Updates to the Integrated Protein-Protein Interaction Benchmarks: Docking Benchmark Version 5 and Affinity Benchmark Version 2

Systematic investigation of machine learning on limited data: A study on predicting protein-protein binding strength

The application of machine learning techniques in biological research, especially when dealing with limited data availability, poses significant challenges. In this study, we leveraged advancements in method development for predicting protein-protein binding strength to conduct a systematic investigation into the application of machine learning on limited data. The binding strength, quantitatively measured as binding affinity, is vital for understanding the processes of recognition, association, and dysfunction that occur within protein complexes. By incorporating transfer learning, integrating domain knowledge, and employing both deep learning and traditional machine learning algorithms, we mitigated the impact of data limitations and made significant advancements in predicting protein-protein binding affinity. In particular, we developed over 20 models, ultimately selecting three representative best-performing ones that belong to distinct categories. The first model is structure-based, consisting of a random forest regression and thirteen handcrafted features. The second model is sequence-based, employing an architecture that combines transferred embedding features with a multilayer perceptron. Finally, we created an ensemble model by averaging the predictions of the two aforementioned models. The comparison with other predictors on three independent datasets confirms the significant improvements achieved by our models in predicting protein-protein binding affinity. The programs for running these three models are available at https://github.com/minghuilab/BindPPI.

EpiScan: accurate high-throughput mapping of antibody-specific epitopes using sequence information

The identification of antibody-specific epitopes on virus proteins is crucial for vaccine development and drug design. Nonetheless, traditional wet-lab approaches for the identification of epitopes are both costly and labor-intensive, underscoring the need for the development of efficient and cost-effective computational tools. Here, EpiScan, an attention-based deep learning framework for predicting antibody-specific epitopes, is presented. EpiScan adopts a multi-input and single-output strategy by designing independent blocks for different parts of antibodies, including variable heavy chain (VH), variable light chain (VL), complementary determining regions (CDRs), and framework regions (FRs). The block predictions are weighted and integrated for the prediction of potential epitopes. Using multiple experimental data samples, we show that EpiScan, which only uses antibody sequence information, can accurately map epitopes on specific antigen structures. The antibody-specific epitopes on the receptor binding domain (RBD) of SARS coronavirus 2 (SARS-CoV-2) were located by EpiScan, and the potentially valuable vaccine epitope was identified. EpiScan can expedite the epitope mapping process for high-throughput antibody sequencing data, supporting vaccine design and drug development. Availability: For the convenience of related wet-experimental researchers, the source code and web server of EpiScan are publicly available at https://github.com/gzBiomedical/EpiScan .

Quantitative Characterization of the Impact of Protein-Protein Interactions on Ligand-Protein Binding: A Multi-Chain Dynamics Perturbation Analysis Method

Many protein-protein interactions (PPIs) affect the ways in which small molecules bind to their constituent proteins, which can impact drug efficacy and regulatory mechanisms. While recent advances have improved our ability to independently predict both PPIs and ligand-protein interactions (LPIs), a comprehensive understanding of how PPIs affect LPIs is still lacking. Here, we examined 63 pairs of ligand-protein complexes in a benchmark dataset for protein-protein docking studies and quantified six typical effects of PPIs on LPIs. A multi-chain dynamics perturbation analysis method, called mcDPA, was developed to model these effects and used to predict small-molecule binding regions in protein-protein complexes. Our results illustrated that the mcDPA can capture the impact of PPI on LPI to varying degrees, with six similar changes in its predicted ligand-binding region. The calculations showed that 52% of the examined complexes had prediction accuracy at or above 50%, and 55% of the predictions had a recall of not less than 50%. When applied to 33 FDA-approved protein-protein-complex-targeting drugs, these numbers improved to 60% and 57% for the same accuracy and recall rates, respectively. The method developed in this study may help to design drug-target interactions in complex environments, such as in the case of protein-protein interactions.

Computational Methods to Predict Conformational B-Cell Epitopes

Accurate computational prediction of B-cell epitopes can greatly enhance biomedical research and rapidly advance efforts to develop therapeutics, monoclonal antibodies, vaccines, and immunodiagnostic reagents. Previous research efforts have primarily focused on the development of computational methods to predict linear epitopes rather than conformational epitopes; however, the latter is much more biologically predominant. Several conformational B-cell epitope prediction methods have recently been published, but their predictive performances are weak. Here, we present a review of the latest computational methods and assess their performances on a diverse test set of 29 non-redundant unbound antigen structures. Our results demonstrate that ISPIPab performs better than most methods and compares favorably with other recent antigen-specific methods. Finally, we suggest new strategies and opportunities to improve computational predictions of conformational B-cell epitopes.

