Antibody structure prediction using interpretable deep learning

Artificial intelligence-driven systems engineering for next-generation plant-derived biopharmaceuticals

Recombinant biopharmaceuticals including antigens, antibodies, hormones, cytokines, single-chain variable fragments, and peptides have been used as vaccines, diagnostics and therapeutics. Plant molecular pharming is a robust platform that uses plants as an expression system to produce simple and complex recombinant biopharmaceuticals on a large scale. Plant system has several advantages over other host systems such as humanized expression, glycosylation, scalability, reduced risk of human or animal pathogenic contaminants, rapid and cost-effective production. Despite many advantages, the expression of recombinant proteins in plant system is hindered by some factors such as non-human post-translational modifications, protein misfolding, conformation changes and instability. Artificial intelligence (AI) plays a vital role in various fields of biotechnology and in the aspect of plant molecular pharming, a significant increase in yield and stability can be achieved with the intervention of AI-based multi-approach to overcome the hindrance factors. Current limitations of plant-based recombinant biopharmaceutical production can be circumvented with the aid of synthetic biology tools and AI algorithms in plant-based glycan engineering for protein folding, stability, viability, catalytic activity and organelle targeting. The AI models, including but not limited to, neural network, support vector machines, linear regression, Gaussian process and regressor ensemble, work by predicting the training and experimental data sets to design and validate the protein structures thereby optimizing properties such as thermostability, catalytic activity, antibody affinity, and protein folding. This review focuses on, integrating systems engineering approaches and AI-based machine learning and deep learning algorithms in protein engineering and host engineering to augment protein production in plant systems to meet the ever-expanding therapeutics market.

A PD-L1xCD3 bispecific nanobody as a novel T-cell engager in treating PD-L1 overexpression melanoma

The development of Immune checkpoint blockade(ICB) therapy and BRAF- and MEK-targeted therapies has reshaped the survival outcomes of the patients with advanced melanoma. PD-1/PD-L1 blockade was an approved strategy in melanoma treatment. Here we design a PD-L1 xCD3 nanobody as a novel bispecific T cell engager (BiTE) in treating PD-L1 overexpression melanoma. BiTE PD-L1×CD3 Nb was predicted to bind near a large acidic surface on CD3-ε similar to UCHT1-scFv antibody based on alpha-fold and molecular docking. BiTE PD-L1×CD3 Nb and anti-CD3 Nb retained the ability to activate T cells to produce TNF-α and IFN-γ in a dose-dependent manner. The IC50 value of BiTE PD-L1×CD3 Nb was 4.208μg/mL. BiTE PD-L1×CD3 Nb showed obvious cytotoxic activity on both A375WT and A375PD-L1 related to PD-L1 expression level.

Structure and Dynamics Guiding Design of Antibody Therapeutics and Vaccines

Antibodies and other new antibody-like formats have emerged as one of the most rapidly growing classes of biotherapeutic proteins. Understanding the structural features that drive antibody function and, consequently, their molecular recognition is critical for engineering antibodies. Here, we present the structural architecture of conventional IgG antibodies alongside other formats. We emphasize the importance of considering antibodies as conformational ensembles in solution instead of focusing on single-static structures because their functions and properties are strongly governed by their dynamic nature. Thus, in this review, we provide an overview of the unique structural and dynamic characteristics of antibodies with respect to their antigen recognition, biophysical properties, and effector functions. We highlight the numerous technical advances in antibody structure prediction and design, enabled by the vast number of experimentally determined high-quality structures recorded with cryo-EM, NMR, and X-ray crystallography. Lastly, we assess antibody and vaccine design strategies in the context of structure and dynamics.

Deep learning in preclinical antibody drug discovery and development

Antibody drugs have become a key part of biotherapeutics. Patients suffering from various diseases have benefited from antibody therapies. However, its development process is rather long, expensive and risky. To speed up the process, reduce cost and improve success rate, artificial intelligence, especially deep learning methods, have been widely used in all aspects of preclinical antibody drug development, from library generation to hit identification, developability screening, lead selection and optimization. In this review, we systematically summarize antibody encodings, deep learning architectures and models used in preclinical antibody drug discovery and development. We also critically discuss challenges and opportunities, problems and possible solutions, current applications and future directions of deep learning in antibody drug development.

Accelerating antibody discovery and design with artificial intelligence: Recent advances and prospects

Therapeutic antibodies are the largest class of biotherapeutics and have been successful in treating human diseases. However, the design and discovery of antibody drugs remains challenging and time-consuming. Recently, artificial intelligence technology has had an incredible impact on antibody design and discovery, resulting in significant advances in antibody discovery, optimization, and developability. This review summarizes major machine learning (ML) methods and their applications for computational predictors of antibody structure and antigen interface/interaction, as well as the evaluation of antibody developability. Additionally, this review addresses the current status of ML-based therapeutic antibodies under preclinical and clinical phases. While many challenges remain, ML may offer a new therapeutic option for the future direction of fully computational antibody design.

Improved computational epitope profiling using structural models identifies a broader diversity of antibodies that bind to the same epitope

The function of an antibody is intrinsically linked to the epitope it engages. Clonal clustering methods, based on sequence identity, are commonly used to group antibodies that will bind to the same epitope. However, such methods neglect the fact that antibodies with highly diverse sequences can exhibit similar binding site geometries and engage common epitopes. In a previous study, we described SPACE1, a method that structurally clustered antibodies in order to predict their epitopes. This methodology was limited by the inaccuracies and incomplete coverage of template-based modeling. In addition, it was only benchmarked at the level of domain-consistency on one virus class. Here, we present SPACE2, which uses the latest machine learning-based structure prediction technology combined with a novel clustering protocol, and benchmark it on binding data that have epitope-level resolution. On six diverse sets of antigen-specific antibodies, we demonstrate that SPACE2 accurately clusters antibodies that engage common epitopes and achieves far higher dataset coverage than clonal clustering and SPACE1. Furthermore, we show that the functionally consistent structural clusters identified by SPACE2 are even more diverse in sequence, genetic lineage, and species origin than those found by SPACE1. These results reiterate that structural data improve our ability to identify antibodies that bind to the same epitope, adding information to sequence-based methods, especially in datasets of antibodies from diverse sources. SPACE2 is openly available on GitHub (https://github.com/oxpig/SPACE2).

