Antibody interface prediction with 3D Zernike descriptors and SVM

Structural trends in antibody-antigen binding interfaces: a computational analysis of 1833 experimentally determined 3D structures

Antibodies are attractive therapeutic candidates due to their ability to bind cognate antigens with high affinity and specificity. Still, the underlying molecular rules governing the antibody-antigen interface remain poorly understood, making in silico antibody design inherently difficult and keeping the discovery and design of novel antibodies a costly and laborious process. This study investigates the characteristics of antibody-antigen binding interfaces through a computational analysis of more than 850,000 atom-atom contacts from the largest reported set of antibody-antigen complexes with 1833 nonredundant, experimentally determined structures. The analysis compares binding characteristics of conventional antibodies and single-domain antibodies (sdAbs) targeting both protein- and peptide antigens. We find clear patterns in the number antibody-antigen contacts and amino acid frequencies in the paratope. The direct comparison of sdAbs and conventional antibodies helps elucidate the mechanisms employed by sdAbs to compensate for their smaller size and the fact that they harbor only half the number of complementarity-determining regions compared to conventional antibodies. Furthermore, we pinpoint antibody interface hotspot residues that are often found at the binding interface and the amino acid frequencies at these positions. These findings have direct potential applications in antibody engineering and the design of improved antibody libraries.

ZMPY3D: accelerating protein structure volume analysis through vectorized 3D Zernike moments and Python-based GPU integration

Motivation

Volumetric 3D object analyses are being applied in research fields such as structural bioinformatics, biophysics, and structural biology, with potential integration of artificial intelligence/machine learning (AI/ML) techniques. One such method, 3D Zernike moments, has proven valuable in analyzing protein structures (e.g., protein fold classification, protein-protein interaction analysis, and molecular dynamics simulations). Their compactness and efficiency make them amenable to large-scale analyses. Established methods for deriving 3D Zernike moments, however, can be inefficient, particularly when higher order terms are required, hindering broader applications. As the volume of experimental and computationally-predicted protein structure information continues to increase, structural biology has become a "big data" science requiring more efficient analysis tools.

Results

This application note presents a Python-based software package, ZMPY3D, to accelerate computation of 3D Zernike moments by vectorizing the mathematical formulae and using graphical processing units (GPUs). The package offers popular GPU-supported libraries such as CuPy and TensorFlow together with NumPy implementations, aiming to improve computational efficiency, adaptability, and flexibility in future algorithm development. The ZMPY3D package can be installed via PyPI, and the source code is available from GitHub. Volumetric-based protein 3D structural similarity scores and transform matrix of superposition functionalities have both been implemented, creating a powerful computational tool that will allow the research community to amalgamate 3D Zernike moments with existing AI/ML tools, to advance research and education in protein structure bioinformatics.

Availability and implementation

ZMPY3D, implemented in Python, is available on GitHub (https://github.com/tawssie/ZMPY3D) and PyPI, released under the GPL License.

The receiver operating characteristic curve accurately assesses imbalanced datasets

Many problems in biology require looking for a "needle in a haystack," corresponding to a binary classification where there are a few positives within a much larger set of negatives, which is referred to as a class imbalance. The receiver operating characteristic (ROC) curve and the associated area under the curve (AUC) have been reported as ill-suited to evaluate prediction performance on imbalanced problems where there is more interest in performance on the positive minority class, while the precision-recall (PR) curve is preferable. We show via simulation and a real case study that this is a misinterpretation of the difference between the ROC and PR spaces, showing that the ROC curve is robust to class imbalance, while the PR curve is highly sensitive to class imbalance. Furthermore, we show that class imbalance cannot be easily disentangled from classifier performance measured via PR-AUC.

Antibody design using deep learning: from sequence and structure design to affinity maturation

Deep learning has achieved impressive results in various fields such as computer vision and natural language processing, making it a powerful tool in biology. Its applications now encompass cellular image classification, genomic studies and drug discovery. While drug development traditionally focused deep learning applications on small molecules, recent innovations have incorporated it in the discovery and development of biological molecules, particularly antibodies. Researchers have devised novel techniques to streamline antibody development, combining in vitro and in silico methods. In particular, computational power expedites lead candidate generation, scaling and potential antibody development against complex antigens. This survey highlights significant advancements in protein design and optimization, specifically focusing on antibodies. This includes various aspects such as design, folding, antibody-antigen interactions docking and affinity maturation.

FP-Zernike: An Open-source Structural Database Construction Toolkit for Fast Structure Retrieval

The release of AlphaFold2 has sparked a rapid expansion in protein model databases. Efficient protein structure retrieval is crucial for the analysis of structure models, while measuring the similarity between structures is the key challenge in structural retrieval. Although existing structure alignment algorithms can address this challenge, they are often time-consuming. Currently, the state-of-the-art approach involves converting protein structures into three-dimensional (3D) Zernike descriptors and assessing similarity using Euclidean distance. However, the methods for computing 3D Zernike descriptors mainly rely on structural surfaces and are predominantly web-based, thus limiting their application in studying custom datasets. To overcome this limitation, we developed FP-Zernike, a user-friendly toolkit for computing different types of Zernike descriptors based on feature points. Users simply need to enter a single line of command to calculate the Zernike descriptors of all structures in customized datasets. FP-Zernike outperforms the leading method in terms of retrieval accuracy and binary classification accuracy across diverse benchmark datasets. In addition, we showed the application of FP-Zernike in the construction of the descriptor database and the protocol used for the Protein Data Bank (PDB) dataset to facilitate the local deployment of this tool for interested readers. Our demonstration contained 590,685 structures, and at this scale, our system required only 4-9 s to complete a retrieval. The experiments confirmed that it achieved the state-of-the-art accuracy level. FP-Zernike is an open-source toolkit, with the source code and related data accessible at https://ngdc.cncb.ac.cn/biocode/tools/BT007365/releases/0.1, as well as through a webserver at http://www.structbioinfo.cn/.