Combining Complementarity and Binding Energetics in the Assessment of Protein Interactions: EnCPdock-A Practical Manual

The combined effect of shape and electrostatic complementarities (Sc, EC) at the interface of the interacting protein partners (PPI) serves as the physical basis for such associations and is a strong determinant of their binding energetics. EnCPdock (https://www.scinetmol.in/EnCPdock/) presents a comprehensive web platform for the direct conjoint comparative analyses of complementarity and binding energetics in PPIs. It elegantly interlinks the dual nature of local (Sc) and nonlocal complementarity (EC) in PPIs using the complementarity plot. It further derives an AI-based ΔGbinding with a prediction accuracy comparable to the state of the art. This book chapter presents a practical manual to conceptualize and implement EnCPdock with its various features and functionalities, collectively having the potential to serve as a valuable protein engineering tool in the design of novel protein interfaces.

Context-aware geometric deep learning for protein sequence design

Protein design and engineering are evolving at an unprecedented pace leveraging the advances in deep learning. Current models nonetheless cannot natively consider non-protein entities within the design process. Here, we introduce a deep learning approach based solely on a geometric transformer of atomic coordinates and element names that predicts protein sequences from backbone scaffolds aware of the restraints imposed by diverse molecular environments. To validate the method, we show that it can produce highly thermostable, catalytically active enzymes with high success rates. This concept is anticipated to improve the versatility of protein design pipelines for crafting desired functions.

Advancing membrane-associated protein docking with improved sampling and scoring in Rosetta

The oligomerization of protein macromolecules on cell membranes plays a fundamental role in regulating cellular function. From modulating signal transduction to directing immune response, membrane proteins (MPs) play a crucial role in biological processes and are often the target of many pharmaceutical drugs. Despite their biological relevance, the challenges in experimental determination have hampered the structural availability of membrane proteins and their complexes. Computational docking provides a promising alternative to model membrane protein complex structures. Here, we present Rosetta-MPDock, a flexible transmembrane (TM) protein docking protocol that captures binding-induced conformational changes. Rosetta-MPDock samples large conformational ensembles of flexible monomers and docks them within an implicit membrane environment. We benchmarked this method on 29 TM-protein complexes of variable backbone flexibility. These complexes are classified based on the root-mean-square deviation between the unbound and bound states (RMSDUB) as: rigid (RMSDUB <1.2 Å), moderately-flexible (RMSDUB ∈ [1.2, 2.2) Å), and flexible targets (RMSDUB > 2.2 Å). In a local docking scenario, i.e. with membrane protein partners starting ≈10 Å apart embedded in the membrane in their unbound conformations, Rosetta-MPDock successfully predicts the correct interface (success defined as achieving 3 near-native structures in the 5 top-ranked models) for 67% moderately flexible targets and 60% of the highly flexible targets, a substantial improvement from the existing membrane protein docking methods. Further, by integrating AlphaFold2-multimer for structure determination and using Rosetta-MPDock for docking and refinement, we demonstrate improved success rates over the benchmark targets from 64% to 73%. Rosetta-MPDock advances the capabilities for membrane protein complex structure prediction and modeling to tackle key biological questions and elucidate functional mechanisms in the membrane environment. The benchmark set and the code is available for public use at github.com/Graylab/MPDock.

Geometric epitope and paratope prediction

MOTIVATION

Identifying the binding sites of antibodies is essential for developing vaccines and synthetic antibodies. In this article, we investigate the optimal representation for predicting the binding sites in the two molecules and emphasize the importance of geometric information.

RESULTS

Specifically, we compare different geometric deep learning methods applied to proteins' inner (I-GEP) and outer (O-GEP) structures. We incorporate 3D coordinates and spectral geometric descriptors as input features to fully leverage the geometric information. Our research suggests that different geometrical representation information is useful for different tasks. Surface-based models are more efficient in predicting the binding of the epitope, while graph models are better in paratope prediction, both achieving significant performance improvements. Moreover, we analyze the impact of structural changes in antibodies and antigens resulting from conformational rearrangements or reconstruction errors. Through this investigation, we showcase the robustness of geometric deep learning methods and spectral geometric descriptors to such perturbations.

AVAILABILITY AND IMPLEMENTATION

The python code for the models, together with the data and the processing pipeline, is open-source and available at https://github.com/Marco-Peg/GEP.

Quantum Chemistry-Based Protein-Protein Docking without Empirical Parameters

This study developed a novel protein-protein docking approach based on quantum chemistry. To judge the appropriateness of complex structures, we introduced two criterion values, EV1 and EV2, computed using the fragment molecular orbital method without any empirical parameters. These criterion values enable us to search complex structures in which patterns of the electrostatic potential of the two proteins are optimally aligned at their interface. The performance of our method was validated using 53 complexes in a benchmark set provided for protein-protein docking. When employing bound state structures, docking success rates reached 64% for EV1 and 76% for EV2. On the other hand, when employing unbound state structures, docking success rates reached 13% for EV1 and 17% for EV2.

