale binning and analysis of entire repertoires.

Current advances in biopharmaceutical informatics: guidelines, impact and challenges in the computational developability assessment of antibody therapeutics

Therapeutic monoclonal antibodies and their derivatives are key components of clinical pipelines in the global biopharmaceutical industry. The availability of large datasets of antibody sequences, structures, and biophysical properties is increasingly enabling the development of predictive models and computational tools for the "developability assessment" of antibody drug candidates. Here, we provide an overview of the antibody informatics tools applicable to the prediction of developability issues such as stability, aggregation, immunogenicity, and chemical degradation. We further evaluate the opportunities and challenges of using biopharmaceutical informatics for drug discovery and optimization. Finally, we discuss the potential of developability guidelines based on in silico metrics that can be used for the assessment of antibody stability and manufacturability.

A perspective toward mass spectrometry-based de novo sequencing of endogenous antibodies

A key step in therapeutic and endogenous humoral antibody characterization is identifying the amino acid sequence. So far, this task has been mainly tackled through sequencing of B-cell receptor (BCR) repertoires at the nucleotide level. Mass spectrometry (MS) has emerged as an alternative tool for obtaining sequence information directly at the – most relevant – protein level. Although several MS methods are now well established, analysis of recombinant and endogenous antibodies comes with a specific set of challenges, requiring approaches beyond the conventional proteomics workflows. Here, we review the challenges in MS-based sequencing of both recombinant as well as endogenous humoral antibodies and outline state-of-the-art methods attempting to overcome these obstacles. We highlight recent examples and discuss remaining challenges. We foresee a great future for these approaches making de novo antibody sequencing and discovery by MS-based techniques feasible, even for complex clinical samples from endogenous sources such as serum and other liquid biopsies.

Epitope profiling using computational structural modelling demonstrated on coronavirus-binding antibodies

Identifying the epitope of an antibody is a key step in understanding its function and its potential as a therapeutic. Sequence-based clonal clustering can identify antibodies with similar epitope complementarity, however, antibodies from markedly different lineages but with similar structures can engage the same epitope. We describe a novel computational method for epitope profiling based on structural modelling and clustering. Using the method, we demonstrate that sequence dissimilar but functionally similar antibodies can be found across the Coronavirus Antibody Database, with high accuracy (92% of antibodies in multiple-occupancy structural clusters bind to consistent domains). Our approach functionally links antibodies with distinct genetic lineages, species origins, and coronavirus specificities. This indicates greater convergence exists in the immune responses to coronaviruses than is suggested by sequence-based approaches. Our results show that applying structural analytics to large class-specific antibody databases will enable high confidence structure-function relationships to be drawn, yielding new opportunities to identify functional convergence hitherto missed by sequence-only analysis.

Antibodies are a key component of the immune system that combat pathogens by binding to a defined region of their molecular surface (known as an ‘epitope’). The ability to map which antibodies target the same epitopes is crucial when designing non-competing antibody therapeutics or predicting the influence of pathogen mutation on population immunity. While one can use laboratory experiments to deduce when pairs of antibodies engage the same epitope, such experiments are very expensive and time consuming if used to compare on the order of thousands of antibodies. In this work, we report a new computational algorithm (SPACE) that clusters antibodies that target the same epitope based on their predicted 3D structure, as binding site structure is a property often conserved between binders complementary to the same epitope. Unlike existing antibody epitope profiling tools which assume two antibodies must share a high sequence identity/similar genetic basis to engage the same region, our orthogonal method can detect broader patterns of convergent evolution across binders to different pathogen strains, and between antibodies with different genetic and even species origins.

Observed Antibody Space: A diverse database of cleaned, annotated, and translated unpaired and paired antibody sequences

The antibody repertoires of individuals and groups have been used to explore disease states, understand vaccine responses, and drive therapeutic development. The arrival of B‐cell receptor repertoire sequencing has enabled researchers to get a snapshot of these antibody repertoires, and as more data are generated, increasingly in‐depth studies are possible. However, most publicly available data only exist as raw FASTQ files, making the data hard to access, process, and compare. The Observed Antibody Space (OAS) database was created in 2018 to offer clean, annotated, and translated repertoire data. In this paper, we describe an update to OAS that has been driven by the increasing volume of data and the appearance of paired (VH/VL) sequence data. OAS is now accessible via a new web server, with standardized search parameters and a new sequence‐based search option. The new database provides both nucleotides and amino acids for every sequence, with additional sequence annotations to make the data Minimal Information about Adaptive Immune Receptor Repertoire compliant, and comments on potential problems with the sequence. OAS now contains 25 new studies, including severe acute respiratory syndrome coronavirus 2 data and paired sequencing data. The new database is accessible at http://opig.stats.ox.ac.uk/webapps/oas/, and all data are freely available for download.

Antibody Repertoire Analysis of Tumor-Infiltrating B Cells Reveals Distinct Signatures and Distributions Across Tissues

The role of B cells in the tumor microenvironment (TME) has largely been under investigated, and data regarding the antibody repertoire encoded by B cells in the TME and the adjacent lymphoid organs are scarce. Here, we utilized B cell receptor high-throughput sequencing (BCR-Seq) to profile the antibody repertoire signature of tumor-infiltrating lymphocyte B cells (TIL−Bs) in comparison to B cells from three anatomic compartments in a mouse model of triple-negative breast cancer. We found that TIL-Bs exhibit distinct antibody repertoire measures, including high clonal polarization and elevated somatic hypermutation rates, suggesting a local antigen-driven B-cell response. Importantly, TIL-Bs were highly mutated but non-class switched, suggesting that class-switch recombination may be inhibited in the TME. Tracing the distribution of TIL-B clones across various compartments indicated that they migrate to and from the TME. The data thus suggests that antibody repertoire signatures can serve as indicators for identifying tumor-reactive B cells.