Revolutionizing Synthetic Antibody Design: Harnessing Artificial Intelligence and Deep Sequencing Big Data for Unprecedented Advances

Synthetic antibodies (Abs) represent a category of engineered proteins meticulously crafted to replicate the functions of their natural counterparts. Such Abs are generated in vitro, enabling advanced molecular alterations associated with antigen recognition, paratope site engineering, and biochemical refinements. In a parallel realm, deep sequencing has brought about a paradigm shift in molecular biology. It facilitates the prompt and cost-effective high-throughput sequencing of DNA and RNA molecules, enabling the comprehensive big data analysis of Ab transcriptomes, including specific regions of interest. Significantly, the integration of artificial intelligence (AI), based on machine- and deep- learning approaches, has fundamentally transformed our capacity to discern patterns hidden within deep sequencing big data, including distinctive Ab features and protein folding free energy landscapes. Ultimately, current AI advances can generate approximations of the most stable Ab structural configurations, enabling the prediction of de novo synthetic Abs. As a result, this manuscript comprehensively examines the latest and relevant literature concerning the intersection of deep sequencing big data and AI methodologies for the design and development of synthetic Abs. Together, these advancements have accelerated the exploration of antibody repertoires, contributing to the refinement of synthetic Ab engineering and optimizations, and facilitating advancements in the lead identification process.

Development and use of machine learning algorithms in vaccine target selection

Computer-aided discovery of vaccine targets has become a cornerstone of rational vaccine design. In this article, I discuss how Machine Learning (ML) can inform and guide key computational steps in rational vaccine design concerned with the identification of B and T cell epitopes and correlates of protection. I provide examples of ML models, as well as types of data and predictions for which they are built. I argue that interpretable ML has the potential to improve the identification of immunogens also as a tool for scientific discovery, by helping elucidate the molecular processes underlying vaccine-induced immune responses. I outline the limitations and challenges in terms of data availability and method development that need to be addressed to bridge the gap between advances in ML predictions and their translational application to vaccine design.

Design of protein-binding peptides with controlled binding affinity: the case of SARS-CoV-2 receptor binding domain and angiotensin-converting enzyme 2 derived peptides

The development of methods able to modulate the binding affinity between proteins and peptides is of paramount biotechnological interest in view of a vast range of applications that imply designed polypeptides capable to impair or favour Protein-Protein Interactions. Here, we applied a peptide design algorithm based on shape complementarity optimization and electrostatic compatibility and provided the first experimental in vitro proof of the efficacy of the design algorithm. Focusing on the interaction between the SARS-CoV-2 Spike Receptor-Binding Domain (RBD) and the human angiotensin-converting enzyme 2 (ACE2) receptor, we extracted a 23-residues long peptide that structurally mimics the major interacting portion of the ACE2 receptor and designed in silico five mutants of such a peptide with a modulated affinity. Remarkably, experimental KD measurements, conducted using biolayer interferometry, matched the in silico predictions. Moreover, we investigated the molecular determinants that govern the variation in binding affinity through molecular dynamics simulation, by identifying the mechanisms driving the different values of binding affinity at a single residue level. Finally, the peptide sequence with the highest affinity, in comparison with the wild type peptide, was expressed as a fusion protein with human H ferritin (HFt) 24-mer. Solution measurements performed on the latter constructs confirmed that peptides still exhibited the expected trend, thereby enhancing their efficacy in RBD binding. Altogether, these results indicate the high potentiality of this general method in developing potent high-affinity vectors for hindering/enhancing protein-protein associations.

Exploiting the Role of Features for Antigens-Antibodies Interaction Site Prediction

Antibodies are a class of proteins that recognize and neutralize pathogens by binding to their antigens. They are the most significant category of biopharmaceuticals for both diagnostic and therapeutic applications. Understanding how antibodies interact with their antigens plays a fundamental role in drug and vaccine design and helps to comprise the complex antigen binding mechanisms. Computational methods for predicting interaction sites of antibody-antigen are of great value due to the overall cost of experimental methods. Machine learning methods and deep learning techniques obtained promising results.In this work, we predict antibody interaction interface sites by applying HSS-PPI, a hybrid method defined to predict the interface sites of general proteins. The approach abstracts the proteins in terms of hierarchical representation and uses a graph convolutional network to classify the amino acids between interface and non-interface. Moreover, we also equipped the amino acids with different sets of physicochemical features together with structural ones to describe the residues. Analyzing the results, we observe that the structural features play a fundamental role in the amino acid descriptions. We compare the obtained performances, evaluated using standard metrics, with the ones obtained with SVM with 3D Zernike descriptors, Parapred, Paratome, and Antibody i-Patch.