Review and Comparative Analysis of Methods and Advancements in Predicting Protein Complex Structure

Protein complexes perform diverse biological functions, and obtaining their three-dimensional structure is critical to understanding and grasping their functions. In many cases, it's not just two proteins interacting to form a dimer; instead, multiple proteins interact to form a multimer. Experimentally resolving protein complex structures can be quite challenging. Recently, there have been efforts and methods that build upon prior predictions of dimer structures to attempt to predict multimer structures. However, in comparison to monomeric protein structure prediction, the accuracy of protein complex structure prediction remains relatively low. This paper provides an overview of recent advancements in efficient computational models for predicting protein complex structures. We introduce protein-protein docking methods in detail and summarize their main ideas, applicable modes, and related information. To enhance prediction accuracy, other critical protein-related information is also integrated, such as predicting interchain residue contact, utilizing experimental data like cryo-EM experiments, and considering protein interactions and non-interactions. In addition, we comprehensively review computational approaches for end-to-end prediction of protein complex structures based on artificial intelligence (AI) technology and describe commonly used datasets and representative evaluation metrics in protein complexes. Finally, we analyze the formidable challenges faced in current protein complex structure prediction tasks, including the structure prediction of heteromeric complex, disordered regions in complex, antibody-antigen complex, and RNA-related complex, as well as the evaluation metrics for complex assessment. We hope that this work will provide comprehensive knowledge of complex structure predictions to contribute to future advanced predictions.

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.

A Survey of Deep Learning Methods for Estimating the Accuracy of Protein Quaternary Structure Models

The quality prediction of quaternary structure models of a protein complex, in the absence of its true structure, is known as the Estimation of Model Accuracy (EMA). EMA is useful for ranking predicted protein complex structures and using them appropriately in biomedical research, such as protein-protein interaction studies, protein design, and drug discovery. With the advent of more accurate protein complex (multimer) prediction tools, such as AlphaFold2-Multimer and ESMFold, the estimation of the accuracy of protein complex structures has attracted increasing attention. Many deep learning methods have been developed to tackle this problem; however, there is a noticeable absence of a comprehensive overview of these methods to facilitate future development. Addressing this gap, we present a review of deep learning EMA methods for protein complex structures developed in the past several years, analyzing their methodologies, data and feature construction. We also provide a prospective summary of some potential new developments for further improving the accuracy of the EMA methods.

Database-Driven Identification of Structurally Similar Protein-Protein Interfaces

Analyzing the similarity of protein interfaces in protein-protein interactions gives new insights into protein function and assists in discovering new drugs. Usually, tools that assess the similarity focus on the interactions between two protein interfaces, while sometimes we only have one predicted interface. Herein, we present PiMine, a database-driven protein interface similarity search. It compares interface residues of one or two interacting chains by calculating and searching tetrahedral geometric patterns of α-carbon atoms and calculating physicochemical and shape-based similarity. On a dedicated, tailor-made dataset, we show that PiMine outperforms commonly used comparison tools in terms of early enrichment when considering interfaces of sequentially and structurally unrelated proteins. In an application example, we demonstrate its usability for protein interaction partner prediction by comparing predicted interfaces to known protein-protein interfaces.

ProInterVal: Validation of Protein-Protein Interfaces through Learned Interface Representations

Proteins are vital components of the biological world and serve a multitude of functions. They interact with other molecules through their interfaces and participate in crucial cellular processes. Disruption of these interactions can have negative effects on organisms, highlighting the importance of studying protein-protein interfaces for developing targeted therapies for diseases. Therefore, the development of a reliable method for investigating protein-protein interactions is of paramount importance. In this work, we present an approach for validating protein-protein interfaces using learned interface representations. The approach involves using a graph-based contrastive autoencoder architecture and a transformer to learn representations of protein-protein interaction interfaces from unlabeled data and then validating them through learned representations with a graph neural network. Our method achieves an accuracy of 0.91 for the test set, outperforming existing GNN-based methods. We demonstrate the effectiveness of our approach on a benchmark data set and show that it provides a promising solution for validating protein-protein interfaces.

A comprehensive review of protein-centric predictors for biomolecular interactions: from proteins to nucleic acids and beyond

Proteins interact with diverse ligands to perform a large number of biological functions, such as gene expression and signal transduction. Accurate identification of these protein-ligand interactions is crucial to the understanding of molecular mechanisms and the development of new drugs. However, traditional biological experiments are time-consuming and expensive. With the development of high-throughput technologies, an increasing amount of protein data is available. In the past decades, many computational methods have been developed to predict protein-ligand interactions. Here, we review a comprehensive set of over 160 protein-ligand interaction predictors, which cover protein-protein, protein-nucleic acid, protein-peptide and protein-other ligands (nucleotide, heme, ion) interactions. We have carried out a comprehensive analysis of the above four types of predictors from several significant perspectives, including their inputs, feature profiles, models, availability, etc. The current methods primarily rely on protein sequences, especially utilizing evolutionary information. The significant improvement in predictions is attributed to deep learning methods. Additionally, sequence-based pretrained models and structure-based approaches are emerging as new trends.