Evaluation of Molecular Simulations and Deep Learning Prediction of Antibodies' Recognition of TRBC1 and TRBC2

T cell receptor β-chain constant (TRBC) is a promising class of cancer targets consisting of two highly homologous proteins, TRBC1 and TRBC2. Developing targeted antibody therapeutics against TRBC1 or TRBC2 is expected to eradicate the malignant T cells and preserve half of the normal T cells. Recently, several antibody engineering strategies have been used to modulate the TRBC1 and TRBC2 specificity of antibodies. Here, we used molecular simulation and artificial intelligence methods to quantify the affinity difference in antibodies with various mutations for TRBC1 and TRBC2. The affinity of the existing mutants was verified by FEP calculations aided by the AI. We also performed long-time molecular dynamics simulations to reveal the dynamical antigen recognition mechanisms of the TRBC antibodies.

Computational Methods in Immunology and Vaccinology: Design and Development of Antibodies and Immunogens

The design of new biomolecules able to harness immune mechanisms for the treatment of diseases is a prime challenge for computational and simulative approaches. For instance, in recent years, antibodies have emerged as an important class of therapeutics against a spectrum of pathologies. In cancer, immune-inspired approaches are witnessing a surge thanks to a better understanding of tumor-associated antigens and the mechanisms of their engagement or evasion from the human immune system. Here, we provide a summary of the main state-of-the-art computational approaches that are used to design antibodies and antigens, and in parallel, we review key methodologies for epitope identification for both B- and T-cell mediated responses. A special focus is devoted to the description of structure- and physics-based models, privileged over purely sequence-based approaches. We discuss the implications of novel methods in engineering biomolecules with tailored immunological properties for possible therapeutic uses. Finally, we highlight the extraordinary challenges and opportunities presented by the possible integration of structure- and physics-based methods with emerging Artificial Intelligence technologies for the prediction and design of novel antigens, epitopes, and antibodies.

Simplifying complex antibody engineering using machine learning

Machine learning is transforming antibody engineering by enabling the generation of drug-like monoclonal antibodies with unprecedented efficiency. Unsupervised algorithms trained on massive and diverse protein sequence datasets facilitate the prediction of panels of antibody variants with native-like intrinsic properties (e.g., high stability), greatly reducing the amount of subsequent experimentation needed to identify specific candidates that also possess desired extrinsic properties (e.g., high affinity). Additionally, supervised algorithms, which are trained on deep sequencing datasets obtained after enrichment of in vitro antibody libraries for one or more specific extrinsic properties, enable the prediction of antibody variants with desired combinations of extrinsic properties without the need for additional screening. Here we review recent advances using both machine learning approaches and how they are impacting the field of antibody engineering as well as key outstanding challenges and opportunities for these paradigm-changing methods.

Improved HIV-1 neutralization breadth and potency of V2-apex antibodies by in silico design

Broadly neutralizing antibodies (bNAbs) against HIV can reduce viral transmission in humans, but an effective therapeutic will require unusually high breadth and potency of neutralization. We employ the OSPREY computational protein design software to engineer variants of two apex-directed bNAbs, PGT145 and PG9RSH, resulting in increases in potency of over 100-fold against some viruses. The top designed variants improve neutralization breadth from 39% to 54% at clinically relevant concentrations (IC80 < 1 μg/mL) and improve median potency (IC80) by up to 4-fold over a cross-clade panel of 208 strains. To investigate the mechanisms of improvement, we determine cryoelectron microscopy structures of each variant in complex with the HIV envelope trimer. Surprisingly, we find the largest increases in breadth to be a result of optimizing side-chain interactions with highly variable epitope residues. These results provide insight into mechanisms of neutralization breadth and inform strategies for antibody design and improvement.

Structural delineation and computational design of SARS-CoV-2-neutralizing antibodies against Omicron subvariants

SARS-CoV-2 Omicron subvariants have evolved to evade receptor-binding site (RBS) antibodies that exist in diverse individuals as public antibody clones. We rationally selected RBS antibodies resilient to mutations in emerging Omicron subvariants. Y489 was identified as a site of virus vulnerability and a common footprint of broadly neutralizing antibodies against the subvariants. Multiple Y489-binding antibodies were encoded by public clonotypes and additionally recognized F486, potentially accounting for the emergence of Omicron subvariants harboring the F486V mutation. However, a subclass of antibodies broadly neutralized BA.4/BA.5 variants via hydrophobic binding sites of rare clonotypes along with high mutation-resilience under escape mutation screening. A computationally designed antibody based on one of the Y489-binding antibodies, NIV-10/FD03, was able to bind XBB with any 486 mutation and neutralized XBB.1.5. The structural basis for the mutation-resilience of this Y489-binding antibody group may provide important insights into the design of therapeutics resistant to viral escape.

Structural modeling of antibody variable regions using deep learning-progress and perspectives on drug discovery

AlphaFold2 has hallmarked a generational improvement in protein structure prediction. In particular, advances in antibody structure prediction have provided a highly translatable impact on drug discovery. Though AlphaFold2 laid the groundwork for all proteins, antibody-specific applications require adjustments tailored to these molecules, which has resulted in a handful of deep learning antibody structure predictors. Herein, we review the recent advances in antibody structure prediction and relate them to their role in advancing biologics discovery.

Structure-based neural network protein-carbohydrate interaction predictions at the residue level

Carbohydrates dynamically and transiently interact with proteins for cell-cell recognition, cellular differentiation, immune response, and many other cellular processes. Despite the molecular importance of these interactions, there are currently few reliable computational tools to predict potential carbohydrate-binding sites on any given protein. Here, we present two deep learning (DL) models named CArbohydrate-Protein interaction Site IdentiFier (CAPSIF) that predicts non-covalent carbohydrate-binding sites on proteins: (1) a 3D-UNet voxel-based neural network model (CAPSIF:V) and (2) an equivariant graph neural network model (CAPSIF:G). While both models outperform previous surrogate methods used for carbohydrate-binding site prediction, CAPSIF:V performs better than CAPSIF:G, achieving test Dice scores of 0.597 and 0.543 and test set Matthews correlation coefficients (MCCs) of 0.599 and 0.538, respectively. We further tested CAPSIF:V on AlphaFold2-predicted protein structures. CAPSIF:V performed equivalently on both experimentally determined structures and AlphaFold2-predicted structures. Finally, we demonstrate how CAPSIF models can be used in conjunction with local glycan-docking protocols, such as GlycanDock, to predict bound protein-carbohydrate structures.