ProteinMAE: masked autoencoder for protein surface self-supervised learning

SUMMARY

The biological functions of proteins are determined by the chemical and geometric properties of their surfaces. Recently, with the booming progress of deep learning, a series of learning-based surface descriptors have been proposed and achieved inspirational performance in many tasks such as protein design, protein-protein interaction prediction, etc. However, they are still limited by the problem of label scarcity, since the labels are typically obtained through wet experiments. Inspired by the great success of self-supervised learning in natural language processing and computer vision, we introduce ProteinMAE, a self-supervised framework specifically designed for protein surface representation to mitigate label scarcity. Specifically, we propose an efficient network and utilize a large number of accessible unlabeled protein data to pretrain it by self-supervised learning. Then we use the pretrained weights as initialization and fine-tune the network on downstream tasks. To demonstrate the effectiveness of our method, we conduct experiments on three different downstream tasks including binding site identification in protein surface, ligand-binding protein pocket classification, and protein-protein interaction prediction. The extensive experiments show that our method not only successfully improves the network's performance on all downstream tasks, but also achieves competitive performance with state-of-the-art methods. Moreover, our proposed network also exhibits significant advantages in terms of computational cost, which only requires less than a tenth of memory cost of previous methods.

AVAILABILITY AND IMPLEMENTATION

https://github.com/phdymz/ProteinMAE.

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

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

DeepBindPPI: Protein-Protein Binding Site Prediction Using Attention Based Graph Convolutional Network

Due to the importance of protein-protein interactions in defence mechanism of living body, attempts were made to investigate its attributes, including, but not limited to, binding affinity, and binding region. Contemporary strategies for binding site prediction largely resort to deep learning techniques but turned out to be low precision models. As laboratory experiments for drug discovery tasks utilize this information, increased false positives devalue the computational methods. This emphasize the need to develop enhanced strategies. DeepBindPPI employs deep learning technique to predict the binding regions of proteins, particularly antigen-antibody interaction sites. The results obtained are applied in a docking environment to confirm their correctness. An integration of graph convolutional network with attention mechanism predicts interacting amino acids with improved precision. The model learns the determining factors in interaction from a general pool of proteins and is then fine-tuned using antigen-antibody data. Comparison of the proposed method with existing techniques shows that the developed model has comparable performance. The use of a separate spatial network clearly improved the precision of the proposed method from 0.4 to 0.5. An attempt to utilize the interface information for docking using the HDOCK server gives promising results, with high-quality structures appearing in the top10 ranks.

Electrostatic complementarity at the interface drives transient protein-protein interactions

Understanding the mechanisms driving bio-molecules binding and determining the resulting complexes' stability is fundamental for the prediction of binding regions, which is the starting point for drug-ability and design. Characteristics like the preferentially hydrophobic composition of the binding interfaces, the role of van der Waals interactions, and the consequent shape complementarity between the interacting molecular surfaces are well established. However, no consensus has yet been reached on the role of electrostatic. Here, we perform extensive analyses on a large dataset of protein complexes for which both experimental binding affinity and pH data were available. Probing the amino acid composition, the disposition of the charges, and the electrostatic potential they generated on the protein molecular surfaces, we found that (i) although different classes of dimers do not present marked differences in the amino acid composition and charges disposition in the binding region, (ii) homodimers with identical binding region show higher electrostatic compatibility with respect to both homodimers with non-identical binding region and heterodimers. Interestingly, (iii) shape and electrostatic complementarity, for patches defined on short-range interactions, behave oppositely when one stratifies the complexes by their binding affinity: complexes with higher binding affinity present high values of shape complementarity (the role of the Lennard-Jones potential predominates) while electrostatic tends to be randomly distributed. Conversely, complexes with low values of binding affinity exploit Coulombic complementarity to acquire specificity, suggesting that electrostatic complementarity may play a greater role in transient (or less stable) complexes. In light of these results, (iv) we provide a novel, fast, and efficient method, based on the 2D Zernike polynomial formalism, to measure electrostatic complementarity without the need of knowing the complex structure. Expanding the electrostatic potential on a basis of 2D orthogonal polynomials, we can discriminate between transient and permanent protein complexes with an AUC of the ROC of [Formula: see text] 0.8. Ultimately, our work helps shedding light on the non-trivial relationship between the hydrophobic and electrostatic contributions in the binding interfaces, thus favoring the development of new predictive methods for binding affinity characterization.

GeoBind: segmentation of nucleic acid binding interface on protein surface with geometric deep learning

Unveiling the nucleic acid binding sites of a protein helps reveal its regulatory functions in vivo. Current methods encode protein sites from the handcrafted features of their local neighbors and recognize them via a classification, which are limited in expressive ability. Here, we present GeoBind, a geometric deep learning method for predicting nucleic binding sites on protein surface in a segmentation manner. GeoBind takes the whole point clouds of protein surface as input and learns the high-level representation based on the aggregation of their neighbors in local reference frames. Testing GeoBind on benchmark datasets, we demonstrate GeoBind is superior to state-of-the-art predictors. Specific case studies are performed to show the powerful ability of GeoBind to explore molecular surfaces when deciphering proteins with multimer formation. To show the versatility of GeoBind, we further extend GeoBind to five other types of ligand binding sites prediction tasks and achieve competitive performances.