In silico models of the macromolecular NaV1.5-KIR2.1 complex

In cardiac cells, the expression of the cardiac voltage-gated Na+ channel (NaV1.5) is reciprocally regulated with the inward rectifying K+ channel (KIR2.1). These channels can form macromolecular complexes that pre-assemble early during forward trafficking (transport to the cell membrane). In this study, we present in silico 3D models of NaV1.5-KIR2.1, generated by rigid-body protein-protein docking programs and deep learning-based AlphaFold-Multimer software. Modeling revealed that the two channels could physically interact with each other along the entire transmembrane region. Structural mapping of disease-associated mutations revealed a hotspot at this interface with several trafficking-deficient variants in close proximity. Thus, examining the role of disease-causing variants is important not only in isolated channels but also in the context of macromolecular complexes. These findings may contribute to a better understanding of the life-threatening cardiovascular diseases underlying KIR2.1 and NaV1.5 malfunctions.

Flexible protein-protein docking with a multitrack iterative transformer

Conventional protein-protein docking algorithms usually rely on heavy candidate sampling and reranking, but these steps are time-consuming and hinder applications that require high-throughput complex structure prediction, for example, structure-based virtual screening. Existing deep learning methods for protein-protein docking, despite being much faster, suffer from low docking success rates. In addition, they simplify the problem to assume no conformational changes within any protein upon binding (rigid docking). This assumption precludes applications when binding-induced conformational changes play a role, such as allosteric inhibition or docking from uncertain unbound model structures. To address these limitations, we present GeoDock, a multitrack iterative transformer network to predict a docked structure from separate docking partners. Unlike deep learning models for protein structure prediction that input multiple sequence alignments, GeoDock inputs just the sequences and structures of the docking partners, which suits the tasks when the individual structures are given. GeoDock is flexible at the protein residue level, allowing the prediction of conformational changes upon binding. On the Database of Interacting Protein Structures (DIPS) test set, GeoDock achieves a 43% top-1 success rate, outperforming all other tested methods. However, in the standard DIPS train/test splits, we discovered contamination of close homologs in the training set. After decontaminating the training set, the success rate is 31%. On the DB5.5 test set and a benchmark dataset of antibody-antigen complexes, GeoDock outperforms the deep learning models trained using the same dataset but falls behind most of the conventional methods and AlphaFold-Multimer. GeoDock attains an average inference speed of under 1 s on a single GPU, enabling its application in large-scale structure screening. Although binding-induced conformational changes are still a challenge owing to limited training and evaluation data, our architecture sets up the foundation to capture this backbone flexibility. Code and a demonstration Jupyter notebook are available at https://github.com/Graylab/GeoDock.

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.

ARCTIC-3D: automatic retrieval and clustering of interfaces in complexes from 3D structural information

The formation of a stable complex between proteins lies at the core of a wide variety of biological processes and has been the focus of countless experiments. The huge amount of information contained in the protein structural interactome in the Protein Data Bank can now be used to characterise and classify the existing biological interfaces. We here introduce ARCTIC-3D, a fast and user-friendly data mining and clustering software to retrieve data and rationalise the interface information associated with the protein input data. We demonstrate its use by various examples ranging from showing the increased interaction complexity of eukaryotic proteins, 20% of which on average have more than 3 different interfaces compared to only 10% for prokaryotes, to associating different functions to different interfaces. In the context of modelling biomolecular assemblies, we introduce the concept of "recognition entropy", related to the number of possible interfaces of the components of a protein-protein complex, which we demonstrate to correlate with the modelling difficulty in classical docking approaches. The identified interface clusters can also be used to generate various combinations of interface-specific restraints for integrative modelling. The ARCTIC-3D software is freely available at github.com/haddocking/arctic3d and can be accessed as a web-service at wenmr.science.uu.nl/arctic3d.

DeepRank-GNN-esm: a graph neural network for scoring protein-protein models using protein language model

Motivation

Protein-Protein interactions (PPIs) play critical roles in numerous cellular processes. By modelling the 3D structures of the correspond protein complexes valuable insights can be obtained, providing, e.g. starting points for drug and protein design. One challenge in the modelling process is however the identification of near-native models from the large pool of generated models. To this end we have previously developed DeepRank-GNN, a graph neural network that integrates structural and sequence information to enable effective pattern learning at PPI interfaces. Its main features are related to the Position Specific Scoring Matrices (PSSMs), which are computationally expensive to generate, significantly limits the algorithm's usability.

Results

We introduce here DeepRank-GNN-esm that includes as additional features protein language model embeddings from the ESM-2 model. We show that the ESM-2 embeddings can actually replace the PSSM features at no cost in-, or even better performance on two PPI-related tasks: scoring docking poses and detecting crystal artifacts. This new DeepRank version bypasses thus the need of generating PSSM, greatly improving the usability of the software and opening new application opportunities for systems for which PSSM profiles cannot be obtained or are irrelevant (e.g. antibody-antigen complexes).

Availability and implementation

DeepRank-GNN-esm is freely available from https://github.com/DeepRank/DeepRank-GNN-esm.