ImmuneBuilder: Deep-Learning models for predicting the structures of immune proteins

Immune receptor proteins play a key role in the immune system and have shown great promise as biotherapeutics. The structure of these proteins is critical for understanding their antigen binding properties. Here, we present ImmuneBuilder, a set of deep learning models trained to accurately predict the structure of antibodies (ABodyBuilder2), nanobodies (NanoBodyBuilder2) and T-Cell receptors (TCRBuilder2). We show that ImmuneBuilder generates structures with state of the art accuracy while being far faster than AlphaFold2. For example, on a benchmark of 34 recently solved antibodies, ABodyBuilder2 predicts CDR-H3 loops with an RMSD of 2.81Å, a 0.09Å improvement over AlphaFold-Multimer, while being over a hundred times faster. Similar results are also achieved for nanobodies, (NanoBodyBuilder2 predicts CDR-H3 loops with an average RMSD of 2.89Å, a 0.55Å improvement over AlphaFold2) and TCRs. By predicting an ensemble of structures, ImmuneBuilder also gives an error estimate for every residue in its final prediction. ImmuneBuilder is made freely available, both to download ( https://github.com/oxpig/ImmuneBuilder ) and to use via our webserver ( http://opig.stats.ox.ac.uk/webapps/newsabdab/sabpred ). We also make available structural models for ~150 thousand non-redundant paired antibody sequences ( https://doi.org/10.5281/zenodo.7258553 ).

Structural Modeling of Nanobodies: A Benchmark of State-of-the-Art Artificial Intelligence Programs

The number of applications for nanobodies is steadily expanding, positioning these molecules as fast-growing biologic products in the biotechnology market. Several of their applications require protein engineering, which in turn would greatly benefit from having a reliable structural model of the nanobody of interest. However, as with antibodies, the structural modeling of nanobodies is still a challenge. With the rise of artificial intelligence (AI), several methods have been developed in recent years that attempt to solve the problem of protein modeling. In this study, we have compared the performance in nanobody modeling of several state-of-the-art AI-based programs, either designed for general protein modeling, such as AlphaFold2, OmegaFold, ESMFold, and Yang-Server, or specifically designed for antibody modeling, such as IgFold, and Nanonet. While all these programs performed rather well in constructing the nanobody framework and CDRs 1 and 2, modeling CDR3 still represents a big challenge. Interestingly, tailoring an AI method for antibody modeling does not necessarily translate into better results for nanobodies.

Fast, accurate antibody structure prediction from deep learning on massive set of natural antibodies

Antibodies have the capacity to bind a diverse set of antigens, and they have become critical therapeutics and diagnostic molecules. The binding of antibodies is facilitated by a set of six hypervariable loops that are diversified through genetic recombination and mutation. Even with recent advances, accurate structural prediction of these loops remains a challenge. Here, we present IgFold, a fast deep learning method for antibody structure prediction. IgFold consists of a pre-trained language model trained on 558 million natural antibody sequences followed by graph networks that directly predict backbone atom coordinates. IgFold predicts structures of similar or better quality than alternative methods (including AlphaFold) in significantly less time (under 25 s). Accurate structure prediction on this timescale makes possible avenues of investigation that were previously infeasible. As a demonstration of IgFold's capabilities, we predicted structures for 1.4 million paired antibody sequences, providing structural insights to 500-fold more antibodies than have experimentally determined structures.

Automated optimisation of solubility and conformational stability of antibodies and proteins

Biologics, such as antibodies and enzymes, are crucial in research, biotechnology, diagnostics, and therapeutics. Often, biologics with suitable functionality are discovered, but their development is impeded by developability issues. Stability and solubility are key biophysical traits underpinning developability potential, as they determine aggregation, correlate with production yield and poly-specificity, and are essential to access parenteral and oral delivery. While advances for the optimisation of individual traits have been made, the co-optimization of multiple traits remains highly problematic and time-consuming, as mutations that improve one property often negatively impact others. In this work, we introduce a fully automated computational strategy for the simultaneous optimisation of conformational stability and solubility, which we experimentally validate on six antibodies, including two approved therapeutics. Our results on 42 designs demonstrate that the computational procedure is highly effective at improving developability potential, while not affecting antigen-binding. We make the method available as a webserver at www-cohsoftware.ch.cam.ac.uk.

IgG1 and IgG4 antibodies sample initial structure dependent local conformational states and exhibit non-identical Fab dynamics

We have investigated the dynamics of two [Formula: see text]-immunoglobulin molecules, IgG1 and IgG4, using long all atom molecular dynamics simulations. We first show that the de novo structures of IgG1 and IgG4 predicted using AlphaFold, with no interactions between the fragment crystallizable (Fc) domain and the antigen fragment binding domain (Fab), eventually relaxes to a state with persistent Fc-Fab interactions that mirrors experimentally resolved structures. We quantified the conformational space sampled by antibody trajectories spawned from six different initial structures and show that the individual trajectories only sample states bound by a local minimum and display very little mixing in their conformational states. Furthermore, the dynamics of the individual Fab domains are strongly dependent on the initial crystal structure and isotype. In all conditions, we observe non-identical dynamics between the Fab arms in an antibody. For a six-bead coarse grained model, we show that non-covalent Fc-Fab interactions can modulate the stiffnesses associated with Fc-Fab distances, angles, and dihedral angles by up to three orders of magnitude. Our results clearly illustrate the inherent complexities in studying antibody dynamics and highlight the need to include non-identical Fab dynamics as an inherent feature in computational models of therapeutic antibodies.

Structure-Based Neural Network Protein-Carbohydrate Interaction Predictions at the Residue Level

Carbohydrates dynamically and transiently interact with proteins for cell-cell recognition, cellular differentiation, immune response, and many other cellular processes. Despite the molecular importance of these interactions, there are currently few reliable computational tools to predict potential carbohydrate binding sites on any given protein. Here, we present two deep learning models named CArbohydrate-Protein interaction Site IdentiFier (CAPSIF) that predict carbohydrate binding sites on proteins: (1) a 3D-UNet voxel-based neural network model (CAPSIF:V) and (2) an equivariant graph neural network model (CAPSIF:G). While both models outperform previous surrogate methods used for carbohydrate binding site prediction, CAPSIF:V performs better than CAPSIF:G, achieving test Dice scores of 0.597 and 0.543 and test set Matthews correlation coefficients (MCCs) of 0.599 and 0.538, respectively. We further tested CAPSIF:V on AlphaFold2-predicted protein structures. CAPSIF:V performed equivalently on both experimentally determined structures and AlphaFold2 predicted structures. Finally, we demonstrate how CAPSIF models can be used in conjunction with local glycan-docking protocols, such as GlycanDock, to predict bound protein-carbohydrate structures.