Computational structural-based GPCR optimization for user-defined ligand: Implications for the development of biosensors

Organisms have developed effective mechanisms to sense the external environment. Human-designed biosensors exploit this natural optimization, where different biological machinery have been adapted to detect the presence of user-defined molecules. Specifically, the pheromone pathway in the model organism Saccharomyces cerevisiae represents a suitable candidate as a synthetic signaling system. Indeed, it expresses just one G-Protein Coupled Receptor (GPCR), Ste2, able to recognize pheromone and initiate the expression of pheromone-dependent genes. To date, the standard procedure to engineer this system relies on the substitution of the yeast GPCR with another one and on the modification of the yeast G-protein to bind the inserted receptor. Here, we propose an innovative computational procedure, based on geometrical and chemical optimization of protein binding pockets, to select the amino acid substitutions required to make the native yeast GPCR able to recognize a user-defined ligand. This procedure would allow the yeast to recognize a wide range of ligands, without a-priori knowledge about a GPCR recognizing them or the corresponding G protein. We used Monte Carlo simulations to design on Ste2 a binding pocket able to recognize epinephrine, selected as a test ligand. We validated Ste2 mutants via molecular docking and molecular dynamics. We verified that the amino acid substitutions we identified make Ste2 able to accommodate and remain firmly bound to epinephrine. Our results indicate that we sampled efficiently the huge space of possible mutants, proposing such a strategy as a promising starting point for the development of a new kind of S.cerevisiae-based biosensors.

Paragraph-antibody paratope prediction using graph neural networks with minimal feature vectors

SUMMARY

The development of new vaccines and antibody therapeutics typically takes several years and requires over $1bn in investment. Accurate knowledge of the paratope (antibody binding site) can speed up and reduce the cost of this process by improving our understanding of antibody-antigen binding. We present Paragraph, a structure-based paratope prediction tool that outperforms current state-of-the-art tools using simpler feature vectors and no antigen information.

AVAILABILITY AND IMPLEMENTATION

Source code is freely available at www.github.com/oxpig/Paragraph.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

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

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

Protein-protein interaction prediction methods: from docking-based to AI-based approaches

Protein-protein interactions (PPIs), such as protein-protein inhibitor, antibody-antigen complex, and supercomplexes play diverse and important roles in cells. Recent advances in structural analysis methods, including cryo-EM, for the determination of protein complex structures are remarkable. Nevertheless, much room remains for improvement and utilization of computational methods to predict PPIs because of the large number and great diversity of unresolved complex structures. This review introduces a wide array of computational methods, including our own, for estimating PPIs including antibody-antigen interactions, offering both historical and forward-looking perspectives.

Antibody variable region engineering for improving cancer immunotherapy

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

Shape Complementarity Optimization of Antibody–Antigen Interfaces: The Application to SARS-CoV-2 Spike Protein

Many factors influence biomolecule binding, and its assessment constitutes an elusive challenge in computational structural biology. In this aspect, the evaluation of shape complementarity at molecular interfaces is one of the main factors to be considered. We focus on the particular case of antibody–antigen complexes to quantify the complementarities occurring at molecular interfaces. We relied on a method we recently developed, which employs the 2D Zernike descriptors, to characterize the investigated regions with an ordered set of numbers summarizing the local shape properties. Collecting a structural dataset of antibody–antigen complexes, we applied this method and we statistically distinguished, in terms of shape complementarity, pairs of the interacting regions from the non-interacting ones. Thus, we set up a novel computational strategy based on in silico mutagenesis of antibody-binding site residues. We developed a Monte Carlo procedure to increase the shape complementarity between the antibody paratope and a given epitope on a target protein surface. We applied our protocol against several molecular targets in SARS-CoV-2 spike protein, known to be indispensable for viral cell invasion. We, therefore, optimized the shape of template antibodies for the interaction with such regions. As the last step of our procedure, we performed an independent molecular docking validation of the results of our Monte Carlo simulations.

A novel computational strategy for defining the minimal protein molecular surface representation

Most proteins perform their biological function by interacting with one or more molecular partners. In this respect, characterizing local features of the molecular surface, that can potentially be involved in the interaction with other molecules, represents a step forward in the investigation of the mechanisms of recognition and binding between molecules. Predictive methods often rely on extensive samplings of molecular patches with the aim to identify hot spots on the surface. In this framework, analysis of large proteins and/or many molecular dynamics frames is often unfeasible due to the high computational cost. Thus, finding optimal ways to reduce the number of points to be sampled maintaining the biological information (including the surface shape) carried by the molecular surface is pivotal. In this perspective, we here present a new theoretical and computational algorithm with the aim of defining a set of molecular surfaces composed of points not uniformly distributed in space, in such a way as to maximize the information of the overall shape of the molecule by minimizing the number of total points. We test our procedure’s ability in recognizing hot-spots by describing the local shape properties of portions of molecular surfaces through a recently developed method based on the formalism of 2D Zernike polynomials. The results of this work show the ability of the proposed algorithm to preserve the key information of the molecular surface using a reduced number of points compared to the complete surface, where all points of the surface are used for the description. In fact, the methodology shows a significant gain of the information stored in the sampling procedure compared to uniform random sampling.

Hierarchical representation for PPI sites prediction

Background

Protein–protein interactions have pivotal roles in life processes, and aberrant interactions are associated with various disorders. Interaction site identification is key for understanding disease mechanisms and design new drugs. Effective and efficient computational methods for the PPI prediction are of great value due to the overall cost of experimental methods. Promising results have been obtained using machine learning methods and deep learning techniques, but their effectiveness depends on protein representation and feature selection.

