Co-optimization of therapeutic antibody affinity and specificity using machine learning models that generalize to novel mutational space

Recent advances in antibody optimization based on deep learning methods

Antibodies currently comprise the predominant treatment modality for a variety of diseases; therefore, optimizing their properties rapidly and efficiently is an indispensable step in antibody-based drug development. Inspired by the great success of artificial intelligence-based algorithms, especially deep learning-based methods in the field of biology, various computational methods have been introduced into antibody optimization to reduce costs and increase the success rate of lead candidate generation and optimization. Herein, we briefly review recent progress in deep learning-based antibody optimization, focusing on the available datasets and algorithm input data types that are crucial for constructing appropriate deep learning models. Furthermore, we discuss the current challenges and potential solutions for the future development of general-purpose deep learning algorithms in antibody optimization.

Accelerating antibody discovery and optimization with high-throughput experimentation and machine learning

The integration of high-throughput experimentation and machine learning is transforming data-driven antibody engineering, revolutionizing the discovery and optimization of antibody therapeutics. These approaches employ extensive datasets comprising antibody sequences, structures, and functional properties to train predictive models that enable rational design. This review highlights the significant advancements in data acquisition and feature extraction, emphasizing the necessity of capturing both sequence and structural information. We illustrate how machine learning models, including protein language models, are used not only to enhance affinity but also to optimize other crucial therapeutic properties, such as specificity, stability, viscosity, and manufacturability. Furthermore, we provide practical examples and case studies to demonstrate how the synergy between experimental and computational approaches accelerates antibody engineering. Finally, this review discusses the remaining challenges in fully realizing the potential of artificial intelligence (AI)-powered antibody discovery pipelines to expedite therapeutic development.

Optimization of synthetic human VH affinity and solubility through in vitro affinity maturation and minimal camelization

An attractive feature of human VHs over camelid VHHs as immunotherapeutics is their perceived lower risk of immunogenicity. While human VHs can readily be obtained from synthetic phage display libraries, they often suffer from low affinity and poor solubility compared to VHHs derived from immune libraries. Using SARS-CoV-2 spike protein as a model antigen, we screened a synthetic human VH phage display library and identified a diverse set of antigen-specific VHs. However, the VHs exhibited low affinity, and many had low solubility; that is, they were prone to aggregation. To explore the feasibility of improving the affinity, we subjected a representative VH to in vitro affinity maturation. We created a yeast surface display library of VH variants employing a site-saturated mutagenesis approach targeting complementarity-determining regions and selected against the target antigen. Next-generation sequencing of the selected variants, combined with structural modeling, identified a set of VHs as potentially improved candidates. Characterization of these candidates revealed several VHs with improved affinities of up to 100-fold (KDs as low as 3 nM) and potent neutralization capabilities; however, they still showed significant aggregation. By introducing as few as two camelid residues into the framework region 2 of a high-affinity VH (a process referred to as camelization), we were able to completely solubilize the VH without compromising its affinity and other important attributes, including thermostability and protein A binding. This study demonstrates the feasibility of generating high-affinity, -solubility, and -stability human VHs from synthetic libraries through a combination of in vitro affinity maturation and minimal camelization.

Prediction of Self-Association and Solution Behavior of Monoclonal Antibodies Using the QCM-D Metric of Loosely Interacting Layer

Despite the increasing availability and success of monoclonal antibodies (mAb), early identification of candidate molecules with desirable developability attributes remains challenging due to self-association and poor solution behavior. Measuring these phenomena experimentally using the available methods is complicated in mAbs development. Quartz crystal microbalance with dissipation monitoring (QCM-D) detects a loosely interacting layer on top of the irreversibly adsorbed layer of molecules, providing information about the mAbs interaction. This work aimed to explore whether the characteristics of this layer can be used as a reliable self-association metric. QCM-D experiments showed a large frequency shift (Δf) associated with loosely interacting layers for omalizumab but a small or absent layer for tocilizumab. Accordingly, the viscosity of omalizumab increased exponentially at high concentrations compared to tocilizumab. Testing eight mAbs with different self-association behaviors revealed a strong rank order correlation between the mostly used metric of self-association, i.e., diffusion interaction parameter (kD-DLS), and Δf, indicating Δf's potential for predicting mAb solution behavior. The study also highlighted the robustness of the metric to impurities and temperature variations compared to the sensitive kD-DLS. Overall, we demonstrate that the loosely interacting layer provides valuable information about mAb self-association, predicting the colloidal stability and solution behavior in therapeutic development.

Revolutionizing oncology: the role of Artificial Intelligence (AI) as an antibody design, and optimization tools

Antibodies play a crucial role in defending the human body against diseases, including life-threatening conditions like cancer. They mediate immune responses against foreign antigens and, in some cases, self-antigens. Over time, antibody-based technologies have evolved from monoclonal antibodies (mAbs) to chimeric antigen receptor T cells (CAR-T cells), significantly impacting biotechnology, diagnostics, and therapeutics. Although these advancements have enhanced therapeutic interventions, the integration of artificial intelligence (AI) is revolutionizing antibody design and optimization. This review explores recent AI advancements, including large language models (LLMs), diffusion models, and generative AI-based applications, which have transformed antibody discovery by accelerating de novo generation, enhancing immune response precision, and optimizing therapeutic efficacy. Through advanced data analysis, AI enables the prediction and design of antibody sequences, 3D structures, complementarity-determining regions (CDRs), paratopes, epitopes, and antigen-antibody interactions. These AI-powered innovations address longstanding challenges in antibody development, significantly improving speed, specificity, and accuracy in therapeutic design. By integrating computational advancements with biomedical applications, AI is driving next-generation cancer therapies, transforming precision medicine, and enhancing patient outcomes.

AntiFold: improved structure-based antibody design using inverse folding

Summary

The design and optimization of antibodies requires an intricate balance across multiple properties. Protein inverse folding models, capable of generating diverse sequences folding into the same structure, are promising tools for maintaining structural integrity during antibody design. Here, we present AntiFold, an antibody-specific inverse folding model, fine-tuned from ESM-IF1 on solved and predicted antibody structures. AntiFold outperforms existing inverse folding tools on sequence recovery across complementarity-determining regions, with designed sequences showing high structural similarity to their solved counterpart. It additionally achieves stronger correlations when predicting antibody-antigen binding affinity in a zero-shot manner. AntiFold assigns low probabilities to mutations that disrupt antigen binding, synergizing with protein language model residue probabilities, and demonstrates promise for guiding antibody optimization while retaining structure-related properties.

