Structure-based cross-docking analysis of antibody-antigen interactions

Machine-learning-based structural analysis of interactions between antibodies and antigens

Computational analysis of paratope-epitope interactions between antibodies and their corresponding antigens can facilitate our understanding of the molecular mechanism underlying humoral immunity and boost the design of new therapeutics for many diseases. The recent breakthrough in artificial intelligence has made it possible to predict protein-protein interactions and model their structures. Unfortunately, detecting antigen-binding sites associated with a specific antibody is still a challenging problem. To tackle this challenge, we implemented a deep learning model to characterize interaction patterns between antibodies and their corresponding antigens. With high accuracy, our model can distinguish between antibody-antigen complexes and other types of protein-protein complexes. More intriguingly, we can identify antigens from other common protein binding regions with an accuracy of higher than 70% even if we only have the epitope information. This indicates that antigens have distinct features on their surface that antibodies can recognize. Additionally, our model was unable to predict the partnerships between antibodies and their particular antigens. This result suggests that one antigen may be targeted by more than one antibody and that antibodies may bind to previously unidentified proteins. Taken together, our results support the precision of antibody-antigen interactions while also suggesting positive future progress in the prediction of specific pairing.

ABAG-docking benchmark: a non-redundant structure benchmark dataset for antibody-antigen computational docking

Accurate prediction of antibody-antigen complex structures is pivotal in drug discovery, vaccine design and disease treatment and can facilitate the development of more effective therapies and diagnostics. In this work, we first review the antibody-antigen docking (ABAG-docking) datasets. Then, we present the creation and characterization of a comprehensive benchmark dataset of antibody-antigen complexes. We categorize the dataset based on docking difficulty, interface properties and structural characteristics, to provide a diverse set of cases for rigorous evaluation. Compared with Docking Benchmark 5.5, we have added 112 cases, including 14 single-domain antibody (sdAb) cases and 98 monoclonal antibody (mAb) cases, and also increased the proportion of Difficult cases. Our dataset contains diverse cases, including human/humanized antibodies, sdAbs, rodent antibodies and other types, opening the door to better algorithm development. Furthermore, we provide details on the process of building the benchmark dataset and introduce a pipeline for periodic updates to keep it up to date. We also utilize multiple complex prediction methods including ZDOCK, ClusPro, HDOCK and AlphaFold-Multimer for testing and analyzing this dataset. This benchmark serves as a valuable resource for evaluating and advancing docking computational methods in the analysis of antibody-antigen interaction, enabling researchers to develop more accurate and effective tools for predicting and designing antibody-antigen complexes. The non-redundant ABAG-docking structure benchmark dataset is available at https://github.com/Zhaonan99/Antibody-antigen-complex-structure-benchmark-dataset.

Machine-learning-based Structural Analysis of Interactions between Antibodies and Antigens

Computational analysis of paratope-epitope interactions between antibodies and their corresponding antigens can facilitate our understanding of the molecular mechanism underlying humoral immunity and boost the design of new therapeutics for many diseases. The recent breakthrough in artificial intelligence has made it possible to predict protein-protein interactions and model their structures. Unfortunately, detecting antigen-binding sites associated with a specific antibody is still a challenging problem. To tackle this challenge, we implemented a deep learning model to characterize interaction patterns between antibodies and their corresponding antigens. With high accuracy, our model can distinguish between antibody-antigen complexes and other types of protein-protein complexes. More intriguingly, we can identify antigens from other common protein binding regions with an accuracy of higher than 70% even if we only have the epitope information. This indicates that antigens have distinct features on their surface that antibodies can recognize. Additionally, our model was unable to predict the partnerships between antibodies and their particular antigens. This result suggests that one antigen may be targeted by more than one antibody and that antibodies may bind to previously unidentified proteins. Taken together, our results support the precision of antibody-antigen interactions while also suggesting positive future progress in the prediction of specific pairing.