Results

We define a new abstraction of the protein structure, called hierarchical representations, considering and quantifying spatial and sequential neighboring among amino acids. We also investigate the effect of molecular abstractions using the Graph Convolutional Networks technique to classify amino acids as interface and no-interface ones. Our study takes into account three abstractions, hierarchical representations, contact map, and the residue sequence, and considers the eight functional classes of proteins extracted from the Protein–Protein Docking Benchmark 5.0. The performance of our method, evaluated using standard metrics, is compared to the ones obtained with some state-of-the-art protein interface predictors. The analysis of the performance values shows that our method outperforms the considered competitors when the considered molecules are structurally similar.

Conclusions

The hierarchical representation can capture the structural properties that promote the interactions and can be used to represent proteins with unknown structures by codifying only their sequential neighboring. Analyzing the results, we conclude that classes should be arranged according to their architectures rather than functions.

Comprehensive analysis of competitive endogenous RNAs network: Identification and validation of prediction model composed of mRNA signature and miRNA signature in gastric cancer

Gastric cancer (GC), one of the most lethal malignant tumors, is highly aggressive with a poor prognosis, while the molecular mechanisms underlying it remain largely unknown. Although advanced imaging techniques and comprehensive treatment facilitate the diagnosis and survival of some GC patients, the precise diagnosis and prognosis are still a challenge. The present study used publicly available gene expression profiles from The Cancer Genome Atlas and Gene Expression Omnibus datasets including mRNA, micro (mi)RNA and circular (circ)RNA of GC to establish a competing endogenous RNA network (ceRNA). Further, the present study performed least absolute shrinkage and selector operator regression analysis on the hub RNAs to establish a prediction model with mRNA and miRNA. The ceRNA network contained 109 edges and 56 nodes and the visible network contains 13 miRNAs, 9 circRNAs and 34 mRNAs. The five mRNA-based signature were CTF1, FKBP5, RNF128, GSTM2 and ADAMTS1. The area under curve (AUC) value of the diagnosis training cohort was 0.9975. The prognosis of the high-risk group (RiskScore >4.664) was worse compared with that of the low-risk group (RiskScore ≤4.664; P<0.05) in the training cohort. The five miRNA-based signature were miR-145-5p, miR-615-3p, miR-6507-5p, miR-937-3p and miR-99a-3p. The AUC value of the diagnosis training cohort was 0.9975. The prognosis of the high-risk group (RiskScore >1.621) was worse compared with that of the low-risk group (RiskScore ≤1.621; P<0.05) in the training cohort. The validation cohorts indicated that both five mRNA and five miRNA-based signatures had strong predictive power in diagnosis and prognosis for GC. In conclusion, a ceRNA network was established for GC and a five mRNA-based signature and a five miRNA-based signature was identified that enabled diagnosis and prognosis of GC by assigning patient to a high-risk group or low-risk group.

Machine learning in postgenomic biology and personalized medicine

In recent years Artificial Intelligence in the form of machine learning has been revolutionizing biology, biomedical sciences, and gene-based agricultural technology capabilities. Massive data generated in biological sciences by rapid and deep gene sequencing and protein or other molecular structure determination, on the one hand, requires data analysis capabilities using machine learning that are distinctly different from classical statistical methods; on the other, these large datasets are enabling the adoption of novel data-intensive machine learning algorithms for the solution of biological problems that until recently had relied on mechanistic model-based approaches that are computationally expensive. This review provides a bird's eye view of the applications of machine learning in post-genomic biology. Attempt is also made to indicate as far as possible the areas of research that are poised to make further impacts in these areas, including the importance of explainable artificial intelligence (XAI) in human health. Further contributions of machine learning are expected to transform medicine, public health, agricultural technology, as well as to provide invaluable gene-based guidance for the management of complex environments in this age of global warming.

Progress and challenges for the machine learning-based design of fit-for-purpose monoclonal antibodies

Although the therapeutic efficacy and commercial success of monoclonal antibodies (mAbs) are tremendous, the design and discovery of new candidates remain a time and cost-intensive endeavor. In this regard, progress in the generation of data describing antigen binding and developability, computational methodology, and artificial intelligence may pave the way for a new era of in silico on-demand immunotherapeutics design and discovery. Here, we argue that the main necessary machine learning (ML) components for an in silico mAb sequence generator are: understanding of the rules of mAb-antigen binding, capacity to modularly combine mAb design parameters, and algorithms for unconstrained parameter-driven in silico mAb sequence synthesis. We review the current progress toward the realization of these necessary components and discuss the challenges that must be overcome to allow the on-demand ML-based discovery and design of fit-for-purpose mAb therapeutic candidates.

Leveraging Sequential and Spatial Neighbors Information by Using CNNs Linked With GCNs for Paratope Prediction

Antibodies consisting of variable and constant regions, are a special type of proteins playing a vital role in immune system of the vertebrate. They have the remarkable ability to bind a large range of diverse antigens with extraordinary affinity and specificity. This malleability of binding makes antibodies an important class of biological drugs and biomarkers. In this article, we propose a method to identify which amino acid residues of an antibody directly interact with its associated antigen based on the features from sequence and structure. Our algorithm uses convolution neural networks (CNNs) linked with graph convolution networks (GCNs) to make use of information from both sequential and spatial neighbors to understand more about the local environment of target amino acid residue. Furthermore, we process the antigen partner of an antibody by employing an attention layer. Our method improves on the state-of-the-art methodology.