Availability and implementation

AntiFold is freely available under the BSD 3-Clause as a web server (https://opig.stats.ox.ac.uk/webapps/antifold/) and pip-installable package (https://github.com/oxpig/AntiFold).

Pareto-optimal sampling for multi-objective protein sequence design

Supervised machine learning (ML) has significantly advanced sequence-based protein property prediction. However, its inverse application, designing protein sequences with desired properties, remains under-explored. The challenges in sequence design stem from the vast search space and the rugged protein fitness landscape. In this work, we present MosPro, an efficient ML algorithm for property-guided protein sequence design. We frame sequence design as a discrete sampling problem. Utilizing a pre-trained differentiable ML model that predicts properties of sequences, MosPro shapes a distribution that assigns high probability mass to regions for high-property sequences. To generate designs, MosPro efficiently samples sequences from this constructed distribution. We further develop a Pareto optimization algorithm to propose sequences that are simultaneously optimized for multiple properties. Evaluations on experimental fitness landscapes demonstrated that MosPro generates sequences that optimally trade off multiple desiderata. Our results suggested an unparalleled potential of generative ML for efficient and controllable design for functional proteins.

Binding Promiscuity of Therapeutic Factor VIII

The binding promiscuity of proteins defines their ability to indiscriminately bind multiple unrelated molecules. Binding promiscuity is implicated, at least in part, in the off-target reactivity, nonspecific biodistribution, immunogenicity, and/or short half-life of potentially efficacious protein drugs, thus affecting their clinical use. In this review, we discuss the current evidence for the binding promiscuity of factor VIII (FVIII), a protein used for the treatment of hemophilia A, which displays poor pharmacokinetics, and elevated immunogenicity. We summarize the different canonical and noncanonical interactions that FVIII may establish in the circulation and that could be responsible for its therapeutic liabilities. We also provide information suggesting that the FVIII light chain, and especially its C1 and C2 domains, could play an important role in the binding promiscuity. We believe that the knowledge accumulated over years of FVIII usage could be exploited for the development of strategies to predict protein binding promiscuity and therefore anticipate drug efficacy and toxicity. This would open a mutational space to reduce the binding promiscuity of emerging protein drugs while conserving their therapeutic potency.

Self-supervised machine learning methods for protein design improve sampling but not the identification of high-fitness variants

Machine learning (ML) is changing the world of computational protein design, with data-driven methods surpassing biophysical-based methods in experimental success. However, they are most often reported as case studies, lack integration and standardization, and are therefore hard to objectively compare. In this study, we established a streamlined and diverse toolbox for methods that predict amino acid probabilities inside the Rosetta software framework that allows for the side-by-side comparison of these models. Subsequently, existing protein fitness landscapes were used to benchmark novel ML methods in realistic protein design settings. We focused on the traditional problems of protein design: sampling and scoring. A major finding of our study is that ML approaches are better at purging the sampling space from deleterious mutations. Nevertheless, scoring resulting mutations without model fine-tuning showed no clear improvement over scoring with Rosetta. We conclude that ML now complements, rather than replaces, biophysical methods in protein design.

AI-assisted protein design to rapidly convert antibody sequences to intrabodies targeting diverse peptides and histone modifications

Intrabodies are engineered antibodies that function inside living cells, enabling therapeutic, diagnostic, and imaging applications. While powerful, their development has been hindered by challenges associated with their folding, solubility, and stability in the reduced intracellular environment. Here, we present an AI-driven pipeline integrating AlphaFold2, ProteinMPNN, and live-cell screening to optimize antibody framework regions while preserving epitope-binding complementarity-determining regions. Using this approach, we successfully converted 19 out of 26 antibody sequences into functional single-chain variable fragment (scFv) intrabodies, including a panel targeting diverse histone modifications for real-time imaging of chromatin dynamics and gene regulation. Notably, 18 of these 19 sequences had failed to convert using the standard approach, demonstrating the unique effectiveness of our method. As antibody sequence databases expand, our method will accelerate intrabody design, making their development easier, more cost-effective, and broadly accessible for biological research.

Adaptation of Aglycosylated Monoclonal Antibodies for Improved Production in Komagataella phaffii

Monoclonal antibodies (mAbs) are a major class of biopharmaceuticals manufactured by well-established processes using Chinese Hamster Ovary (CHO) cells. Next-generation biomanufacturing using alternative hosts like Komagataella phaffii could improve the accessibility of these medicines, address broad societal goals for sustainability, and offer financial advantages for accelerated development of new products. Antibodies produced by K. phaffii, however, may manifest unique molecular quality attributes, like host-dependent, product-related variants, that could raise potential concerns for clinical use. We demonstrate here conservative modifications to the amino acid sequence of aglycosylated antibodies based on the human IgG1 isotype that minimize product-related variations when secreted by K. phaffii. A combination of 2-3 changes of amino acids reduced variations across six different aglycosylated versions of commercial mAbs. Expression of a modified sequence of NIST mAb in both K. phaffii and CHO cells showed comparable biophysical properties and molecular variations. These results suggest a path toward the production of high-quality mAbs that could be expressed interchangeably by either yeast or mammalian cells. Improving molecular designs of proteins to enable a range of manufacturing strategies for well-characterized biopharmaceuticals could accelerate global accessibility and innovations.

Navigating the landscape: A comprehensive overview of computational approaches in therapeutic antibody design and analysis

Immunotherapy, harnessing components like antibodies, cells, and cytokines, has become a cornerstone in treating diseases such as cancer and autoimmune disorders. Therapeutic antibodies, in particular, have transformed modern medicine, providing a targeted approach that destroys disease-causing cells while sparing healthy tissues, thereby reducing the side effects commonly associated with chemotherapy. Beyond oncology, these antibodies also hold promise in addressing chronic infections where conventional therapeutics may fall short. However, antibodies identified through in vivo or in vitro methods often require extensive engineering to enhance their therapeutic potential. This optimization process, aimed at improving affinity, specificity, and reducing immunogenicity, is both challenging and costly, often involving trade-offs between critical properties. Traditional methods of antibody development, such as hybridoma technology and display techniques, are resource-intensive and time-consuming. In contrast, computational approaches offer a faster, more efficient alternative, enabling the precise design and analysis of therapeutic antibodies. These methods include sequence and structural bioinformatics approaches, next-generation sequencing-based data mining, machine learning algorithms, systems biology, immuno-informatics, and integrative approaches. These approaches are advancing the field by providing new insights and enhancing the accuracy of antibody design and analysis. In conclusion, computational approaches are essential in the development of therapeutic antibodies, significantly improving the precision and speed of discovery, optimization, and validation. Integrating these methods with experimental approaches accelerates therapeutic antibody development, paving the way for innovative strategies and treatments for various diseases ranging from cancers to autoimmune and infectious diseases.