Docking Foundations: From Rigid to Flexible Docking

Despite the development of methods for the experimental determination of protein structures, the dissonance between the number of known sequences and their solved structures is still enormous. This is particularly evident in protein-protein complexes. To fill this gap, diverse technologies have been developed to study protein-protein interactions (PPIs) in a cellular context including a range of biological and computational methods. The latter derive from techniques originally published and applied almost half a century ago and are based on interdisciplinary knowledge from the nexus of the fields of biology, chemistry, and physics about protein sequences, structures, and their folding. Protein-protein docking, the main protagonist of this chapter, is routinely treated as an integral part of protein research. Herein, we describe the basic foundations of the whole process in general terms, but step by step from protein representations through docking methods and evaluation of complexes to their final validation.

Challenges of Protein-Protein Docking of the Membrane Proteins

Despite the recent advances in the determination of high-resolution membrane protein (MP) structures, the structural and functional characterization of MPs remains extremely challenging, mainly due to the hydrophobic nature, low abundance, poor expression, purification, and crystallization difficulties associated with MPs. Whereby the major challenges/hurdles for MP structure determination are associated with the expression, purification, and crystallization procedures. Although there have been significant advances in the experimental determination of MP structures, only a limited number of MP structures (approximately less than 1% of all) are available in the Protein Data Bank (PDB). Therefore, the structures of a large number of MPs still remain unresolved, which leads to the availability of widely unplumbed structural and functional information related to MPs. As a result, recent developments in the drug discovery realm and the significant biological contemplation have led to the development of several novel, low-cost, and time-efficient computational methods that overcome the limitations of experimental approaches, supplement experiments, and provide alternatives for the characterization of MPs. Whereby the fine tuning and optimizations of these computational approaches remains an ongoing endeavor.Computational methods offer a potential way for the elucidation of structural features and the augmentation of currently available MP information. However, the use of computational modeling can be extremely challenging for MPs mainly due to insufficient knowledge of (or gaps in) atomic structures of MPs. Despite the availability of numerous in silico methods for 3D structure determination the applicability of these methods to MPs remains relatively low since all methods are not well-suited or adequate for MPs. However, sophisticated methods for MP structure predictions are constantly being developed and updated to integrate the modifications required for MPs. Currently, different computational methods for (1) MP structure prediction, (2) stability analysis of MPs through molecular dynamics simulations, (3) modeling of MP complexes through docking, (4) prediction of interactions between MPs, and (5) MP interactions with its soluble partner are extensively used. Towards this end, MP docking is widely used. It is notable that the MP docking methods yet few in number might show greater potential in terms of filling the knowledge gap. In this chapter, MP docking methods and associated challenges have been reviewed to improve the applicability, accuracy, and the ability to model macromolecular complexes.

Software for Predicting Binding Free Energy of Protein-Protein Complexes and Their Mutants

Protein-protein binding affinity prediction is important for understanding complex biochemical pathways and to uncover protein interaction networks. Quantitative estimation of the binding affinity changes caused by mutations can provide critical information for protein function annotation and genetic disease diagnoses. The binding free energies of protein-protein complexes can be predicted using several computational tools. This chapter is a summary of software developed for the prediction of binding free energies for protein-protein complexes and their mutants.

Evaluation of AlphaFold antibody-antigen modeling with implications for improving predictive accuracy

High resolution antibody-antigen structures provide critical insights into immune recognition and can inform therapeutic design. The challenges of experimental structural determination and the diversity of the immune repertoire underscore the necessity of accurate computational tools for modeling antibody-antigen complexes. Initial benchmarking showed that despite overall success in modeling protein-protein complexes, AlphaFold and AlphaFold-Multimer have limited success in modeling antibody-antigen interactions. In this study, we performed a thorough analysis of AlphaFold's antibody-antigen modeling performance on 427 nonredundant antibody-antigen complex structures, identifying useful confidence metrics for predicting model quality, and features of complexes associated with improved modeling success. Notably, we found that the latest version of AlphaFold improves near-native modeling success to over 30%, versus approximately 20% for a previous version, while increased AlphaFold sampling gives approximately 50% success. With this improved success, AlphaFold can generate accurate antibody-antigen models in many cases, while additional training or other optimization may further improve performance.

ComDock: A novel approach for protein-protein docking with an efficient fusing strategy

Protein-protein interaction plays an important role in studying the mechanism of protein functions from the structural perspective. Molecular docking is a powerful approach to detect protein-protein complexes using computational tools, due to the high cost and time-consuming of the traditional experimental methods. Among existing technologies, the template-based method utilizes the structural information of known homologous 3D complexes as available and reliable templates to achieve high accuracy and low computational complexity. However, the performance of the template-based method depends on the quality and quantity of templates. When insufficient or even no templates, the ab initio docking method is necessary and largely enriches the docking conformations. Therefore, it's a feasible strategy to fuse the effectivity of the template-based model and the universality of ab initio model to improve the docking performance. In this study, we construct a new, diverse, comprehensive template library derived from PDB, containing 77,685 complexes. We propose a template-based method (named TemDock), which retrieves the evolutionary relationship between the target sequence and samples in the template library and transfers similar structural information. Then, the target structure is built by superposing on the homologous template complex with TM-align. Moreover, we develop a consensus-based method (named ComDock) to integrate our TemDock and an existing ab initio method (ZDOCK). On 105 targets with templates from Benchmark 5.0, the TemDock and ComDock achieve a success rate of 68.57 % and 71.43 % in the top 10 conformations, respectively. Compared with the HDOCK, ComDock obtains better I-RMSD of hit configurations on 9 targets and more hit models in the top 100 conformations. As an efficient method for protein-protein docking, the ComDock is expected to study protein-protein recognition and reveal the various biological passways that are critical for developing drug discovery. The final results are stored at https://github.com/guofei-tju/mqz_ComDock_docking.