Comprehensive analysis of lectin-glycan interactions reveals determinants of lectin specificity

Lectin-glycan interactions facilitate inter- and intracellular communication in many processes including protein trafficking, host-pathogen recognition, and tumorigenesis promotion. Specific recognition of glycans by lectins is also the basis for a wide range of applications in areas including glycobiology research, cancer screening, and antiviral therapeutics. To provide a better understanding of the determinants of lectin-glycan interaction specificity and support such applications, this study comprehensively investigates specificity-conferring features of all available lectin-glycan complex structures. Systematic characterization, comparison, and predictive modeling of a set of 221 complementary physicochemical and geometric features representing these interactions highlighted specificity-conferring features with potential mechanistic insight. Univariable comparative analyses with weighted Wilcoxon-Mann-Whitney tests revealed strong statistical associations between binding site features and specificity that are conserved across unrelated lectin binding sites. Multivariable modeling with random forests demonstrated the utility of these features for predicting the identity of bound glycans based on generalized patterns learned from non-homologous lectins. These analyses revealed global determinants of lectin specificity, such as sialic acid glycan recognition in deep, concave binding sites enriched for positively charged residues, in contrast to high mannose glycan recognition in fairly shallow but well-defined pockets enriched for non-polar residues. Focused fine specificity analysis of hemagglutinin interactions with human-like and avian-like glycans uncovered features representing both known and novel mutations related to shifts in influenza tropism from avian to human tissues. As the approach presented here relies on co-crystallized lectin-glycan pairs for studying specificity, it is limited in its inferences by the quantity, quality, and diversity of the structural data available. Regardless, the systematic characterization of lectin binding sites presented here provides a novel approach to studying lectin specificity and is a step towards confidently predicting new lectin-glycan interactions.

Quantitative Description of Surface Complementarity of Antibody-Antigen Interfaces

Antibodies have the remarkable ability to recognise their cognate antigens with extraordinary affinity and specificity. Discerning the rules that define antibody-antigen recognition is a fundamental step in the rational design and engineering of functional antibodies with desired properties. In this study we apply the 3D Zernike formalism to the analysis of the surface properties of the antibody complementary determining regions (CDRs). Our results show that shape and electrostatic 3DZD descriptors of the surface of the CDRs are predictive of antigen specificity, with classification accuracy of 81% and area under the receiver operating characteristic curve (AUC) of 0.85. Additionally, while in terms of surface size, solvent accessibility and amino acid composition, antibody epitopes are typically not distinguishable from non-epitope, solvent-exposed regions of the antigen, the 3DZD descriptors detect significantly higher surface complementarity to the paratope, and are able to predict correct paratope-epitope interaction with an AUC = 0.75.

RAPID: A Rep-Seq Dataset Analysis Platform With an Integrated Antibody Database

The antibody repertoire is a critical component of the adaptive immune system and is believed to reflect an individual’s immune history and current immune status. Delineating the antibody repertoire has advanced our understanding of humoral immunity, facilitated antibody discovery, and showed great potential for improving the diagnosis and treatment of disease. However, no tool to date has effectively integrated big Rep-seq data and prior knowledge of functional antibodies to elucidate the remarkably diverse antibody repertoire. We developed a Rep-seq dataset Analysis Platform with an Integrated antibody Database (RAPID; https://rapid.zzhlab.org/), a free and web-based tool that allows researchers to process and analyse Rep-seq datasets. RAPID consolidates 521 WHO-recognized therapeutic antibodies, 88,059 antigen- or disease-specific antibodies, and 306 million clones extracted from 2,449 human IGH Rep-seq datasets generated from individuals with 29 different health conditions. RAPID also integrates a standardized Rep-seq dataset analysis pipeline to enable users to upload and analyse their datasets. In the process, users can also select set of existing repertoires for comparison. RAPID automatically annotates clones based on integrated therapeutic and known antibodies, and users can easily query antibodies or repertoires based on sequence or optional keywords. With its powerful analysis functions and rich set of antibody and antibody repertoire information, RAPID will benefit researchers in adaptive immune studies.

In-Silico Evidence for a Two Receptor Based Strategy of SARS-CoV-2

We propose a computational investigation on the interaction mechanisms between SARS-CoV-2 spike protein and possible human cell receptors. In particular, we make use of our newly developed numerical method able to determine efficiently and effectively the relationship of complementarity between portions of protein surfaces. This innovative and general procedure, based on the representation of the molecular isoelectronic density surface in terms of 2D Zernike polynomials, allows the rapid and quantitative assessment of the geometrical shape complementarity between interacting proteins, which was unfeasible with previous methods. Our results indicate that SARS-CoV-2 uses a dual strategy: in addition to the known interaction with angiotensin-converting enzyme 2, the viral spike protein can also interact with sialic-acid receptors of the cells in the upper airways.