Functional and in silico Characterization of Neutralizing Interactions Between Antibodies and the Foot-and-Mouth Disease Virus Immunodominant Antigenic Site

Molecular knowledge of virus–antibody interactions is essential for the development of better vaccines and for a timely assessment of the spread and severity of epidemics. For foot-and-mouth disease virus (FMDV) research, in particular, computational methods for antigen–antibody (Ag–Ab) interaction, and cross-antigenicity characterization and prediction are critical to design engineered vaccines with robust, long-lasting, and wider response against different strains. We integrated existing structural modeling and prediction algorithms to study the surface properties of FMDV Ags and Abs and their interaction. First, we explored four modeling and two Ag–Ab docking methods and implemented a computational pipeline based on a reference Ag–Ab structure for FMDV of serotype C, to be used as a source protocol for the study of unknown interaction pairs of Ag–Ab. Next, we obtained the variable region sequence of two monoclonal IgM and IgG antibodies that recognize and neutralize antigenic site A (AgSA) epitopes from South America serotype A FMDV and developed two peptide ELISAs for their fine epitope mapping. Then, we applied the previous Ag–Ab molecular structure modeling and docking protocol further scored by functional peptide ELISA data. This work highlights a possible different behavior in the immune response of IgG and IgM Ab isotypes. The present method yielded reliable Ab models with differential paratopes and Ag interaction topologies in concordance with their isotype classes. Moreover, it demonstrates the applicability of computational prediction techniques to the interaction phenomena between the FMDV immunodominant AgSA and Abs, and points out their potential utility as a metric for virus-related, massive Ab repertoire analysis or as a starting point for recombinant vaccine design.

Animal immunization merges with innovative technologies: A new paradigm shift in antibody discovery

Animal-derived antibody sources, particularly, transgenic mice that are engineered with human immunoglobulin loci, along with advanced antibody generation technology platforms have facilitated the discoveries of human antibody therapeutics. For example, isolation of antigen-specific B cells, microfluidics, and next-generation sequencing have emerged as powerful tools for identifying and developing monoclonal antibodies (mAbs). These technologies enable not only antibody drug discovery but also lead to the understanding of B cell biology, immune mechanisms and immunogenetics of antibodies. In this perspective article, we discuss the scientific merits of animal immunization combined with advanced methods for antibody generation as compared to animal-free alternatives through in-vitro-generated antibody libraries. The knowledge gained from animal-derived antibodies concerning the recombinational diversity, somatic hypermutation patterns, and physiochemical properties is found more valuable and prerequisite for developing in vitro libraries, as well as artificial intelligence/machine learning methods to discover safe and effective mAbs.

Protein design and variant prediction using autoregressive generative models

The ability to design functional sequences and predict effects of variation is central to protein engineering and biotherapeutics. State-of-art computational methods rely on models that leverage evolutionary information but are inadequate for important applications where multiple sequence alignments are not robust. Such applications include the prediction of variant effects of indels, disordered proteins, and the design of proteins such as antibodies due to the highly variable complementarity determining regions. We introduce a deep generative model adapted from natural language processing for prediction and design of diverse functional sequences without the need for alignments. The model performs state-of-art prediction of missense and indel effects and we successfully design and test a diverse 105-nanobody library that shows better expression than a 1000-fold larger synthetic library. Our results demonstrate the power of the alignment-free autoregressive model in generalizing to regions of sequence space traditionally considered beyond the reach of prediction and design.

Immune repertoire: Revealing the "real-time" adaptive immune response in autoimmune diseases

The diversity of the immune repertoire (IR) enables the human immune system to distinguish multifarious antigens (Ags) that humans may encounter throughout life. At the same time, bias or abnormalities in the IR also pay a contribution to the pathogenesis of autoimmune diseases. Rapid advancements in high-throughput sequencing (HTS) technology have ushered in a new era of immune studies, revealing novel molecules and pathways that might result in autoimmunity. In the field of IR, HTS can monitor the immune response status and identify disease-specific immune repertoires. In this review, we summarize updated progress on the mechanisms of the IR and current related studies on four autoimmune diseases, particularly focusing on systemic lupus erythematosus (SLE). These autoimmune diseases can exhibit slightly or significantly skewed IRs and provide novel insights that inform our comprehending of disease pathogenesis and provide potential targets for diagnosis and treatment.

Progress toward improved understanding of antibody maturation

Upon encountering an antigen, antibodies mature through various rounds of somatic mutations, resulting in higher affinities and specificities to the particular antigen. We review recent progress in four areas of antibody maturation studies. (1) Next-generation and single-cell sequencing have revolutionized the analysis of antibody repertoires by dramatically increasing the sequences available to study the state and evolution of the immune system. Computational methods, including machine learning tools, have been developed for reconstituting antibody clonal lineages and for general repertoire analysis. (2) The availability of X-ray structures, thermodynamic and kinetic data, and molecular dynamics simulations provide information on the biophysical mechanisms responsible for improved affinity. (3) In addition to improved binding to a specific antigen, providing affinity-independent diversity and self/nonself discrimination are fundamental functions of the immune system. Recent studies, including X-ray structures, yield improved understanding of both mechanisms. (4) Results from in vivo maturation help to develop methods of in vitro maturation to improve antibody properties for therapeutic applications, frequently combining computational and experimental approaches.