Efficient isolation of rare B cells using next-generation antigen barcoding

The ability to efficiently isolate antigen-specific B cells in high throughput will greatly accelerate the discovery of therapeutic monoclonal antibodies (mAbs) and catalyze rational vaccine development. Traditional mAb discovery is a costly and labor-intensive process, although recent advances in single-cell genomics using emulsion microfluidics allow simultaneous processing of thousands of individual cells. Here we present a streamlined method for isolation and analysis of large numbers of antigen-specific B cells, including next generation antigen barcoding and an integrated computational framework for B cell multi-omics. We demonstrate the power of this approach by recovering thousands of antigen-specific mAbs, including the efficient isolation of extremely rare precursors of VRC01-class and IOMA-class broadly neutralizing HIV mAbs.

Artificial intelligence in cancer immunotherapy: Applications in neoantigen recognition, antibody design and immunotherapy response prediction

Cancer immunotherapy is a method of controlling and eliminating tumors by reactivating the body's cancer-immunity cycle and restoring its antitumor immune response. The increased availability of data, combined with advancements in high-performance computing and innovative artificial intelligence (AI) technology, has resulted in a rise in the use of AI in oncology research. State-of-the-art AI models for functional classification and prediction in immunotherapy research are increasingly used to support laboratory-based experiments. This review offers a glimpse of the current AI applications in immunotherapy, including neoantigen recognition, antibody design, and prediction of immunotherapy response. Advancing in this direction will result in more robust predictive models for developing better targets, drugs, and treatments, and these advancements will eventually make their way into the clinical setting, pushing AI forward in the field of precision oncology.

Comparative Binding Ability of Human Monoclonal Antibodies against Omicron Variants of SARS-CoV-2: An In Silico Investigation

Mutation(s) in the spike protein is the major characteristic trait of newly emerged SARS-CoV-2 variants such as Alpha (B.1.1.7), Beta (B.1.351), Gamma (P.1), Delta (B.1.617.2), and Delta-plus. Omicron (B.1.1.529) is the latest addition and it has been characterized by high transmissibility and the ability to escape host immunity. Recently developed vaccines and repurposed drugs exert limited action on Omicron strains and hence new therapeutics are immediately needed. Herein, we have explored the efficiency of twelve therapeutic monoclonal antibodies (mAbs) targeting the RBD region of the spike glycoprotein against all the Omicron variants bearing a mutation in spike protein through molecular docking and molecular dynamics simulation. Our in silico evidence reveals that adintivimab, beludivimab, and regadanivimab are the most potent mAbs to form strong biophysical interactions and neutralize most of the Omicron variants. Considering the efficacy of mAbs, we incorporated CDRH3 of beludavimab within the framework of adintrevimab, which displayed a more intense binding affinity towards all of the Omicron variants viz. BA.1, BA.2, BA.2.12.1, BA.4, and BA.5. Furthermore, the cDNA of chimeric mAb was cloned in silico within pET30ax for recombinant production. In conclusion, the present study represents the candidature of human mAbs (beludavimab and adintrevimab) and the therapeutic potential of designed chimeric mAb for treating Omicron-infected patients.

Quality assessment of VHH models

Heavy Chain Only Antibodies are specific to Camelid species. Despite the lack of the light chain variable domain, their heavy chain variable domain (VH) domain, named VHH or nanobody, has promising potential applications in research and therapeutic fields. The structural study of VHH is therefore of great interest. Unfortunately, considering the huge amount of sequences that might be produced, only about one thousand of VHH experimental structures are publicly available in the Protein Data Bank, implying that structural model prediction of VHH is a necessary alternative to obtaining 3D information besides its sequence. The present study aims to assess and compare the quality of predictions from different modelling methodologies. Established comparative & homology modelling approaches to recent Deep Learning-based modelling strategies were applied, i.e. Modeller using single or multiple structural templates, ModWeb, SwissModel (with two evaluation schema), RoseTTAfold, AlphaFold 2 and NanoNet. The prediction accuracy was evaluated using RMSD, TM-score, GDT-TS, GDT-HA and Protein Blocks distance metrics. Besides the global structure assessment, we performed specific analyses of Frameworks and CDRs structures. We observed that AlphaFold 2 and especially NanoNet performed better than the other evaluated softwares. Importantly, we performed molecular dynamics simulations of an experimental structure and a NanoNet predicted model of a VHH in order to compare the global structural flexibility and local conformations using Protein Blocks. Despite rather similar structures, substantial differences in dynamical properties were observed, which underlies the complexity of the task of model evaluation.Communicated by Ramaswamy H. Sarma.

De novo design and Rosetta-based assessment of high-affinity antibody variable regions (Fv) against the SARS-CoV-2 spike receptor binding domain (RBD)

The continued emergence of new SARS-CoV-2 variants has accentuated the growing need for fast and reliable methods for the design of potentially neutralizing antibodies (Abs) to counter immune evasion by the virus. Here, we report on the de novo computational design of high-affinity Ab variable regions (Fv) through the recombination of VDJ genes targeting the most solvent-exposed hACE2-binding residues of the SARS-CoV-2 spike receptor binding domain (RBD) protein using the software tool OptMAVEn-2.0. Subsequently, we carried out computational affinity maturation of the designed variable regions through amino acid substitutions for improved binding with the target epitope. Immunogenicity of designs was restricted by preferring designs that match sequences from a 9-mer library of "human Abs" based on a human string content score. We generated 106 different antibody designs and reported in detail on the top five that trade-off the greatest computational binding affinity for the RBD with human string content scores. We further describe computational evaluation of the top five designs produced by OptMAVEn-2.0 using a Rosetta-based approach. We used Rosetta SnugDock for local docking of the designs to evaluate their potential to bind the spike RBD and performed "forward folding" with DeepAb to assess their potential to fold into the designed structures. Ultimately, our results identified one designed Ab variable region, P1.D1, as a particularly promising candidate for experimental testing. This effort puts forth a computational workflow for the de novo design and evaluation of Abs that can quickly be adapted to target spike epitopes of emerging SARS-CoV-2 variants or other antigenic targets.