Use of Affimer technology for inhibition of α2-antiplasmin and enhancement of fibrinolysis

ABSTRACT

Hypofibrinolysis is a documented abnormality in conditions with high risk of vascular occlusion. A key inhibitor of fibrinolysis is α2-antiplasmin (α2AP), and we hypothesize that the Affimer technology, comprising small conformational proteins with 2 9-amino-acid variable regions, can be used to modulate α2AP activity and facilitate fibrinolysis. Using a phage display system, a library of Affimers was screened against α2AP. A total of 28 α2AP-specific Affimers were isolated, of which 1, termed Affimer A11, inhibited protein function and enhanced fibrinolysis. Affimer A11 displayed a monomeric form and consistently reduced the lysis time of clots made from plasma samples of individuals with type 2 diabetes mellitus (n = 15; from 150.8 ± 100.9 to 109.8 ± 104.8 minutes) and those with cardiovascular disease (n = 15; 117.6 ± 40.6 to 79.7 ± 33.3 minutes; P < .01 for both groups). The effects of A11 on fibrinolysis were maintained when clots were made from whole blood samples. Mechanistic studies demonstrated that A11 did not affect clot structure or interfere with the incorporation of α2AP into fibrin networks but significantly enhanced plasmin activity and accelerated plasmin generation. Affimer A11 reduced α2AP binding to plasmin(ogen), whereas molecular modeling demonstrated interactions with α2AP in an area responsible for binding to plasminogen, explaining the effects on both plasmin activity and generation. Affimer A11, at 0.15 to 0.60 mg/mL, had the ability to bind 70% to 90% of plasma α2AP. In conclusion, we demonstrate that Affimers are viable tools for inhibiting α2AP function and facilitating fibrinolysis, making them potential future therapeutic agents to reduce thrombosis risk.

Reading the repertoire: Progress in adaptive immune receptor analysis using machine learning

The adaptive immune system holds invaluable information on past and present immune responses in the form of B and T cell receptor sequences, but we are limited in our ability to decode this information. Machine learning approaches are under active investigation for a range of tasks relevant to understanding and manipulating the adaptive immune receptor repertoire, including matching receptors to the antigens they bind, generating antibodies or T cell receptors for use as therapeutics, and diagnosing disease based on patient repertoires. Progress on these tasks has the potential to substantially improve the development of vaccines, therapeutics, and diagnostics, as well as advance our understanding of fundamental immunological principles. We outline key challenges for the field, highlighting the need for software benchmarking, targeted large-scale data generation, and coordinated research efforts.

Predicting adaptive immune receptor specificities by machine learning is a data generation problem

Determining the specificity of adaptive immune receptors-B cell receptors (BCRs), their secreted form antibodies, and T cell receptors (TCRs)-is critical for understanding immune responses and advancing immunotherapy and drug discovery. Immune receptors exhibit extensive diversity in their variable domains, enabling them to interact with a plethora of antigens. Despite the significant progress made by AI tools such as AlphaFold in predicting protein structures, challenges remain in accurately modeling the structure and specificity of immune receptors, primarily due to the limited availability of high-quality crystal structures and the complexity of immune receptor-antigen interactions. In this perspective, we highlight recent advancements in sequence-based and structure-based data generation for immune receptors, which are crucial for training machine learning models that predict receptor specificity. We discuss the current bottlenecks and potential future directions in generating and utilizing high-dimensional datasets for predicting and designing the specificity of antibodies and TCRs.

A potent pan-sarbecovirus neutralizing antibody resilient to epitope diversification

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) evolution has resulted in viral escape from clinically authorized monoclonal antibodies (mAbs), creating a need for mAbs that are resilient to epitope diversification. Broadly neutralizing coronavirus mAbs that are sufficiently potent for clinical development and retain activity despite viral evolution remain elusive. We identified a human mAb, designated VIR-7229, which targets the viral receptor-binding motif (RBM) with unprecedented cross-reactivity to all sarbecovirus clades, including non-ACE2-utilizing bat sarbecoviruses, while potently neutralizing SARS-CoV-2 variants since 2019, including the recent EG.5, BA.2.86, and JN.1. VIR-7229 tolerates extraordinary epitope variability, partly attributed to its high binding affinity, receptor molecular mimicry, and interactions with RBM backbone atoms. Consequently, VIR-7229 features a high barrier for selection of escape mutants, which are rare and associated with reduced viral fitness, underscoring its potential to be resilient to future viral evolution. VIR-7229 is a strong candidate to become a next-generation medicine.

Molecular dynamics and machine learning stratify motion-dependent activity profiles of S-layer destabilizing nanobodies

Nanobody (Nb)-induced disassembly of surface array protein (Sap) S-layers, a two-dimensional paracrystalline protein lattice from Bacillus anthracis, has been presented as a therapeutic intervention for lethal anthrax infections. However, only a subset of existing Nbs with affinity to Sap exhibit depolymerization activity, suggesting that affinity and epitope recognition are not enough to explain inhibitory activity. In this study, we performed all-atom molecular dynamics simulations of each Nb bound to the Sap binding site and trained a collection of machine learning classifiers to predict whether each Nb induces depolymerization. We used feature importance analysis to filter out unnecessary features and engineered remaining features to regularize the feature landscape and encourage learning of the depolymerization mechanism. We find that, while not enforced in training, a gradient-boosting decision tree is able to reproduce the experimental activities of inhibitory Nbs while maintaining high classification accuracy, whereas neural networks were only able to discriminate between classes. Further feature analysis revealed that inhibitory Nbs restrain Sap motions toward an inhibitory conformational state described by domain-domain clamping and induced twisting of domains normal to the lattice plane. We believe these motions drive Sap lattice depolymerization and can be used as design targets for improved Sap-inhibitory Nbs. Finally, we expect our method of study to apply to S-layers that serve as virulence factors in other pathogens, paving the way forward for Nb therapeutics that target depolymerization mechanisms.