OctSurf: Efficient hierarchical voxel-based molecular surface representation for protein-ligand affinity prediction

Voxel-based 3D convolutional neural networks (CNNs) have been applied to predict protein-ligand binding affinity. However, the memory usage and computation cost of these voxel-based approaches increase cubically with respect to spatial resolution and sometimes make volumetric CNNs intractable at higher resolutions. Therefore, it is necessary to develop memory-efficient alternatives that can accelerate the convolutional operation on 3D volumetric representations of the protein-ligand interaction. In this study, we implement a novel volumetric representation, OctSurf, to characterize the 3D molecular surface of protein binding pockets and bound ligands. The OctSurf surface representation is built based on the octree data structure, which has been widely used in computer graphics to efficiently represent and store 3D object data. Vanilla 3D-CNN approaches often divide the 3D space of objects into equal-sized voxels. In contrast, OctSurf recursively partitions the 3D space containing the protein-ligand pocket into eight subspaces called octants. Only those octants containing van der Waals surface points of protein or ligand atoms undergo the recursive subdivision process until they reach the predefined octree depth, whereas unoccupied octants are kept intact to reduce the memory cost. Resulting non-empty leaf octants approximate molecular surfaces of the protein pocket and bound ligands. These surface octants, along with their chemical and geometric features, are used as the input to 3D-CNNs. Two kinds of CNN architectures, VGG and ResNet, are applied to the OctSurf representation to predict binding affinity. The OctSurf representation consumes much less memory than the conventional voxel representation at the same resolution. By restricting the convolution operation to only octants of the smallest size, our method also alleviates the overall computational overhead of CNN. A series of experiments are performed to demonstrate the disk storage and computational efficiency of the proposed learning method. Our code is available at the following GitHub repository: https://github.uconn.edu/mldrugdiscovery/OctSurf.

Molecular Mechanisms Behind Anti SARS-CoV-2 Action of Lactoferrin

Despite the huge effort to contain the infection, the novel SARS-CoV-2 coronavirus has rapidly become pandemic, mainly due to its extremely high human-to-human transmission capability, and a surprisingly high viral charge of symptom-less people. While the seek for a vaccine is still ongoing, promising results have been obtained with antiviral compounds. In particular, lactoferrin is regarded to have beneficial effects both in preventing and soothing the infection. Here, we explore the possible molecular mechanisms with which lactoferrin interferes with SARS-CoV-2 cell invasion, preventing attachment and/or entry of the virus. To this aim, we search for possible interactions lactoferrin may have with virus structural proteins and host receptors. Representing the molecular iso-electron surface of proteins in terms of 2D-Zernike descriptors, we 1) identified putative regions on the lactoferrin surface able to bind sialic acid present on the host cell membrane, sheltering the cell from the virus attachment; 2) showed that no significant shape complementarity is present between lactoferrin and the ACE2 receptor, while 3) two high complementarity regions are found on the N- and C-terminal domains of the SARS-CoV-2 spike protein, hinting at a possible competition between lactoferrin and ACE2 for the binding to the spike protein.

Machine Learning Feature Selection for Predicting High Concentration Therapeutic Antibody Aggregation

Protein aggregation can hinder the development, safety and efficacy of therapeutic antibody-based drugs. Developing a predictive model that evaluates aggregation behaviors during early stage development is therefore desirable. Machine learning is a widely used tool to train models that predict data with different attributes. However, most machine learning techniques require more data than is typically available in antibody development. In this work, we describe a rational feature selection framework to develop accurate models with a small number of features. We applied this framework to predict aggregation behaviors of 21 approved monospecific monoclonal antibodies at high concentration (150 mg/mL), yielding a correlation coefficient of 0.71 on validation tests with only two features using a linear model. The nearest neighbors and support vector regression models further improved the performance, which have correlation coefficients of 0.86 and 0.80, respectively. This framework can be extended to train other models that predict different physical properties.

2D Zernike polynomial expansion: Finding the protein-protein binding regions

We present a method for efficiently and effectively assessing whether and where two proteins can interact with each other to form a complex. This is still largely an open problem, even for those relatively few cases where the 3D structure of both proteins is known. In fact, even if much of the information about the interaction is encoded in the chemical and geometric features of the structures, the set of possible contact patches and of their relative orientations are too large to be computationally affordable in a reasonable time, thus preventing the compilation of reliable interactome. Our method is able to rapidly and quantitatively measure the geometrical shape complementarity between interacting proteins, comparing their molecular iso-electron density surfaces expanding the surface patches in term of 2D Zernike polynomials. We first test the method against the real binding region of a large dataset of known protein complexes, reaching a success rate of 0.72. We then apply the method for the blind recognition of binding sites, identifying the real region of interaction in about 60% of the analyzed cases. Finally, we investigate how the efficiency in finding the right binding region depends on the surface roughness as a function of the expansion order.

Computational approaches to therapeutic antibody design: established methods and emerging trends

Antibodies are proteins that recognize the molecular surfaces of potentially noxious molecules to mount an adaptive immune response or, in the case of autoimmune diseases, molecules that are part of healthy cells and tissues. Due to their binding versatility, antibodies are currently the largest class of biotherapeutics, with five monoclonal antibodies ranked in the top 10 blockbuster drugs. Computational advances in protein modelling and design can have a tangible impact on antibody-based therapeutic development. Antibody-specific computational protocols currently benefit from an increasing volume of data provided by next generation sequencing and application to related drug modalities based on traditional antibodies, such as nanobodies. Here we present a structured overview of available databases, methods and emerging trends in computational antibody analysis and contextualize them towards the engineering of candidate antibody therapeutics.