The RESP AI model accelerates the identification of tight-binding antibodies

High-affinity antibodies are often identified through directed evolution, which may require many iterations of mutagenesis and selection to find an optimal candidate. Deep learning techniques hold the potential to accelerate this process but the existing methods cannot provide the confidence interval or uncertainty needed to assess the reliability of the predictions. Here we present a pipeline called RESP for efficient identification of high affinity antibodies. We develop a learned representation trained on over 3 million human B-cell receptor sequences to encode antibody sequences. We then develop a variational Bayesian neural network to perform ordinal regression on a set of the directed evolution sequences binned by off-rate and quantify their likelihood to be tight binders against an antigen. Importantly, this model can assess sequences not present in the directed evolution library and thus greatly expand the search space to uncover the best sequences for experimental evaluation. We demonstrate the power of this pipeline by achieving a 17-fold improvement in the KD of the PD-L1 antibody Atezolizumab and this success illustrates the potential of RESP in facilitating general antibody development.

Challenges in antibody structure prediction

Advances in structural biology and the exponential increase in the amount of high-quality experimental structural data available in the Protein Data Bank has motivated numerous studies to tackle the grand challenge of predicting protein structures. In 2020 AlphaFold2 revolutionized the field using a combination of artificial intelligence and the evolutionary information contained in multiple sequence alignments. Antibodies are one of the most important classes of biotherapeutic proteins. Accurate structure models are a prerequisite to advance biophysical property predictions and consequently antibody design. Specialized tools used to predict antibody structures based on different principles have profited from current advances in protein structure prediction based on artificial intelligence. Here, we emphasize the importance of reliable protein structure models and highlight the enormous advances in the field, but we also aim to increase awareness that protein structure models, and in particular antibody models, may suffer from structural inaccuracies, namely incorrect cis-amide bonds, wrong stereochemistry or clashes. We show that these inaccuracies affect biophysical property predictions such as surface hydrophobicity. Thus, we stress the importance of carefully reviewing protein structure models before investing further computing power and setting up experiments. To facilitate the assessment of model quality, we provide a tool "TopModel" to validate structure models.

Approaches to expand the conventional toolbox for discovery and selection of antibodies with drug-like physicochemical properties

Antibody drugs should exhibit not only high-binding affinity for their target antigens but also favorable physicochemical drug-like properties. Such drug-like biophysical properties are essential for the successful development of antibody drug products. The traditional approaches used in antibody drug development require significant experimentation to produce, optimize, and characterize many candidates. Therefore, it is attractive to integrate new methods that can optimize the process of selecting antibodies with both desired target-binding and drug-like biophysical properties. Here, we summarize a selection of techniques that can complement the conventional toolbox used to de-risk antibody drug development. These techniques can be integrated at different stages of the antibody development process to reduce the frequency of physicochemical liabilities in antibody libraries during initial discovery and to co-optimize multiple antibody features during early-stage antibody engineering and affinity maturation. Moreover, we highlight biophysical and computational approaches that can be used to predict physical degradation pathways relevant for long-term storage and in-use stability to reduce the need for extensive experimentation.

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.

Learned features of antibody-antigen binding affinity

Defining predictors of antigen-binding affinity of antibodies is valuable for engineering therapeutic antibodies with high binding affinity to their targets. However, this task is challenging owing to the huge diversity in the conformations of the complementarity determining regions of antibodies and the mode of engagement between antibody and antigen. In this study, we used the structural antibody database (SAbDab) to identify features that can discriminate high- and low-binding affinity across a 5-log scale. First, we abstracted features based on previously learned representations of protein-protein interactions to derive 'complex' feature sets, which include energetic, statistical, network-based, and machine-learned features. Second, we contrasted these complex feature sets with additional 'simple' feature sets based on counts of contacts between antibody and antigen. By investigating the predictive potential of 700 features contained in the eight complex and simple feature sets, we observed that simple feature sets perform comparably to complex feature sets in classification of binding affinity. Moreover, combining features from all eight feature-sets provided the best classification performance (median cross-validation AUROC and F1-score of 0.72). Of note, classification performance is substantially improved when several sources of data leakage (e.g., homologous antibodies) are not removed from the dataset, emphasizing a potential pitfall in this task. We additionally observe a classification performance plateau across diverse featurization approaches, highlighting the need for additional affinity-labeled antibody-antigen structural data. The findings from our present study set the stage for future studies aimed at multiple-log enhancement of antibody affinity through feature-guided engineering.

Homology Modeling of Antibody Variable Regions: Methods and Applications

Adaptive immunity specifically protects us from antigenic challenges. Antibodies are key effector proteins of adaptive immunity, and they are remarkable in their ability to recognize a virtually limitless number of antigens. Fragment variable (F_V), the antigen-binding region of antibodies, can be split into two main components, namely, framework and complementarity determining regions. The framework (FR) consists of light-chain framework (FR_L) and heavy-chain framework (FR_H). Similarly, the complementarity determining regions (CDRs) comprises of light-chain CDRs 1-3 (CDRs L1-3) and heavy-chain CDRs 1-3 (CDRs H1-3). While FRs are relatively constant in sequence and structure across diverse antibodies, sequence variation in CDRs leading to differential conformations of CDR loops accounts for the distinct antigenic specificities of diverse antibodies. The conserved structural features in FRs and conformity of CDRs to a limited set of standard conformations allow for the accurate prediction of F_V models using homology modeling techniques. Antibody structure prediction from its amino acid sequence has numerous important applications including prediction of antibody-antigen interaction interfaces and redesign of therapeutically and biotechnologically useful antibodies with improved affinity. This chapter summarizes the current practices employed in the successful homology modeling of antibody variable regions and the potential applications of the generated homology models.