Bispecific Antibodies, Immune Checkpoint Inhibitors, and Antibody-Drug Conjugates Directing Antitumor Immune Responses: Challenges and Prospects

Tumor immunotherapy includes bispecific antibodies (BsAbs), immune checkpoint inhibitors (ICIs), vaccines, and adoptive cell immunotherapy. BsAbs belong to the family of antibodies that can specifically target two or more different antigens and are a promising option for tumor immunotherapy. Immune checkpoints are antibodies targeting PD-1, PD-L1, and CTLA4 and have demonstrated remarkable therapeutic efficacy in the treatment of hematological and solid tumors, whose combination therapies have been shown to synergistically enhance the antitumor effects of BsAbs. In addition, the clinical efficacy of existing monoclonal antibodies targeting PD-1 (e.g., ipilimumab, nivolumab, pembrolizumab, and cemiplimab) and PD-L1 (e.g., atezolizumab, avelumab, and durvalumab) could also be enhanced by conjugation to small drugs as antibody-drug conjugates (ADCs). The development of truly effective therapies for patients with treatment-resistant cancers can be achieved by optimizing the various components of ADCs.

AntiBinder: utilizing bidirectional attention and hybrid encoding for precise antibody-antigen interaction prediction

Antibodies play a key role in medical diagnostics and therapeutics. Accurately predicting antibody-antigen binding is essential for developing effective treatments. Traditional protein-protein interaction prediction methods often fall short because they do not account for the unique structural and dynamic properties of antibodies and antigens. In this study, we present AntiBinder, a novel predictive model specifically designed to address these challenges. AntiBinder integrates the unique structural and sequence characteristics of antibodies and antigens into its framework and employs a bidirectional cross-attention mechanism to automatically learn the intrinsic mechanisms of antigen-antibody binding, eliminating the need for manual feature engineering. Our comprehensive experiments, which include predicting interactions between known antigens and new antibodies, predicting the binding of previously unseen antigens, and predicting cross-species antigen-antibody interactions, demonstrate that AntiBinder outperforms existing state-of-the-art methods. Notably, AntiBinder excels in predicting interactions with unseen antigens and maintains a reasonable level of predictive capability in challenging cross-species prediction tasks. AntiBinder's ability to model complex antigen-antibody interactions highlights its potential applications in biomedical research and therapeutic development, including the design of vaccines and antibody therapies for rapidly emerging infectious diseases.

Prediction of antibody-antigen interaction based on backbone aware with invariant point attention

BACKGROUND

Antibodies play a crucial role in disease treatment, leveraging their ability to selectively interact with the specific antigen. However, screening antibody gene sequences for target antigens via biological experiments is extremely time-consuming and labor-intensive. Several computational methods have been developed to predict antibody-antigen interaction while suffering from the lack of characterizing the underlying structure of the antibody.

RESULTS

Beneficial from the recent breakthroughs in deep learning for antibody structure prediction, we propose a novel neural network architecture to predict antibody-antigen interaction. We first introduce AbAgIPA: an antibody structure prediction network to obtain the antibody backbone structure, where the structural features of antibodies and antigens are encoded into representation vectors according to the amino acid physicochemical features and Invariant Point Attention (IPA) computation methods. Finally, the antibody-antigen interaction is predicted by global max pooling, feature concatenation, and a fully connected layer. We evaluated our method on antigen diversity and antigen-specific antibody-antigen interaction datasets. Additionally, our model exhibits a commendable level of interpretability, essential for understanding underlying interaction mechanisms.

CONCLUSIONS

Quantitative experimental results demonstrate that the new neural network architecture significantly outperforms the best sequence-based methods as well as the methods based on residue contact maps and graph convolution networks (GCNs). The source code is freely available on GitHub at https://github.com/gmthu66/AbAgIPA .

Recent Advances in Artificial Intelligence to Improve Immunotherapy and the Use of Digital Twins to Identify Prognosis of Patients with Solid Tumors

To date, the public health system has been impacted by the increasing costs of many diagnostic and therapeutic pathways due to limited resources. At the same time, we are constantly seeking to improve these paths through approaches aimed at personalized medicine. To achieve the required levels of diagnostic and therapeutic precision, it is necessary to integrate data from different sources and simulation platforms. Today, artificial intelligence (AI), machine learning (ML), and predictive computer models are more efficient at guiding decisions regarding better therapies and medical procedures. The evolution of these multiparametric and multimodal systems has led to the creation of digital twins (DTs). The goal of our review is to summarize AI applications in discovering new immunotherapies and developing predictive models for more precise immunotherapeutic decision-making. The findings from this literature review highlight that DTs, particularly predictive mathematical models, will be pivotal in advancing healthcare outcomes. Over time, DTs will indeed bring the benefits of diagnostic precision and personalized treatment to a broader spectrum of patients.

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.

Human antibody polyreactivity is governed primarily by the heavy-chain complementarity-determining regions

Although antibody variable regions mediate antigen-specific binding, they can also mediate non-specific interactions with non-cognate antigens, impacting diverse immunological processes and the efficacy, safety, and half-life of antibody therapeutics. To understand the molecular basis of antibody non-specificity, we sorted two dissimilar human naïve antibody libraries against multiple reagents to enrich for variants with different levels of polyreactivity. Sequence analysis of >300,000 paired antibody variable regions revealed that the heavy chain primarily mediates human antibody polyreactivity, and this is due to the high positive charge, high hydrophobicity, and combinations thereof in the corresponding complementarity-determining regions, which can be predicted using a machine learning model developed in this work. Notably, a subset of the most important features governing antibody non-specific interactions, namely those that contain tyrosine, also govern specific antigen recognition. Our findings are broadly relevant for understanding fundamental aspects of antibody molecular recognition and the applied aspects of antibody-drug design.

Monoclonal antibodies: From magic bullet to precision weapon

Monoclonal antibodies (mAbs) are used to prevent, detect, and treat a broad spectrum of non-communicable and communicable diseases. Over the past few years, the market for mAbs has grown exponentially with an expected compound annual growth rate (CAGR) of 11.07% from 2024 (237.64 billion USD estimated at the end of 2023) to 2033 (679.03 billion USD expected by the end of 2033). Ever since the advent of hybridoma technology introduced in 1975, antibody-based therapeutics were realized using murine antibodies which further progressed into humanized and fully human antibodies, reducing the risk of immunogenicity. Some benefits of using mAbs over conventional drugs include a drastic reduction in the chances of adverse reactions, interactions between drugs, and targeting specific proteins. While antibodies are very efficient, their higher production costs impede the process of commercialization. However, their cost factor has been improved by developing biosimilar antibodies as affordable versions of therapeutic antibodies. Along with the recent advancements and innovations in antibody engineering have helped and will furtherly help to design bio-better antibodies with improved efficacy than the conventional ones. These novel mAb-based therapeutics are set to revolutionize existing drug therapies targeting a wide spectrum of diseases, thereby meeting several unmet medical needs. This review provides comprehensive insights into the current fundamental landscape of mAbs development and applications and the key factors influencing the future projections, advancement, and incorporation of such promising immunotherapeutic candidates as a confrontation approach against a wide list of diseases, with a rationalistic mentioning of any limitations facing this field.