In-silico analysis of potent Mosquirix vaccine adjuvant leads

BACKGROUND

World Health Organization recommend the use of malaria vaccine, Mosquirix, as a malaria prevention strategy. However, Mosquirix has failed to reduce the global burden of malaria because of its inefficacy. The Mosquirix vaccine's modest effectiveness against malaria, 36% among kids aged 5 to 17 months who need at least four doses, fails to aid malaria eradication. Therefore, highly effective and efficacious malaria vaccines are required. The well-characterized P. falciparum circumsporozoite surface protein can be used to discover adjuvants that can increase the efficacy of Mosquirix. Therefore, the study sought to undertake an in-silico discovery of Plasmodium falciparum circumsporozoite surface protein inhibitors with pharmacological properties on Mosquirix using hierarchical virtual screening and molecular dynamics simulation.

RESULTS

Monoclonal antibody L9, an anti-Plasmodium falciparum circumsporozoite surface protein molecule, was used to identify Plasmodium falciparum circumsporozoite surface protein inhibitors with pharmacological properties on Mosquirix during a virtual screening process in ZINCPHARMER that yielded 23 hits. After drug-likeness and absorption, distribution, metabolism, excretion, and toxicity property analysis in the SwissADME web server, only 9 of the 23 hits satisfied the requirements. The 9 compounds were docked with Plasmodium falciparum circumsporozoite surface protein using the PyRx software to understand their interactions. ZINC25374360 (-8.1 kcal/mol), ZINC40144754 (-8.3 kcal/mol), and ZINC71996727 (-8.9 kcal/mol) bound strongly to Plasmodium falciparum circumsporozoite surface protein with binding affinities of less than -8.0 kcal/mol. The stability of these molecularly docked Plasmodium falciparum circumsporozoite surface protein-inhibitor complexes were assessed through molecular dynamics simulation using GROMACS 2022. ZINC25374360 and ZINC71996727 formed stable complexes with Plasmodium falciparum circumsporozoite surface protein. They were subjected to in vitro validation for their inhibitory potential. The IC50 values ranging between 250 and 350 ng/ml suggest inhibition of parasite development.

CONCLUSION

Therefore, the two Plasmodium falciparum circumsporozoite surface protein inhibitors can be used as vaccine adjuvants to increase the efficacy of the existing Mosquirix vaccine. Nevertheless, additional in vivo tests, structural optimization studies, and homogenization analysis are essential to determine the anti-plasmodial action of these adjuvants in humans.

Reliable protein-protein docking with AlphaFold, Rosetta and replica-exchange

Despite the recent breakthrough of AlphaFold (AF) in the field of protein sequence-to-structure prediction, modeling protein interfaces and predicting protein complex structures remains challenging, especially when there is a significant conformational change in one or both binding partners. Prior studies have demonstrated that AF-multimer (AFm) can predict accurate protein complexes in only up to 43% of cases. In this work, we combine AlphaFold as a structural template generator with a physics-based replica exchange docking algorithm. Using a curated collection of 254 available protein targets with both unbound and bound structures, we first demonstrate that AlphaFold confidence measures (pLDDT) can be repurposed for estimating protein flexibility and docking accuracy for multimers. We incorporate these metrics within our ReplicaDock 2.0 protocol to complete a robust in-silico pipeline for accurate protein complex structure prediction. AlphaRED (AlphaFold-initiated Replica Exchange Docking) successfully docks failed AF predictions including 97 failure cases in Docking Benchmark Set 5.5. AlphaRED generates CAPRI acceptable-quality or better predictions for 66% of benchmark targets. Further, on a subset of antigen-antibody targets, which is challenging for AFm (19% success rate), AlphaRED demonstrates a success rate of 51%. This new strategy demonstrates the success possible by integrating deep-learning based architectures trained on evolutionary information with physics-based enhanced sampling. The pipeline is available at github.com/Graylab/AlphaRED.