Toward generalizable prediction of antibody thermostability using machine learning on sequence and structure features

Over the last three decades, the appeal for monoclonal antibodies (mAbs) as therapeutics has been steadily increasing as evident with FDA's recent landmark approval of the 100th mAb. Unlike mAbs that bind to single targets, multispecific biologics (msAbs) have garnered particular interest owing to the advantage of engaging distinct targets. One important modular component of msAbs is the single-chain variable fragment (scFv). Despite the exquisite specificity and affinity of these scFv modules, their relatively poor thermostability often hampers their development as a potential therapeutic drug. In recent years, engineering antibody sequences to enhance their stability by mutations has gained considerable momentum. As experimental methods for antibody engineering are time-intensive, laborious and expensive, computational methods serve as a fast and inexpensive alternative to conventional routes. In this work, we show two machine learning approaches - one with pre-trained language models (PTLM) capturing functional effects of sequence variation, and second, a supervised convolutional neural network (CNN) trained with Rosetta energetic features - to better classify thermostable scFv variants from sequence. Both of these models are trained over temperature-specific data (TS50 measurements) derived from multiple libraries of scFv sequences. On out-of-distribution (refers to the fact that the out-of-distribution sequnes are blind to the algorithm) sequences, we show that a sufficiently simple CNN model performs better than general pre-trained language models trained on diverse protein sequences (average Spearman correlation coefficient, ρ , of 0.4 as opposed to 0.15). On the other hand, an antibody-specific language model performs comparatively better than the CNN model on the same task (ρ = 0.52). Further, we demonstrate that for an independent mAb with available thermal melting temperatures for 20 experimentally characterized thermostable mutations, these models trained on TS50 data could identify 18 residue positions and 5 identical amino-acid mutations showing remarkable generalizability. Our results suggest that such models can be broadly applicable for improving the biological characteristics of antibodies. Further, transferring such models for alternative physicochemical properties of scFvs can have potential applications in optimizing large-scale production and delivery of mAbs or bsAbs.

Multi-state modeling of antibody-antigen complexes with SAXS profiles and deep-learning models

Antibodies are an established class of human therapeutics. Epitope characterization is an important part of therapeutic antibody discovery. However, structural characterization of antibody-antigen complexes remains challenging. On the one hand, X-ray crystallography or cryo-electron microscopy provide atomic resolution characterization of the epitope, but the data collection process is typically long and the success rate is low. On the other hand, computational methods for modeling antibody-antigen structures from the individual components frequently suffer from a high false positive rate, rarely resulting in a unique solution. Recent deep learning models for structure prediction are also successful in predicting protein-protein complexes. However, they do not perform well for antibody-antigen complexes. Small Angle X-ray Scattering (SAXS) is a reliable technique for rapid structural characterization of protein samples in solution albeit at low resolution. Here, we present an integrative approach for modeling antigen-antibody complexes using the antibody sequence, antigen structure, and experimentally determined SAXS profiles of the antibody, antigen, and the complex. The method models antibody structures using a novel deep-learning approach, NanoNet. The structures of the antibodies and antigens are represented using multiple 3D conformations to account for compositional and conformational heterogeneity of the protein samples that are used to collect the SAXS data. The complexes are predicted by integrating the SAXS profiles with scoring functions for protein-protein interfaces that are based on statistical potentials and antibody-specific deep-learning models. We validated the method via application to four Fab:EGFR and one Fab:PCSK9 antibody:antigen complexes with experimentally available SAXS datasets. The integrative approach returns accurate predictions (interface RMSD<4Å) in the top five predictions for four out of five complexes (respective interface RMSD values of 1.95, 2.18, 2.66 and 3.87Å), providing support for the utility of such a computational pipeline for epitope characterization during therapeutic antibody discovery.

Unconstrained generation of synthetic antibody-antigen structures to guide machine learning methodology for antibody specificity prediction

Machine learning (ML) is a key technology for accurate prediction of antibody-antigen binding. Two orthogonal problems hinder the application of ML to antibody-specificity prediction and the benchmarking thereof: the lack of a unified ML formalization of immunological antibody-specificity prediction problems and the unavailability of large-scale synthetic datasets to benchmark real-world relevant ML methods and dataset design. Here we developed the Absolut! software suite that enables parameter-based unconstrained generation of synthetic lattice-based three-dimensional antibody-antigen-binding structures with ground-truth access to conformational paratope, epitope and affinity. We formalized common immunological antibody-specificity prediction problems as ML tasks and confirmed that for both sequence- and structure-based tasks, accuracy-based rankings of ML methods trained on experimental data hold for ML methods trained on Absolut!-generated data. The Absolut! framework has the potential to enable real-world relevant development and benchmarking of ML strategies for biotherapeutics design.

Neutralizing antibody creation technologies: case of SARS-CoV-2

Monoclonal antibodies (mAbs) are the most promising and most intensively replenished type of bioactive pharmaceuticals. Currently, there are over 100 different mAbs approved by the FDA and other regulating agencies for treatment of oncological, infectious, systemic, autoimmune and other diseases. Design of antibodies neutralizing pathogens of socially significant infections, such as HIV, hepatitis viruses, SARS-CoV-2, is a separate direction. The SARS-CoV-2 pandemic has shown how urgent it is to have a technological platform enabling production of fully human antibodies. The development of recombinant DNA technology and antibody phage display enabled compilation of libraries of antigen-binding fragments and screening with target antigens. This review discusses the advantages and disadvantages of phage display, including use of single-domain antibody technology based on the heavy chain variable domain. We describe the state-of-the-art (and practical results of its application) technology enabling production of human antibodies by sorting and sequencing the genome of individual memory B cells, using monoclonal virus-neutralizing antibodies against SARS-CoV-2 as an example. The prospects of further development of the recombinant human antibody production technology are discussed; in particular, we consider creation of sequences of variable fragments of antibodies with the help of artificial intelligence.