Inference and design of antibody specificity: From experiments to models and back

Exquisite binding specificity is essential for many protein functions but is difficult to engineer. Many biotechnological or biomedical applications require the discrimination of very similar ligands, which poses the challenge of designing protein sequences with highly specific binding profiles. Experimental methods for generating specific binders rely on in vitro selection, which is limited in terms of library size and control over specificity profiles. Additional control was recently demonstrated through high-throughput sequencing and downstream computational analysis. Here we follow such an approach to demonstrate the design of specific antibodies beyond those probed experimentally. We do so in a context where very similar epitopes need to be discriminated, and where these epitopes cannot be experimentally dissociated from other epitopes present in the selection. Our approach involves the identification of different binding modes, each associated with a particular ligand against which the antibodies are either selected or not. Using data from phage display experiments, we show that the model successfully disentangles these modes, even when they are associated with chemically very similar ligands. Additionally, we demonstrate and validate experimentally the computational design of antibodies with customized specificity profiles, either with specific high affinity for a particular target ligand, or with cross-specificity for multiple target ligands. Overall, our results showcase the potential of leveraging a biophysical model learned from selections against multiple ligands to design proteins with tailored specificity, with applications to protein engineering extending beyond the design of antibodies.

Advancements in mammalian display technology for therapeutic antibody development and beyond: current landscape, challenges, and future prospects

The evolving development landscape of biotherapeutics and their growing complexity from simple antibodies into bi- and multi-specific molecules necessitates sophisticated discovery and engineering platforms. This review focuses on mammalian display technology as a potential solution to the pressing challenges in biotherapeutic development. We provide a comparative analysis with established methodologies, highlighting key aspects of mammalian display technology, including genetic engineering, construction of display libraries, and its pivotal role in hit selection and/or developability engineering. The review delves into the mechanisms underpinning developability-driven selection via mammalian display and their broader implications. Applications beyond antibody discovery are also explored, alongside advancements towards function-first screening technologies, precision genome engineering and AI/ML-enhanced libraries, situating them in the context of mammalian display. Overall, the review provides a comprehensive overview of the current mammalian display technology landscape, underscores the expansive potential of the technology for biotherapeutic development, addresses the critical challenges for the full realisation of this potential, and examines advances in related disciplines that might impact the future application of mammalian display technologies.

Artificial intelligence and machine learning applications for cultured meat

Cultured meat has the potential to provide a complementary meat industry with reduced environmental, ethical, and health impacts. However, major technological challenges remain which require time-and resource-intensive research and development efforts. Machine learning has the potential to accelerate cultured meat technology by streamlining experiments, predicting optimal results, and reducing experimentation time and resources. However, the use of machine learning in cultured meat is in its infancy. This review covers the work available to date on the use of machine learning in cultured meat and explores future possibilities. We address four major areas of cultured meat research and development: establishing cell lines, cell culture media design, microscopy and image analysis, and bioprocessing and food processing optimization. In addition, we have included a survey of datasets relevant to CM research. This review aims to provide the foundation necessary for both cultured meat and machine learning scientists to identify research opportunities at the intersection between cultured meat and machine learning.

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.

Neural network extrapolation to distant regions of the protein fitness landscape

Machine learning (ML) has transformed protein engineering by constructing models of the underlying sequence-function landscape to accelerate the discovery of new biomolecules. ML-guided protein design requires models, trained on local sequence-function information, to accurately predict distant fitness peaks. In this work, we evaluate neural networks' capacity to extrapolate beyond their training data. We perform model-guided design using a panel of neural network architectures trained on protein G (GB1)-Immunoglobulin G (IgG) binding data and experimentally test thousands of GB1 designs to systematically evaluate the models' extrapolation. We find each model architecture infers markedly different landscapes from the same data, which give rise to unique design preferences. We find simpler models excel in local extrapolation to design high fitness proteins, while more sophisticated convolutional models can venture deep into sequence space to design proteins that fold but are no longer functional. We also find that implementing a simple ensemble of convolutional neural networks enables robust design of high-performing variants in the local landscape. Our findings highlight how each architecture's inductive biases prime them to learn different aspects of the protein fitness landscape and how a simple ensembling approach makes protein engineering more robust.

Unsupervised evolution of protein and antibody complexes with a structure-informed language model

Large language models trained on sequence information alone can learn high-level principles of protein design. However, beyond sequence, the three-dimensional structures of proteins determine their specific function, activity, and evolvability. Here, we show that a general protein language model augmented with protein structure backbone coordinates can guide evolution for diverse proteins without the need to model individual functional tasks. We also demonstrate that ESM-IF1, which was only trained on single-chain structures, can be extended to engineer protein complexes. Using this approach, we screened about 30 variants of two therapeutic clinical antibodies used to treat severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection. We achieved up to 25-fold improvement in neutralization and 37-fold improvement in affinity against antibody-escaped viral variants of concern BQ.1.1 and XBB.1.5, respectively. These findings highlight the advantage of integrating structural information to identify efficient protein evolution trajectories without requiring any task-specific training data.

New technologies in therapeutic antibody development: The next frontier for treating infectious diseases

Adaptive immunity to viral infections requires time to neutralize and clear viruses to resolve infection. Fast growing and pathogenic viruses are quickly established, are highly transmissible and cause significant disease burden making it difficult to mount effective responses, thereby prolonging infection. Antibody-based passive immunotherapies can provide initial protection during acute infection, assist in mounting an adaptive immune response, or provide protection for those who are immune suppressed or immune deficient. Historically, plasma-derived antibodies have demonstrated some success in treating diseases caused by viral pathogens; nonetheless, limitations in access to product and antibody titer reduce success of this treatment modality. Monoclonal antibodies (mAbs) have proven an effective alternative, as it is possible to manufacture highly potent and specific mAbs against viral targets on an industrial scale. As a result, innovative technologies to discover, engineer and manufacture specific and potent antibodies have become an essential part of the first line of treatment in pathogenic viral infections. However, a mAb targeting a specific epitope will allow escape variants to outgrow, causing new variant strains to become dominant and resistant to treatment with that mAb. Methods to mitigate escape have included combining mAbs into cocktails, creating bi-specific or antibody drug conjugates but these strategies have also been challenged by the potential development of escape mutations. New technologies in developing antibodies made as recombinant polyclonal drugs can integrate the strength of poly-specific antibody responses to prevent mutational escape, while also incorporating antibody engineering to prevent antibody dependent enhancement and direct adaptive immune responses.