Quantification of biases in predictions of protein-protein binding affinity changes upon mutations

Understanding the impact of mutations on protein-protein binding affinity is a key objective for a wide range of biotechnological applications and for shedding light on disease-causing mutations, which are often located at protein-protein interfaces. Over the past decade, many computational methods using physics-based and/or machine learning approaches have been developed to predict how protein binding affinity changes upon mutations. They all claim to achieve astonishing accuracy on both training and test sets, with performances on standard benchmarks such as SKEMPI 2.0 that seem overly optimistic. Here we benchmarked eight well-known and well-used predictors and identified their biases and dataset dependencies, using not only SKEMPI 2.0 as a test set but also deep mutagenesis data on the severe acute respiratory syndrome coronavirus 2 spike protein in complex with the human angiotensin-converting enzyme 2. We showed that, even though most of the tested methods reach a significant degree of robustness and accuracy, they suffer from limited generalizability properties and struggle to predict unseen mutations. Interestingly, the generalizability problems are more severe for pure machine learning approaches, while physics-based methods are less affected by this issue. Moreover, undesirable prediction biases toward specific mutation properties, the most marked being toward destabilizing mutations, are also observed and should be carefully considered by method developers. We conclude from our analyses that there is room for improvement in the prediction models and suggest ways to check, assess and improve their generalizability and robustness.

Predictive Modeling and Structure Analysis of Genetic Variants in Familial Hypercholesterolemia: Implications for Diagnosis and Protein Interaction Studies

PURPOSE OF REVIEW

Familial hypercholesterolemia (FH) is a hereditary condition characterized by elevated levels of low-density lipoprotein cholesterol (LDL-C), which increases the risk of cardiovascular disease if left untreated. This review aims to discuss the role of bioinformatics tools in evaluating the pathogenicity of missense variants associated with FH. Specifically, it highlights the use of predictive models based on protein sequence, structure, evolutionary conservation, and other relevant features in identifying genetic variants within LDLR, APOB, and PCSK9 genes that contribute to FH.

RECENT FINDINGS

In recent years, various bioinformatics tools have emerged as valuable resources for analyzing missense variants in FH-related genes. Tools such as REVEL, Varity, and CADD use diverse computational approaches to predict the impact of genetic variants on protein function. These tools consider factors such as sequence conservation, structural alterations, and receptor binding to aid in interpreting the pathogenicity of identified missense variants. While these predictive models offer valuable insights, the accuracy of predictions can vary, especially for proteins with unique characteristics that might not be well represented in the databases used for training. This review emphasizes the significance of utilizing bioinformatics tools for assessing the pathogenicity of FH-associated missense variants. Despite their contributions, a definitive diagnosis of a genetic variant necessitates functional validation through in vitro characterization or cascade screening. This step ensures the precise identification of FH-related variants, leading to more accurate diagnoses. Integrating genetic data with reliable bioinformatics predictions and functional validation can enhance our understanding of the genetic basis of FH, enabling improved diagnosis, risk stratification, and personalized treatment for affected individuals. The comprehensive approach outlined in this review promises to advance the management of this inherited disorder, potentially leading to better health outcomes for those affected by FH.

Topological links in predicted protein complex structures reveal limitations of AlphaFold

AlphaFold is making great progress in protein structure prediction, not only for single-chain proteins but also for multi-chain protein complexes. When using AlphaFold-Multimer to predict protein‒protein complexes, we observed some unusual structures in which chains are looped around each other to form topologically intertwining links at the interface. Based on physical principles, such topological links should generally not exist in native protein complex structures unless covalent modifications of residues are involved. Although it is well known and has been well studied that protein structures may have topologically complex shapes such as knots and links, existing methods are hampered by the chain closure problem and show poor performance in identifying topologically linked structures in protein‒protein complexes. Therefore, we address the chain closure problem by using sliding windows from a local perspective and propose an algorithm to measure the topological-geometric features that can be used to identify topologically linked structures. An application of the method to AlphaFold-Multimer-predicted protein complex structures finds that approximately 1.72% of the predicted structures contain topological links. The method presented in this work will facilitate the computational study of protein‒protein interactions and help further improve the structural prediction of multi-chain protein complexes.

CSM-Potential2: A comprehensive deep learning platform for the analysis of protein interacting interfaces

Proteins are molecular machinery that participate in virtually all essential biological functions within the cell, which are tightly related to their 3D structure. The importance of understanding protein structure-function relationship is highlighted by the exponential growth of experimental structures, which has been greatly expanded by recent breakthroughs in protein structure prediction, most notably RosettaFold, and AlphaFold2. These advances have prompted the development of several computational approaches that leverage these data sources to explore potential biological interactions. However, most methods are generally limited to analysis of single types of interactions, such as protein-protein or protein-ligand interactions, and their complexity limits the usability to expert users. Here we report CSM-Potential2, a deep learning platform for the analysis of binding interfaces on protein structures. In addition to prediction of protein-protein interactions binding sites and classification of biological ligands, our new platform incorporates prediction of interactions with nucleic acids at the residue level and allows for ligand transplantation based on sequence and structure similarity to experimentally determined structures. We anticipate our platform to be a valuable resource that provides easy access to a range of state-of-the-art methods to expert and non-expert users for the study of biological interactions. Our tool is freely available as an easy-to-use web server and API available at https://biosig.lab.uq.edu.au/csm_potential.