Structural Modeling of Adaptive Immune Responses to Infection

Antibody and TCR modeling are becoming important as more and more sequence data becomes available to the public. One of the pressing questions now is how to use such data to understand adaptive immune responses to disease. Infectious disease is of particular interest because the antigens driving such responses are often known to some extent. Here, we describe tips for gathering data and cleaning it for use in downstream analysis. We present a method for high-throughput structural modeling of antibodies or TCRs using Repertoire Builder and its extensions. AbAdapt is an extension of Repertoire Builder for antibody-antigen docking from antibody and antigen sequences. ImmuneScape is a corresponding extension for TCR-pMHC 3D modeling. Together, these pipelines can help researchers to understand immune responses to infection from a structural point of view.

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.

Prediction of Antibody-Antigen Binding via Machine Learning: Development of Data Sets and Evaluation of Methods

BACKGROUND

The mammalian immune system is able to generate antibodies against a huge variety of antigens, including bacteria, viruses, and toxins. The ultradeep DNA sequencing of rearranged immunoglobulin genes has considerable potential in furthering our understanding of the immune response, but it is limited by the lack of a high-throughput, sequence-based method for predicting the antigen(s) that a given immunoglobulin recognizes.

OBJECTIVE

As a step toward the prediction of antibody-antigen binding from sequence data alone, we aimed to compare a range of machine learning approaches that were applied to a collated data set of antibody-antigen pairs in order to predict antibody-antigen binding from sequence data.

METHODS

Data for training and testing were extracted from the Protein Data Bank and the Coronavirus Antibody Database, and additional antibody-antigen pair data were generated by using a molecular docking protocol. Several machine learning methods, including the weighted nearest neighbor method, the nearest neighbor method with the BLOSUM62 matrix, and the random forest method, were applied to the problem.

RESULTS

The final data set contained 1157 antibodies and 57 antigens that were combined in 5041 antibody-antigen pairs. The best performance for the prediction of interactions was obtained by using the nearest neighbor method with the BLOSUM62 matrix, which resulted in around 82% accuracy on the full data set. These results provide a useful frame of reference, as well as protocols and considerations, for machine learning and data set creation in the prediction of antibody-antigen binding.

CONCLUSIONS

Several machine learning approaches were compared to predict antibody-antigen interaction from protein sequences. Both the data set (in CSV format) and the machine learning program (coded in Python) are freely available for download on GitHub.

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.

From complete cross-docking to partners identification and binding sites predictions

Proteins ensure their biological functions by interacting with each other. Hence, characterising protein interactions is fundamental for our understanding of the cellular machinery, and for improving medicine and bioengineering. Over the past years, a large body of experimental data has been accumulated on who interacts with whom and in what manner. However, these data are highly heterogeneous and sometimes contradictory, noisy, and biased. Ab initio methods provide a means to a "blind" protein-protein interaction network reconstruction. Here, we report on a molecular cross-docking-based approach for the identification of protein partners. The docking algorithm uses a coarse-grained representation of the protein structures and treats them as rigid bodies. We applied the approach to a few hundred of proteins, in the unbound conformations, and we systematically investigated the influence of several key ingredients, such as the size and quality of the interfaces, and the scoring function. We achieved some significant improvement compared to previous works, and a very high discriminative power on some specific functional classes. We provide a readout of the contributions of shape and physico-chemical complementarity, interface matching, and specificity, in the predictions. In addition, we assessed the ability of the approach to account for protein surface multiple usages, and we compared it with a sequence-based deep learning method. This work may contribute to guiding the exploitation of the large amounts of protein structural models now available toward the discovery of unexpected partners and their complex structure characterisation.

Understanding memory B cell selection

The mammalian adaptive immune system has evolved over millions of years to become an incredibly effective defense against foreign antigens. The adaptive immune system's humoral response creates plasma B cells and memory B cells, each with their own immunological objectives. The affinity maturation process is widely viewed as a heuristic to solve the global optimization problem of finding B cells with high affinity to the antigen. However, memory B cells appear to be purposely selected earlier in the affinity maturation process and have lower affinity. We propose that this memory B cell selection process may be an approximate solution to two optimization problems: optimizing for affinity to similar antigens in the future despite mutations or other minor differences, and optimizing to warm start the generation of plasma B cells in the future. We use simulations to provide evidence for our hypotheses, taking into account data showing that certain B cell mutations are more likely than others. Our findings are consistent with memory B cells having high-affinity to mutated antigens, but do not provide strong evidence that memory B cells will be more useful than selected naive B cells for seeding the secondary germinal centers.