Local and Global Rigidification Upon Antibody Affinity Maturation

During the affinity maturation process the immune system produces antibodies with higher specificity and activity through various rounds of somatic hypermutations in response to an antigen. Elucidating the affinity maturation process is fundamental in understanding immunity and in the development of biotherapeutics. Therefore, we analyzed 10 pairs of antibody fragments differing in their specificity and in distinct stages of affinity maturation using metadynamics in combination with molecular dynamics (MD) simulations. We investigated differences in flexibility of the CDR-H3 loop and global changes in plasticity upon affinity maturation. Among all antibody pairs we observed a substantial rigidification in flexibility and plasticity reflected in a substantial decrease of conformational diversity. To visualize and characterize these findings we used Markov-states models to reconstruct the kinetics of CDR-H3 loop dynamics and for the first time provide a method to define and localize surface plasticity upon affinity maturation.

Methods for sequence and structural analysis of B and T cell receptor repertoires

B cell receptors (BCRs) and T cell receptors (TCRs) make up an essential network of defense molecules that, collectively, can distinguish self from non-self and facilitate destruction of antigen-bearing cells such as pathogens or tumors. The analysis of BCR and TCR repertoires plays an important role in both basic immunology as well as in biotechnology. Because the repertoires are highly diverse, specialized software methods are needed to extract meaningful information from BCR and TCR sequence data. Here, we review recent developments in bioinformatics tools for analysis of BCR and TCR repertoires, with an emphasis on those that incorporate structural features. After describing the recent sequencing technologies for immune receptor repertoires, we survey structural modeling methods for BCR and TCRs, along with methods for clustering such models. We review downstream analyses, including BCR and TCR epitope prediction, antibody-antigen docking and TCR-peptide-MHC Modeling. We also briefly discuss molecular dynamics in this context.

mmCSM-AB: guiding rational antibody engineering through multiple point mutations

While antibodies are becoming an increasingly important therapeutic class, especially in personalized medicine, their development and optimization has been largely through experimental exploration. While there have been many efforts to develop computational tools to guide rational antibody engineering, most approaches are of limited accuracy when applied to antibody design, and have largely been limited to analysing a single point mutation at a time. To overcome this gap, we have curated a dataset of 242 experimentally determined changes in binding affinity upon multiple point mutations in antibody-target complexes (89 increasing and 153 decreasing binding affinity). Here, we have shown that by using our graph-based signatures and atomic interaction information, we can accurately analyse the consequence of multi-point mutations on antigen binding affinity. Our approach outperformed other available tools across cross-validation and two independent blind tests, achieving Pearson's correlations of up to 0.95. We have implemented our new approach, mmCSM-AB, as a web-server that can help guide the process of affinity maturation in antibody design. mmCSM-AB is freely available at http://biosig.unimelb.edu.au/mmcsm_ab/.

Learning context-aware structural representations to predict antigen and antibody binding interfaces

MOTIVATION

Understanding how antibodies specifically interact with their antigens can enable better drug and vaccine design, as well as provide insights into natural immunity. Experimental structural characterization can detail the 'ground truth' of antibody-antigen interactions, but computational methods are required to efficiently scale to large-scale studies. To increase prediction accuracy as well as to provide a means to gain new biological insights into these interactions, we have developed a unified deep learning-based framework to predict binding interfaces on both antibodies and antigens.

RESULTS

Our framework leverages three key aspects of antibody-antigen interactions to learn predictive structural representations: (i) since interfaces are formed from multiple residues in spatial proximity, we employ graph convolutions to aggregate properties across local regions in a protein; (ii) since interactions are specific between antibody-antigen pairs, we employ an attention layer to explicitly encode the context of the partner; (iii) since more data are available for general protein-protein interactions, we employ transfer learning to leverage this data as a prior for the specific case of antibody-antigen interactions. We show that this single framework achieves state-of-the-art performance at predicting binding interfaces on both antibodies and antigens, and that each of its three aspects drives additional improvement in the performance. We further show that the attention layer not only improves performance, but also provides a biologically interpretable perspective into the mode of interaction.

AVAILABILITY AND IMPLEMENTATION

The source code is freely available on github at https://github.com/vamships/PECAN.git.

How repertoire data are changing antibody science

Antibodies are vital proteins of the immune system that recognize potentially harmful molecules and initiate their removal. Mammals can efficiently create vast numbers of antibodies with different sequences capable of binding to any antigen with high affinity and specificity. Because they can be developed to bind to many disease agents, antibodies can be used as therapeutics. In an organism, after antigen exposure, antibodies specific to that antigen are enriched through clonal selection, expansion, and somatic hypermutation. The antibodies present in an organism therefore report on its immune status, describe its innate ability to deal with harmful substances, and reveal how it has previously responded. Next-generation sequencing technologies are being increasingly used to query the antibody, or B-cell receptor (BCR), sequence repertoire, and the amount of BCR data in public repositories is growing. The Observed Antibody Space database, for example, currently contains over a billion sequences from 68 different studies. Repertoires are available that represent both the naive state (i.e. antigen-inexperienced) and that after immunization. This wealth of data has created opportunities to learn more about our immune system. In this review, we discuss the many ways in which BCR repertoire data have been or could be exploited. We highlight its utility for providing insights into how the naive immune repertoire is generated and how it responds to antigens. We also consider how structural information can be used to enhance these data and may lead to more accurate depictions of the sequence space and to applications in the discovery of new therapeutics.