Protein language models enable prediction of polyreactivity of monospecific, bispecific, and heavy-chain-only antibodies

Background

Early assessment of antibody off-target binding is essential for mitigating developability risks such as fast clearance, reduced efficacy, toxicity, and immunogenicity. The baculovirus particle (BVP) binding assay has been widely utilized to evaluate polyreactivity of antibodies. As a complementary approach, computational prediction of polyreactivity is desirable for counter-screening antibodies from in silico discovery campaigns. However, there is a lack of such models.

Methods

Herein, we present the development of an ensemble of three deep learning models based on two pan-protein foundational protein language models (ESM2 and ProtT5) and an antibody-specific protein language model (PLM) (Antiberty). These models were trained in a transfer learning network to predict the outcomes in the BVP assay and the bovine serum albumin binding assay, which was developed as a complement to the BVP assay. The training was conducted on a large dataset of antibody sequences augmented with experimental conditions, which were collected through a highly efficient application system.

Results

The resulting models demonstrated robust performance on canonical mAbs (monospecific with heavy and light chain), bispecific Abs, and single-domain Fc (VHH-Fc). PLMs outperformed a model built using molecular descriptors calculated from AlphaFold 2 predicted structures. Embeddings from the antibody-specific and foundational PLMs resulted in similar performance.

Conclusion

To our knowledge, this represents the first application of PLMs to predict assay data on bispecifics and VHH-Fcs.

Linguistics-based formalization of the antibody language as a basis for antibody language models

Apparent parallels between natural language and antibody sequences have led to a surge in deep language models applied to antibody sequences for predicting cognate antigen recognition. However, a linguistic formal definition of antibody language does not exist, and insight into how antibody language models capture antibody-specific binding features remains largely uninterpretable. Here we describe how a linguistic formalization of the antibody language, by characterizing its tokens and grammar, could address current challenges in antibody language model rule mining.

A Comprehensive Review of Monoclonal Antibodies in Modern Medicine: Tracing the Evolution of a Revolutionary Therapeutic Approach

Monoclonal antibodies (mAbs) have emerged as potent therapeutic agents, revolutionizing the landscape of modern medicine. This comprehensive review traces the evolution of mAbs from their inception to their current prominence, highlighting key milestones in their development and exploring their diverse therapeutic applications. Beginning with an overview of their molecular structure and mechanisms of action, we delve into the production and engineering of mAbs, including hybridoma technology and recombinant DNA techniques. Therapeutic applications across various medical disciplines, including cancer treatment, autoimmune diseases, and infectious diseases, are examined in detail, showcasing the significant clinical successes of mAbs. Furthermore, this review discusses the challenges and opportunities in manufacturing scalability, cost-effectiveness, and access to therapies. Looking ahead, the implications of mAbs in future research and clinical practice are explored, emphasizing the potential for next-generation mAbs, personalized medicine, and integration with emerging modalities such as immunotherapy and gene therapy. In conclusion, the evolution of monoclonal antibodies underscores their transformative impact on healthcare and their continued promise to advance the frontiers of medicine.

Prospects for the computational humanization of antibodies and nanobodies

To be viable therapeutics, antibodies must be tolerated by the human immune system. Rational approaches to reduce the risk of unwanted immunogenicity involve maximizing the 'humanness' of the candidate drug. However, despite the emergence of new discovery technologies, many of which start from entirely human gene fragments, most antibody therapeutics continue to be derived from non-human sources with concomitant humanization to increase their human compatibility. Early experimental humanization strategies that focus on CDR loop grafting onto human frameworks have been critical to the dominance of this discovery route but do not consider the context of each antibody sequence, impacting their success rate. Other challenges include the simultaneous optimization of other drug-like properties alongside humanness and the humanization of fundamentally non-human modalities such as nanobodies. Significant efforts have been made to develop in silico methodologies able to address these issues, most recently incorporating machine learning techniques. Here, we outline these recent advancements in antibody and nanobody humanization, focusing on computational strategies that make use of the increasing volume of sequence and structural data available and the validation of these tools. We highlight that structural distinctions between antibodies and nanobodies make the application of antibody-focused in silico tools to nanobody humanization non-trivial. Furthermore, we discuss the effects of humanizing mutations on other essential drug-like properties such as binding affinity and developability, and methods that aim to tackle this multi-parameter optimization problem.

An integrated technology for quantitative wide mutational scanning of human antibody Fab libraries

Antibodies are engineerable quantities in medicine. Learning antibody molecular recognition would enable the in silico design of high affinity binders against nearly any proteinaceous surface. Yet, publicly available experiment antibody sequence-binding datasets may not contain the mutagenic, antigenic, or antibody sequence diversity necessary for deep learning approaches to capture molecular recognition. In part, this is because limited experimental platforms exist for assessing quantitative and simultaneous sequence-function relationships for multiple antibodies. Here we present MAGMA-seq, an integrated technology that combines multiple antigens and multiple antibodies and determines quantitative biophysical parameters using deep sequencing. We demonstrate MAGMA-seq on two pooled libraries comprising mutants of nine different human antibodies spanning light chain gene usage, CDR H3 length, and antigenic targets. We demonstrate the comprehensive mapping of potential antibody development pathways, sequence-binding relationships for multiple antibodies simultaneously, and identification of paratope sequence determinants for binding recognition for broadly neutralizing antibodies (bnAbs). MAGMA-seq enables rapid and scalable antibody engineering of multiple lead candidates because it can measure binding for mutants of many given parental antibodies in a single experiment.

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.