Functional antibody characterization via direct structural analysis and information-driven protein-protein docking

Detailed description of the mechanism of action of the therapeutic antibodies is essential for the functional characterization and future optimization of potential clinical agents. We recently developed KD035, a fully human antibody targeting vascular endothelial growth factor receptor 2 (VEGFR2). KD035 blocked VEGF‐A, and VEGF‐C‐mediated VEGFR2 activation, as demonstrated by the in vitro binding and competition assays and functional cellular assays. Here, we report a computational model of the complex between the variable fragment of KD035 (KD035(Fv)) and the domains 2 and 3 of the extracellular portion of VEGFR2 (VEGFR2(D2‐3)). Our modeling was guided by a priori experimental information including the X‐ray structures of KD035 and related antibodies, binding assays, target domain mapping and comparison of KD035 affinity for VEGFR2 from different species. The accuracy of the model was assessed by molecular dynamics simulations, and subsequently validated by mutagenesis and binding analysis. Importantly, the steps followed during the generation of this model can set a precedent for future in silico efforts aimed at the accurate description of the antibody–antigen and more broadly protein–protein complexes.

DLAB: deep learning methods for structure-based virtual screening of antibodies

MOTIVATION

Antibodies are one of the most important classes of pharmaceuticals, with over 80 approved molecules currently in use against a wide variety of diseases. The drug discovery process for antibody therapeutic candidates however is time- and cost-intensive and heavily reliant on in vivo and in vitro high throughput screens. Here, we introduce a framework for structure-based deep learning for antibodies (DLAB) which can virtually screen putative binding antibodies against antigen targets of interest. DLAB is built to be able to predict antibody-antigen binding for antigens with no known antibody binders.

RESULTS

We demonstrate that DLAB can be used both to improve antibody-antigen docking and structure-based virtual screening of antibody drug candidates. DLAB enables improved pose ranking for antibody docking experiments as well as selection of antibody-antigen pairings for which accurate poses are generated and correctly ranked. We also show that DLAB can identify binding antibodies against specific antigens in a case study. Our results demonstrate the promise of deep learning methods for structure-based virtual screening of antibodies.

AVAILABILITY AND IMPLEMENTATION

The DLAB source code and pre-trained models are available at https://github.com/oxpig/dlab-public.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

A compact vocabulary of paratope-epitope interactions enables predictability of antibody-antigen binding

Antibody-antigen binding relies on the specific interaction of amino acids at the paratope-epitope interface. The predictability of antibody-antigen binding is a prerequisite for de novo antibody and (neo-)epitope design. A fundamental premise for the predictability of antibody-antigen binding is the existence of paratope-epitope interaction motifs that are universally shared among antibody-antigen structures. In a dataset of non-redundant antibody-antigen structures, we identify structural interaction motifs, which together compose a commonly shared structure-based vocabulary of paratope-epitope interactions. We show that this vocabulary enables the machine learnability of antibody-antigen binding on the paratope-epitope level using generative machine learning. The vocabulary (1) is compact, less than 10⁴ motifs; (2) distinct from non-immune protein-protein interactions; and (3) mediates specific oligo- and polyreactive interactions between paratope-epitope pairs. Our work leverages combined structure- and sequence-based learning to demonstrate that machine-learning-driven predictive paratope and epitope engineering is feasible.

An expanded benchmark for antibody-antigen docking and affinity prediction reveals insights into antibody recognition determinants

Accurate predictive modeling of antibody-antigen complex structures and structure-based antibody design remain major challenges in computational biology, with implications for biotherapeutics, immunity, and vaccines. Through a systematic search for high-resolution structures of antibody-antigen complexes and unbound antibody and antigen structures, in conjunction with identification of experimentally determined binding affinities, we have assembled a non-redundant set of test cases for antibody-antigen docking and affinity prediction. This benchmark more than doubles the number of antibody-antigen complexes and corresponding affinities available in our previous benchmarks, providing an unprecedented view of the determinants of antibody recognition and insights into molecular flexibility. Initial assessments of docking and affinity prediction tools highlight the challenges posed by this diverse set of cases, which includes camelid nanobodies, therapeutic monoclonal antibodies, and broadly neutralizing antibodies targeting viral glycoproteins. This dataset will enable development of advanced predictive modeling and design methods for this therapeutically relevant class of protein-protein interactions.