Development of therapeutic antibodies for the treatment of diseases

Since the first monoclonal antibody drug, muromonab-CD3, was approved for marketing in 1986, 165 antibody drugs have been approved or are under regulatory review worldwide. With the approval of new drugs for treating a wide range of diseases, including cancer and autoimmune and metabolic disorders, the therapeutic antibody drug market has experienced explosive growth. Monoclonal antibodies have been sought after by many biopharmaceutical companies and scientific research institutes due to their high specificity, strong targeting abilities, low toxicity, side effects, and high development success rate. The related industries and markets are growing rapidly, and therapeutic antibodies are one of the most important research and development areas in the field of biology and medicine. In recent years, great progress has been made in the key technologies and theoretical innovations provided by therapeutic antibodies, including antibody-drug conjugates, antibody-conjugated nuclides, bispecific antibodies, nanobodies, and other antibody analogs. Additionally, therapeutic antibodies can be combined with technologies used in other fields to create new cross-fields, such as chimeric antigen receptor T cells (CAR-T), CAR-natural killer cells (CAR-NK), and other cell therapy. This review summarizes the latest approved or in regulatory review therapeutic antibodies that have been approved or that are under regulatory review worldwide, as well as clinical research on these approaches and their development, and outlines antibody discovery strategies that have emerged during the development of therapeutic antibodies, such as hybridoma technology, phage display, preparation of fully human antibody from transgenic mice, single B-cell antibody technology, and artificial intelligence-assisted antibody discovery.

Hallucinating structure-conditioned antibody libraries for target-specific binders

Antibodies are widely developed and used as therapeutics to treat cancer, infectious disease, and inflammation. During development, initial leads routinely undergo additional engineering to increase their target affinity. Experimental methods for affinity maturation are expensive, laborious, and time-consuming and rarely allow the efficient exploration of the relevant design space. Deep learning (DL) models are transforming the field of protein engineering and design. While several DL-based protein design methods have shown promise, the antibody design problem is distinct, and specialized models for antibody design are desirable. Inspired by hallucination frameworks that leverage accurate structure prediction DL models, we propose the FvHallucinator for designing antibody sequences, especially the CDR loops, conditioned on an antibody structure. Such a strategy generates targeted CDR libraries that retain the conformation of the binder and thereby the mode of binding to the epitope on the antigen. On a benchmark set of 60 antibodies, FvHallucinator generates sequences resembling natural CDRs and recapitulates perplexity of canonical CDR clusters. Furthermore, the FvHallucinator designs amino acid substitutions at the VH-VL interface that are enriched in human antibody repertoires and therapeutic antibodies. We propose a pipeline that screens FvHallucinator designs to obtain a library enriched in binders for an antigen of interest. We apply this pipeline to the CDR H3 of the Trastuzumab-HER2 complex to generate in silico designs predicted to improve upon the binding affinity and interfacial properties of the original antibody. Thus, the FvHallucinator pipeline enables generation of inexpensive, diverse, and targeted antibody libraries enriched in binders for antibody affinity maturation.

Advances in antibody discovery from human BCR repertoires

Antibodies make up an important and growing class of compounds used for the diagnosis or treatment of disease. While traditional antibody discovery utilized immunization of animals to generate lead compounds, technological innovations have made it possible to search for antibodies targeting a given antigen within the repertoires of B cells in humans. Here we group these innovations into four broad categories: cell sorting allows the collection of cells enriched in specificity to one or more antigens; BCR sequencing can be performed on bulk mRNA, genomic DNA or on paired (heavy-light) mRNA; BCR repertoire analysis generally involves clustering BCRs into specificity groups or more in-depth modeling of antibody-antigen interactions, such as antibody-specific epitope predictions; validation of antibody-antigen interactions requires expression of antibodies, followed by antigen binding assays or epitope mapping. Together with innovations in Deep learning these technologies will contribute to the future discovery of diagnostic and therapeutic antibodies directly from humans.

Improved Antibody-Specific Epitope Prediction Using AlphaFold and AbAdapt

Antibodies recognize their cognate antigens with high affinity and specificity, but the prediction of binding sites on the antigen (epitope) corresponding to a specific antibody remains a challenging problem. To address this problem, we developed AbAdapt, a pipeline that integrates antibody and antigen structural modeling with rigid docking in order to derive antibody-antigen specific features for epitope prediction. In this study, we systematically assessed the impact of integrating the state-of-the-art protein modeling method AlphaFold with the AbAdapt pipeline. By incorporating more accurate antibody models, we observed improvement in docking, paratope prediction, and prediction of antibody-specific epitopes. We further applied AbAdapt-AF in an anti-receptor binding domain (RBD) antibody complex benchmark and found AbAdapt-AF outperformed three alternative docking methods. Also, AbAdapt-AF demonstrated higher epitope prediction accuracy than other tested epitope prediction tools in the anti-RBD antibody complex benchmark. We anticipate that AbAdapt-AF will facilitate prediction of antigen-antibody interactions in a wide range of applications.

Local Normal Mode Analysis for Fast Loop Conformational Sampling

We propose and validate a novel method to efficiently explore local protein loop conformations based on a new formalism for constrained normal mode analysis (NMA) in internal coordinates. The manifold of possible loop configurations imposed by the position and orientation of the fixed loop ends is reduced to an orthogonal set of motions (or modes) encoding concerted rotations of all the backbone dihedral angles. We validate the sampling power on a set of protein loops with highly variable experimental structures and demonstrate that our approach can efficiently explore the conformational space of closed loops. We also show an acceptable resemblance of the ensembles around equilibrium conformations generated by long molecular simulations and constrained NMA on a set of exposed and diverse loops. In comparison with other methods, the main advantage is the lack of restrictions on the number of dihedrals that can be altered simultaneously. Furthermore, the method is computationally efficient since it only requires the diagonalization of a tiny matrix, and the modes of motions are energetically contextualized by the elastic network model, which includes both the loop and the neighboring residues.

Antibody variable region engineering for improving cancer immunotherapy

The efficacy and specificity of conventional monoclonal antibody (mAb) drugs in the clinic require further improvement. Currently, the development and application of novel antibody formats for improving cancer immunotherapy have attracted much attention. Variable region-retaining antibody fragments, such as antigen-binding fragment (Fab), single-chain variable fragment (scFv), bispecific antibody, and bi/trispecific cell engagers, are engineered with humanization, multivalent antibody construction, affinity optimization and antibody masking for targeting tumor cells and killer cells to improve antibody-based therapy potency, efficacy and specificity. In this review, we summarize the application of antibody variable region engineering and discuss the future direction of antibody engineering for improving cancer therapies.