Machine learning to predict continuous protein properties from binary cell sorting data and map unseen sequence space

Proteins are a diverse class of biomolecules responsible for wide-ranging cellular functions, from catalyzing reactions to recognizing pathogens. The ability to evolve proteins rapidly and inexpensively toward improved properties is a common objective for protein engineers. Powerful high-throughput methods like fluorescent activated cell sorting and next-generation sequencing have dramatically improved directed evolution experiments. However, it is unclear how to best leverage these data to characterize protein fitness landscapes more completely and identify lead candidates. In this work, we develop a simple yet powerful framework to improve protein optimization by predicting continuous protein properties from simple directed evolution experiments using interpretable, linear machine learning models. Importantly, we find that these models, which use data from simple but imprecise experimental estimates of protein fitness, have predictive capabilities that approach more precise but expensive data. Evaluated across five diverse protein engineering tasks, continuous properties are consistently predicted from readily available deep sequencing data, demonstrating that protein fitness space can be reasonably well modeled by linear relationships among sequence mutations. To prospectively test the utility of this approach, we generated a library of stapled peptides and applied the framework to predict affinity and specificity from simple cell sorting data. We then coupled integer linear programming, a method to optimize protein fitness from linear weights, with mutation scores from machine learning to identify variants in unseen sequence space that have improved and co-optimal properties. This approach represents a versatile tool for improved analysis and identification of protein variants across many domains of protein engineering.

Bispecific antibody shuttles targeting CD98hc mediate efficient and long-lived brain delivery of IgGs

The inability of antibodies to penetrate the blood-brain barrier (BBB) is a key limitation to their use in diverse applications. One promising strategy is to deliver IgGs using a bispecific BBB shuttle, which involves fusing an IgG to a second affinity ligand that engages a cerebrovascular endothelial target and facilitates transport across the BBB. Nearly all prior efforts have focused on shuttles that target transferrin receptor (TfR-1) despite inherent delivery and safety challenges. Here, we report bispecific antibody shuttles that engage CD98hc, the heavy chain of the large neutral amino acid transporter (LAT1), and efficiently transport IgGs into the brain. Notably, CD98hc shuttles lead to much longer-lived brain retention of IgGs than TfR-1 shuttles while enabling more specific targeting due to limited CD98hc engagement in the brain parenchyma, which we demonstrate for IgGs that either agonize a neuronal receptor (TrkB) or target other endogenous cell-surface proteins on neurons and astrocytes.

Machine learning for functional protein design

Recent breakthroughs in AI coupled with the rapid accumulation of protein sequence and structure data have radically transformed computational protein design. New methods promise to escape the constraints of natural and laboratory evolution, accelerating the generation of proteins for applications in biotechnology and medicine. To make sense of the exploding diversity of machine learning approaches, we introduce a unifying framework that classifies models on the basis of their use of three core data modalities: sequences, structures and functional labels. We discuss the new capabilities and outstanding challenges for the practical design of enzymes, antibodies, vaccines, nanomachines and more. We then highlight trends shaping the future of this field, from large-scale assays to more robust benchmarks, multimodal foundation models, enhanced sampling strategies and laboratory automation.

Meta learning addresses noisy and under-labeled data in machine learning-guided antibody engineering

Machine learning-guided protein engineering is rapidly progressing; however, collecting high-quality, large datasets remains a bottleneck. Directed evolution and protein engineering studies often require extensive experimental processes to eliminate noise and label protein sequence-function data. Meta learning has proven effective in other fields in learning from noisy data via bi-level optimization given the availability of a small dataset with trusted labels. Here, we leverage meta learning approaches to overcome noisy and under-labeled data and expedite workflows in antibody engineering. We generate yeast display antibody mutagenesis libraries and screen them for target antigen binding followed by deep sequencing. We then create representative learning tasks, including learning from noisy training data, positive and unlabeled learning, and learning out of distribution properties. We demonstrate that meta learning has the potential to reduce experimental screening time and improve the robustness of machine learning models by training with noisy and under-labeled training data.

An integrated technology for quantitative wide mutational scanning of human antibody Fab libraries

Antibodies are engineerable quantities in medicine. Learning antibody molecular recognition would enable the in silico design of high affinity binders against nearly any proteinaceous surface. Yet, publicly available experiment antibody sequence-binding datasets may not contain the mutagenic, antigenic, or antibody sequence diversity necessary for deep learning approaches to capture molecular recognition. In part, this is because limited experimental platforms exist for assessing quantitative and simultaneous sequence-function relationships for multiple antibodies. Here we present MAGMA-seq, an integrated technology that combines multiple antigens and multiple antibodies and determines quantitative biophysical parameters using deep sequencing. We demonstrate MAGMA-seq on two pooled libraries comprising mutants of ten different human antibodies spanning light chain gene usage, CDR H3 length, and antigenic targets. We demonstrate the comprehensive mapping of potential antibody development pathways, sequence-binding relationships for multiple antibodies simultaneously, and identification of paratope sequence determinants for binding recognition for broadly neutralizing antibodies (bnAbs). MAGMA-seq enables rapid and scalable antibody engineering of multiple lead candidates because it can measure binding for mutants of many given parental antibodies in a single experiment.

Advancing Antibody Engineering through Synthetic Evolution and Machine Learning

Abs are versatile molecules with the potential to achieve exceptional binding to target Ags, while also possessing biophysical properties suitable for therapeutic drug development. Protein display and directed evolution systems have transformed synthetic Ab discovery, engineering, and optimization, vastly expanding the number of Ab clones able to be experimentally screened for binding. Moreover, the burgeoning integration of high-throughput screening, deep sequencing, and machine learning has further augmented in vitro Ab optimization, promising to accelerate the design process and massively expand the Ab sequence space interrogated. In this Brief Review, we discuss the experimental and computational tools employed in synthetic Ab engineering and optimization. We also explore the therapeutic challenges posed by developing Abs for infectious diseases, and the prospects for leveraging machine learning-guided protein engineering to prospectively design Abs resistant to viral escape.

Developability considerations for bispecific and multispecific antibodies

Bispecific antibodies (bsAb) and multispecific antibodies (msAb) encompass a diverse variety of formats that can concurrently bind multiple epitopes, unlocking mechanisms to address previously difficult-to-treat or incurable diseases. Early assessment of candidate developability enables demotion of antibodies with low potential and promotion of the most promising candidates for further development. Protein-based therapies have a stringent set of developability requirements in order to be competitive (e.g. high-concentration formulation, and long half-life) and their assessment requires a robust toolkit of methods, few of which are validated for interrogating bsAbs/msAbs. Important considerations when assessing the developability of bsAbs/msAbs include their molecular format, likelihood for immunogenicity, specificity, stability, and potential for high-volume production. Here, we summarize the critical aspects of developability assessment, and provide guidance on how to develop a comprehensive plan tailored to a given bsAb/msAb.