Structure and Functional Binding Epitope of V-domain Ig Suppressor of T Cell Activation

V-domain immunoglobulin (Ig) suppressor of T cell activation (VISTA) is an immune checkpoint protein that inhibits the T cell response against cancer. Similar to PD-1 and CTLA-4, a blockade of VISTA promotes tumor clearance by the immune system. Here, we report a 1.85 Å crystal structure of the elusive human VISTA extracellular domain, whose lack of homology necessitated a combinatorial MR-Rosetta approach for structure determination. We highlight features that make the VISTA immunoglobulin variable (IgV)-like fold unique among B7 family members, including two additional disulfide bonds and an extended loop region with an attached helix that we show forms a contiguous binding epitope for a clinically relevant anti-VISTA antibody. We propose an overlap of this antibody-binding region with the binding epitope for V-set and Ig domain containing 3 (VSIG3), a purported functional binding partner of VISTA. The structure and functional epitope presented here will help guide future drug development efforts against this important checkpoint target.

Computational Design of a Novel VLP-Based Vaccine for Hepatitis B Virus

Hepatitis B virus (HBV) is a global virus responsible for a universal disease burden for millions of people. Various vaccination strategies have been developed using viral vector, nucleic acid, protein, peptide, and virus-like particles (VLPs) to stimulate favorable immune responses against HBV. Given the pivotal role of specific immune responses of hepatitis B surface antigen (HBsAg) and hepatitis B core antigen (HBcAg) in infection control, we designed a VLP-based vaccine by placing the antibody-binding fragments of HBsAg in the major immunodominant region (MIR) epitope of HBcAg to stimulate multilateral immunity. A computational approach was employed to predict and evaluate the conservation, antigenicity, allergenicity, and immunogenicity of the construct. Modeling and molecular dynamics (MD) demonstrated the folding stability of HBcAg as a carrier in inserting Myrcludex and "a" determinant of HBsAg. Regions 1-50 and 118-150 of HBsAg were considered to have the highest stability to be involved in the designed vaccine. Molecular docking revealed appropriate interactions between the B cell epitope of the designed vaccine and the antibodies. Totally, the final construct was promising for inducing humoral and cellular responses against HBV.

Guidelines To Predict Binding Poses of Antibody-Integrin Complexes

Integrins are cell adhesion receptors that transmit bidirectional signals across the plasma membrane. They are noncovalently linked heterodimeric molecules consisting of two subunits and act as biomarkers in several pathologies. Thus, according to the increase of therapeutic antibody production, some efforts have been applied to produce anti-integrin antibodies. Here, we purposed to evaluate methods of generation and identification of the binding pose of integrin-antibody complexes, through protein-protein docking and molecular dynamics simulations, and propose a strategy to assure the confidence of the final model and avoid false-positive poses. The results show that ClusPro and GRAMM-X were the best programs to generate the native pose of integrin-antibody complexes. Furthermore, we were able to recover and to ensure that the selected pose is the native one by using a simple rule. All complexes from ClusPro in which the first model had the lowest energy, at least 5% more negative than the second one, were correctly predicted. Therefore, our methodology seems to be efficient to avoid misranking of wrong poses for integrin-antibody complexes. In cases where the rule is inconclusive, we proposed the use of heated molecular dynamics to identify the native pose characterized by RMSDi <0.5 nm. We believe that the set of methods presented here helps in the rational design of anti-integrin antibodies, giving some insights on the development of new biopharmaceuticals.