NanoNet: Rapid and accurate end-to-end nanobody modeling by deep learning

Antibodies are a rapidly growing class of therapeutics. Recently, single domain camelid VHH antibodies, and their recognition nanobody domain (Nb) appeared as a cost-effective highly stable alternative to full-length antibodies. There is a growing need for high-throughput epitope mapping based on accurate structural modeling of the variable domains that share a common fold and differ in the Complementarity Determining Regions (CDRs). We develop a deep learning end-to-end model, NanoNet, that given a sequence directly produces the 3D coordinates of the backbone and Cβ atoms of the entire VH domain. For the Nb test set, NanoNet achieves 3.16Å average RMSD for the most variable CDR3 loops and 2.65Å, 1.73Å for the CDR1, CDR2 loops, respectively. The accuracy for antibody VH domains is even higher: 2.38Å RMSD for CDR3 and 0.89Å, 0.96Å for the CDR1, CDR2 loops, respectively. NanoNet run times allow generation of ∼1M nanobody structures in less than 4 hours on a standard CPU computer enabling high-throughput structure modeling. NanoNet is available at GitHub: https://github.com/dina-lab3D/NanoNet.

Viral informatics: bioinformatics-based solution for managing viral infections

Several new viral infections have emerged in the human population and establishing as global pandemics. With advancements in translation research, the scientific community has developed potential therapeutics to eradicate or control certain viral infections, such as smallpox and polio, responsible for billions of disabilities and deaths in the past. Unfortunately, some viral infections, such as dengue virus (DENV) and human immunodeficiency virus-1 (HIV-1), are still prevailing due to a lack of specific therapeutics, while new pathogenic viral strains or variants are emerging because of high genetic recombination or cross-species transmission. Consequently, to combat the emerging viral infections, bioinformatics-based potential strategies have been developed for viral characterization and developing new effective therapeutics for their eradication or management. This review attempts to provide a single platform for the available wide range of bioinformatics-based approaches, including bioinformatics methods for the identification and management of emerging or evolved viral strains, genome analysis concerning the pathogenicity and epidemiological analysis, computational methods for designing the viral therapeutics, and consolidated information in the form of databases against the known pathogenic viruses. This enriched review of the generally applicable viral informatics approaches aims to provide an overview of available resources capable of carrying out the desired task and may be utilized to expand additional strategies to improve the quality of translation viral informatics research.

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.

Machine-designed biotherapeutics: opportunities, feasibility and advantages of deep learning in computational antibody discovery

Antibodies are versatile molecular binders with an established and growing role as therapeutics. Computational approaches to developing and designing these molecules are being increasingly used to complement traditional lab-based processes. Nowadays, in silico methods fill multiple elements of the discovery stage, such as characterizing antibody-antigen interactions and identifying developability liabilities. Recently, computational methods tackling such problems have begun to follow machine learning paradigms, in many cases deep learning specifically. This paradigm shift offers improvements in established areas such as structure or binding prediction and opens up new possibilities such as language-based modeling of antibody repertoires or machine-learning-based generation of novel sequences. In this review, we critically examine the recent developments in (deep) machine learning approaches to therapeutic antibody design with implications for fully computational antibody design.

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.

Simultaneous prediction of antibody backbone and side-chain conformations with deep learning

Antibody engineering is becoming increasingly popular in medicine for the development of diagnostics and immunotherapies. Antibody function relies largely on the recognition and binding of antigenic epitopes via the loops in the complementarity determining regions. Hence, accurate high-resolution modeling of these loops is essential for effective antibody engineering and design. Deep learning methods have previously been shown to effectively predict antibody backbone structures described as a set of inter-residue distances and orientations. However, antigen binding is also dependent on the specific conformations of surface side-chains. To address this shortcoming, we created DeepSCAb: a deep learning method that predicts inter-residue geometries as well as side-chain dihedrals of the antibody variable fragment. The network requires only sequence as input, rendering it particularly useful for antibodies without any known backbone conformations. Rotamer predictions use an interpretable self-attention layer, which learns to identify structurally conserved anchor positions across several species. We evaluate the performance of the model for discriminating near-native structures from sets of decoys and find that DeepSCAb outperforms similar methods lacking side-chain context. When compared to alternative rotamer repacking methods, which require an input backbone structure, DeepSCAb predicts side-chain conformations competitively. Our findings suggest that DeepSCAb improves antibody structure prediction with accurate side-chain modeling and is adaptable to applications in docking of antibody-antigen complexes and design of new therapeutic antibody sequences.

Ig-VAE: Generative modeling of protein structure by direct 3D coordinate generation

While deep learning models have seen increasing applications in protein science, few have been implemented for protein backbone generation—an important task in structure-based problems such as active site and interface design. We present a new approach to building class-specific backbones, using a variational auto-encoder to directly generate the 3D coordinates of immunoglobulins. Our model is torsion- and distance-aware, learns a high-resolution embedding of the dataset, and generates novel, high-quality structures compatible with existing design tools. We show that the Ig-VAE can be used with Rosetta to create a computational model of a SARS-CoV2-RBD binder via latent space sampling. We further demonstrate that the model’s generative prior is a powerful tool for guiding computational protein design, motivating a new paradigm under which backbone design is solved as constrained optimization problem in the latent space of a generative model.

Many essential biochemical processes are governed by protein-protein interactions (PPIs), and our ability to make binding proteins that modulate PPIs is crucial to the creation of therapeutics and the study of cell-signaling. One critical aspect of PPI design is to capture protein conformational flexibility. Deep generative models are a class of mathematical models that are able to synthesize novel data from a finite set of training examples. Here, we make advances in computational protein design methodology by developing a deep generative model that creates protein backbones adopting the immunoglobulin fold, which is found in natural binding proteins such as antibodies. While generative models have been powerful in tasks such as image generation, using them to create proteins has remained a challenge. We solve this problem with a new model that allows for the direct generation of novel 3D molecules and show that they are of high chemical accuracy. Generated structures work well with existing protein design methods such as Rosetta, providing access to a large collection of novel immunoglobulin structures. Finally, we present a new protein design framework, called “generative design,” that shows how deep generative models such as ours can be applied to virtually any protein design problem.

Advances in computational structure-based antibody design

Antibodies are currently the most important class of biotherapeutics and are used to treat numerous diseases. Recent advances in computational methods are ushering in a new era of antibody design, driven in part by accurate structure prediction. Previously, structure-based antibody design has been limited to a relatively small number of cases where accurate structures or models of both the target antigen and antibody were available. As we move towards a time where it is possible to accurately model most antibodies and antigens, and to reliably predict their binding site, there is vast potential for true computational antibody design. In this review, we describe the latest methods that promise to launch a paradigm shift towards entirely in silico structure-based antibody design.