Reduction of monoclonal antibody viscosity using interpretable machine learning

Early identification of antibody candidates with drug-like properties is essential for simplifying the development of safe and effective antibody therapeutics. For subcutaneous administration, it is important to identify candidates with low self-association to enable their formulation at high concentration while maintaining low viscosity, opalescence, and aggregation. Here, we report an interpretable machine learning model for predicting antibody (IgG1) variants with low viscosity using only the sequences of their variable (Fv) regions. Our model was trained on antibody viscosity data (>100 mg/mL mAb concentration) obtained at a common formulation pH (pH 5.2), and it identifies three key Fv features of antibodies linked to viscosity, namely their isoelectric points, hydrophobic patch sizes, and numbers of negatively charged patches. Of the three features, most predicted antibodies at risk for high viscosity, including antibodies with diverse antibody germlines in our study (79 mAbs) as well as clinical-stage IgG1s (94 mAbs), are those with low Fv isoelectric points (Fv pIs < 6.3). Our model identifies viscous antibodies with relatively high accuracy not only in our training and test sets, but also for previously reported data. Importantly, we show that the interpretable nature of the model enables the design of mutations that significantly reduce antibody viscosity, which we confirmed experimentally. We expect that this approach can be readily integrated into the drug development process to reduce the need for experimental viscosity screening and improve the identification of antibody candidates with drug-like properties.

Optimization of therapeutic antibodies for reduced self-association and non-specific binding via interpretable machine learning

Antibody development, delivery, and efficacy are influenced by antibody-antigen affinity interactions, off-target interactions that reduce antibody bioavailability and pharmacokinetics, and repulsive self-interactions that increase the stability of concentrated antibody formulations and reduce their corresponding viscosity. Yet identifying antibody variants with optimal combinations of these three types of interactions is challenging. Here we show that interpretable machine-learning classifiers, leveraging antibody structural features descriptive of their variable regions and trained on experimental data for a panel of 80 clinical-stage monoclonal antibodies, can identify antibodies with optimal combinations of low off-target binding in a common physiological-solution condition and low self-association in a common antibody-formulation condition. For three clinical-stage antibodies with suboptimal combinations of off-target binding and self-association, the classifiers predicted variable-region mutations that optimized non-affinity interactions while maintaining high-affinity antibody-antigen interactions. Interpretable machine-learning models may facilitate the optimization of antibody candidates for therapeutic applications.

Nonspecificity fingerprints for clinical-stage antibodies in solution

Monoclonal antibodies (mAbs) have successfully been developed for the treatment of a wide range of diseases. The clinical success of mAbs does not solely rely on optimal potency and safety but also require good biophysical properties to ensure a high developability potential. In particular, nonspecific interactions are a key developability parameter to monitor during discovery and development. Despite an increased focus on the detection of nonspecific interactions, their underlying physicochemical origins remain poorly understood. Here, we employ solution-based microfluidic technologies to characterize a set of clinical-stage mAbs and their interactions with commonly used nonspecificity ligands to generate nonspecificity fingerprints, providing quantitative data on the underlying physical chemistry. Furthermore, the solution-based analysis enables us to measure binding affinities directly, and we evaluate the contribution of avidity in nonspecific binding by mAbs. We find that avidity can increase the apparent affinity by two orders of magnitude. Notably, we find that a subset of these highly developed mAbs show nonspecific electrostatic interactions, even at physiological pH and ionic strength, and that they can form microscale particles with charge-complementary polymers. The group of mAb constructs flagged here for nonspecificity are among the worst performers in independent reports of surface and column-based screens. The solution measurements improve on the state-of-the-art by providing a stand-alone result for individual mAbs without the need to benchmark against cohort data. Based on our findings, we propose a quantitative solution-based nonspecificity score, which can be integrated in the development workflow for biological therapeutics and more widely in protein engineering.

Machine learning-based meta-analysis of colorectal cancer and inflammatory bowel disease

Colorectal cancer (CRC) is a major global health concern, resulting in numerous cancer-related deaths. CRC detection, treatment, and prevention can be improved by identifying genes and biomarkers. Despite extensive research, the underlying mechanisms of CRC remain elusive, and previously identified biomarkers have not yielded satisfactory insights. This shortfall may be attributed to the predominance of univariate analysis methods, which overlook potential combinations of variants and genes contributing to disease development. Here, we address this knowledge gap by presenting a novel multivariate machine-learning strategy to pinpoint genes associated with CRC. Additionally, we applied our analysis pipeline to Inflammatory Bowel Disease (IBD), as IBD patients face substantial CRC risk. The importance of the identified genes was substantiated by rigorous validation across numerous independent datasets. Several of the discovered genes have been previously linked to CRC, while others represent novel findings warranting further investigation. A Python implementation of our pipeline can be accessed publicly at https://github.com/AriaSar/CRCIBD-ML.

Inverse folding of protein complexes with a structure-informed language model enables unsupervised antibody evolution

Large language models trained on sequence information alone are capable of learning high level principles of protein design. However, beyond sequence, the three-dimensional structures of proteins determine their specific function, activity, and evolvability. Here we show that a general protein language model augmented with protein structure backbone coordinates and trained on the inverse folding problem can guide evolution for diverse proteins without needing to explicitly model individual functional tasks. We demonstrate inverse folding to be an effective unsupervised, structure-based sequence optimization strategy that also generalizes to multimeric complexes by implicitly learning features of binding and amino acid epistasis. Using this approach, we screened ~30 variants of two therapeutic clinical antibodies used to treat SARS-CoV-2 infection and achieved up to 26-fold improvement in neutralization and 37-fold improvement in affinity against antibody-escaped viral variants-of-concern BQ.1.1 and XBB.1.5, respectively. In addition to substantial overall improvements in protein function, we find inverse folding performs with leading experimental success rates among other reported machine learning-guided directed evolution methods, without requiring any task-specific training data.

Advances in ligand-specific biosensing for structurally similar molecules

The specificity of biological systems makes it possible to develop biosensors targeting specific metabolites, toxins, and pollutants in complex medical or environmental samples without interference from structurally similar compounds. For the last two decades, great efforts have been devoted to creating proteins or nucleic acids with novel properties through synthetic biology strategies. Beyond augmenting biocatalytic activity, expanding target substrate scopes, and enhancing enzymes' enantioselectivity and stability, an increasing research area is the enhancement of molecular specificity for genetically encoded biosensors. Here, we summarize recent advances in the development of highly specific biosensor systems and their essential applications. First, we describe the rational design principles required to create libraries containing potential mutants with less promiscuity or better specificity. Next, we review the emerging high-throughput screening techniques to engineer biosensing specificity for the desired target. Finally, we examine the computer-aided evaluation and prediction methods to facilitate the construction of ligand-specific biosensors.