Prediction of Contact Residues in Anti-HIV Neutralizing Antibody by Deep Learning

The monoclonal antibody 1C10 targets the V3 loop of HIV-1 and neutralizes a broad range of clade B viruses. However, the mode of interaction between 1C10 and the V3 loop remains unclear because crystallization of 1C10 and the V3 peptide was unsuccessful due to the flexible regions present in both 1C10 and the V3 peptide. In this study, we predicted the 1C10 amino acid residues that make contact with the V3 loop using a deep learning (DL)-based method. Inputs from ROSIE for docking simulation and FastContact, Naccess, and PDBtools, to approximate interactions were processed by Chainer for DL, and outputs were obtained as probabilities of contact residues. Using this DL algorithm, D95, D97, P100a, and D100b of CDRH3; D53, and D56 of CDRH2; and D61 of FR3 were highly ranked as contact residues of 1C10. Substitution of these residues with alanine significantly decreased the affinity of 1C10 to the V3 peptide. Moreover, the higher the rank of the residue, the more the binding activity diminished. This study demonstrates that the prediction of contact residues using a DL-based approach is a precise and useful tool for the analysis of antibody-antigen interactions.

Trends in therapeutic antibody affinity maturation: From in-vitro towards next-generation sequencing approaches

Current advances in antibody engineering driving the strongest growth area in biotherapeutic agents development. Affinity improvement that is mainly important for biological activity and clinical efficacy of therapeutic antibodies, has still remained a challenging task. In the human body, during a course of immune response affinity maturation increase antibody activity by several rounds of somatic hypermutation and clonal selection in the germinal center. The final outputs are antibodies representing higher affinity and specificity against a particular antigen. In the realm of biotechnology, exploring of mutations which improve antibody affinity while preserving its specificity and stability is an extremely time-consuming and laborious process. Recent advances in computational algorithms and DNA sequencing technologies help researchers to redesign antibody structure to achieve desired properties such as improved binding affinity. In this review, we briefly described the principle of affinity maturation and different corresponding in vitro techniques. Also, we recapitulated the most recent advancements in the field of antibody affinity maturation including computational approaches and next-generation sequencing (NGS).

Homology modeling identified for purported drug targets to the neuroprotective effects of levodopa and asiaticoside-D in degenerated cerebral ganglions of Lumbricus terrestris

CONTEXT:

Homology modeling plays role in determining the therapeutic targets dreadful for condition such as neurodegenerative diseases (NDD), which pose challenge in achieving the effective managements. The structures of the serotonin transporter (SERT), aquaporin (AQP), and tropomyosin receptor kinase (TrkA) which are implicated in NDD pathology are still unknown for Lumbricus terrestris, but the three-dimensional (3D) structure of the human counterpart for modeling.

AIM:

This study aims to generate and evaluate the 3D structure of TrkA, SERT, and AQP proteins and their interaction with the ligands, namely Asiaticoside-D (AD) and levodopa (L-DOPA) the anti-NDD agents.

SUBJECTS AND METHODS:

Homology modeling for SERT, AQP, and TrkA proteins of Lumbricus terrestris using SWISS-MODEL Server and the modeled structure was validated using Rampage Server. Wet-lab analysis of their correspondent m-RNA levels was also done to validate the in silico data.

RESULTS:

It was found that TrkA had moderately high homology (67%) to human while SERT and AQP could exhibit 58% and 42%, respectively. The reliability of the model was assessed by Ramachandran plot analysis. Interactions of AD with the SERT, AQP-4, and TrkA showed the binding energies as −9.93, 8.88, and −7.58 of Kcal/mol, respectively, while for L-DOPA did show −3.93, −5.13, and −6.0 Kcal/mol, respectively. The levels of SERT, TrkA, and AQP-4 were significantly reduced (P < 0.001) on ROT induced when compared to those of control worms. On ROT + AD supplementation group (III), m-RNA levels were significantly increased (P < 0.05) when compared to those of ROT induced worms (group II).

CONCLUSION:

Our pioneering docking data propose the possible of target which is proved useful for therapeutic investigations against the unconquered better of NDD.

Antibody Specific B-Cell Epitope Predictions: Leveraging Information From Antibody-Antigen Protein Complexes

B-cells can neutralize pathogenic molecules by targeting them with extreme specificity using receptors secreted or expressed on their surface (antibodies). This is achieved via molecular interactions between the paratope (i.e., the antibody residues involved in the binding) and the interacting region (epitope) of its target molecule (antigen). Discerning the rules that define this specificity would have profound implications for our understanding of humoral immunogenicity and its applications. The aim of this work is to produce improved, antibody-specific epitope predictions by exploiting features derived from the antigens and their cognate antibodies structures, and combining them using statistical and machine learning algorithms. We have identified several geometric and physicochemical features that are correlated in interacting paratopes and epitopes, used them to develop a Monte Carlo algorithm to generate putative epitopes-paratope pairs, and train a machine-learning model to score them. We show that, by including the structural and physicochemical properties of the paratope, we improve the prediction of the target of a given B-cell receptor. Moreover, we demonstrate a gain in predictive power both in terms of identifying the cognate antigen target for a given antibody and the antibody target for a given antigen, exceeding the results of other available tools.

Non-H3 CDR template selection in antibody modeling through machine learning

Antibodies are proteins generated by the adaptive immune system to recognize and counteract a plethora of pathogens through specific binding. This adaptive binding is mediated by structural diversity in the six complementary determining region (CDR) loops (H1, H2, H3, L1, L2 and L3), which also makes accurate structural modeling of CDRs challenging. Both homology and de novo modeling approaches have been used; to date, the former has achieved greater accuracy for the non-H3 loops. The homology modeling of non-H3 CDRs is more accurate because non-H3 CDR loops of the same length and type can be grouped into a few structural clusters. Most antibody-modeling suites utilize homology modeling for the non-H3 CDRs, differing only in the alignment algorithm and how/if they utilize structural clusters. While RosettaAntibody and SAbPred do not explicitly assign query CDR sequences to clusters, two other approaches, PIGS and Kotai Antibody Builder, utilize sequence-based rules to assign CDR sequences to clusters. While the manually curated sequence rules can identify better structural templates, because their curation requires extensive literature search and human effort, they lag behind the deposition of new antibody structures and are infrequently updated. In this study, we propose a machine learning approach (Gradient Boosting Machine [GBM]) to learn the structural clusters of non-H3 CDRs from sequence alone. The GBM method simplifies feature selection and can easily integrate new data, compared to manual sequence rule curation. We compare the classification results using the GBM method to that of RosettaAntibody in a 3-repeat 10-fold cross-validation (CV) scheme on the cluster-annotated antibody database PyIgClassify and we observe an improvement in the classification accuracy of the concerned loops from 84.5% ± 0.24% to 88.16% ± 0.056%. The GBM models reduce the errors in specific cluster membership misclassifications when the involved clusters have relatively abundant data. Based on the factors identified, we suggest methods that can enrich structural classes with sparse data to further improve prediction accuracy in future studies.

Third generation antibody discovery methods: in silico rational design

Owing to their outstanding performances in molecular recognition, antibodies are extensively used in research and applications in molecular biology, biotechnology and medicine. Recent advances in experimental and computational methods are making it possible to complement well-established in vivo (first generation) and in vitro (second generation) methods of antibody discovery with novel in silico (third generation) approaches. Here we describe the principles of computational antibody design and review the state of the art in this field. We then present Modular, a method that implements the rational design of antibodies in a modular manner, and describe the opportunities offered by this approach.

A combined computational-experimental approach to define the structural origin of antibody recognition of sialyl-Tn, a tumor-associated carbohydrate antigen

Anti-carbohydrate monoclonal antibodies (mAbs) hold great promise as cancer therapeutics and diagnostics. However, their specificity can be mixed, and detailed characterization is problematic, because antibody-glycan complexes are challenging to crystallize. Here, we developed a generalizable approach employing high-throughput techniques for characterizing the structure and specificity of such mAbs, and applied it to the mAb TKH2 developed against the tumor-associated carbohydrate antigen sialyl-Tn (STn). The mAb specificity was defined by apparent K_D values determined by quantitative glycan microarray screening. Key residues in the antibody combining site were identified by site-directed mutagenesis, and the glycan-antigen contact surface was defined using saturation transfer difference NMR (STD-NMR). These features were then employed as metrics for selecting the optimal 3D-model of the antibody-glycan complex, out of thousands plausible options generated by automated docking and molecular dynamics simulation. STn-specificity was further validated by computationally screening of the selected antibody 3D-model against the human sialyl-Tn-glycome. This computational-experimental approach would allow rational design of potent antibodies targeting carbohydrates.