Spatom: a graph neural network for structure-based protein-protein interaction site prediction

Accurate identification of protein-protein interaction (PPI) sites remains a computational challenge. We propose Spatom, a novel framework for PPI site prediction. This framework first defines a weighted digraph for a protein structure to precisely characterize the spatial contacts of residues, then performs a weighted digraph convolution to aggregate both spatial local and global information and finally adds an improved graph attention layer to drive the predicted sites to form more continuous region(s). Spatom was tested on a diverse set of challenging protein-protein complexes and demonstrated the best performance among all the compared methods. Furthermore, when tested on multiple popular proteins in a case study, Spatom clearly identifies the interaction interfaces and captures the majority of hotspots. Spatom is expected to contribute to the understanding of protein interactions and drug designs targeting protein binding.

ComplexQA: a deep graph learning approach for protein complex structure assessment

MOTIVATION

In recent years, the end-to-end deep learning method for single-chain protein structure prediction has achieved high accuracy. For example, the state-of-the-art method AlphaFold, developed by Google, has largely increased the accuracy of protein structure predictions to near experimental accuracy in some of the cases. At the same time, there are few methods that can evaluate the quality of protein complexes at the residue level. In particular, evaluating the quality of residues at the interface of protein complexes can lead to a wide range of applications, such as protein function analysis and drug design. In this paper, we introduce a new deep graph neural network-based method ComplexQA, to evaluate the local quality of interfaces for protein complexes by utilizing the residue-level structural information in 3D space and the sequence-level constraints.

RESULTS

We benchmark our method to other state-of-the-art quality assessment approaches on the HAF2 and DBM55-AF2 datasets (high-quality structural models predicted by AlphaFold-Multimer), and the BM5 docking dataset. The experimental results show that our proposed method achieves better or similar performance compared with other state-of-the-art methods, especially on difficult targets which only contain a few acceptable models. Our method is able to suggest a score for each interfac e residue, which demonstrates a powerful assessment tool for the ever-increasing number of protein complexes.

AVAILABILITY

https://github.com/Cao-Labs/ComplexQA.git. Contact: caora@plu.edu.

Enhanced antibody-antigen structure prediction from molecular docking using AlphaFold2

Predicting the structure of antibody-antigen complexes has tremendous value in biomedical research but unfortunately suffers from a poor performance in real-life applications. AlphaFold2 (AF2) has provided renewed hope for improvements in the field of protein-protein docking but has shown limited success against antibody-antigen complexes due to the lack of co-evolutionary constraints. In this study, we used physics-based protein docking methods for building decoy sets consisting of low-energy docking solutions that were either geometrically close to the native structure (positives) or not (negatives). The docking models were then fed into AF2 to assess their confidence with a novel composite score based on normalized pLDDT and pTMscore metrics after AF2 structural refinement. We show benefits of the AF2 composite score for rescoring docking poses both in terms of (1) classification of positives/negatives and of (2) success rates with particular emphasis on early enrichment. Docking models of at least medium quality present in the decoy set, but not necessarily highly ranked by docking methods, benefitted most from AF2 rescoring by experiencing large advances towards the top of the reranked list of models. These improvements, obtained without any calibration or novel methodologies, led to a notable level of performance in antibody-antigen unbound docking that was never achieved previously.

Discriminating physiological from non-physiological interfaces in structures of protein complexes: A community-wide study

Reliably scoring and ranking candidate models of protein complexes and assigning their oligomeric state from the structure of the crystal lattice represent outstanding challenges. A community-wide effort was launched to tackle these challenges. The latest resources on protein complexes and interfaces were exploited to derive a benchmark dataset consisting of 1677 homodimer protein crystal structures, including a balanced mix of physiological and non-physiological complexes. The non-physiological complexes in the benchmark were selected to bury a similar or larger interface area than their physiological counterparts, making it more difficult for scoring functions to differentiate between them. Next, 252 functions for scoring protein-protein interfaces previously developed by 13 groups were collected and evaluated for their ability to discriminate between physiological and non-physiological complexes. A simple consensus score generated using the best performing score of each of the 13 groups, and a cross-validated Random Forest (RF) classifier were created. Both approaches showed excellent performance, with an area under the Receiver Operating Characteristic (ROC) curve of 0.93 and 0.94, respectively, outperforming individual scores developed by different groups. Additionally, AlphaFold2 engines recalled the physiological dimers with significantly higher accuracy than the non-physiological set, lending support to the reliability of our benchmark dataset annotations. Optimizing the combined power of interface scoring functions and evaluating it on challenging benchmark datasets appears to be a promising strategy.

LEVELNET to visualize, explore, and compare protein-protein interaction networks

Physical interactions between proteins are central to all biological processes. Yet, the current knowledge of who interacts with whom in the cell and in what manner relies on partial, noisy, and highly heterogeneous data. Thus, there is a need for methods comprehensively describing and organizing such data. LEVELNET is a versatile and interactive tool for visualizing, exploring, and comparing protein-protein interaction (PPI) networks inferred from different types of evidence. LEVELNET helps to break down the complexity of PPI networks by representing them as multi-layered graphs and by facilitating the direct comparison of their subnetworks toward biological interpretation. It focuses primarily on the protein chains whose 3D structures are available in the Protein Data Bank. We showcase some potential applications, such as investigating the structural evidence supporting PPIs associated to specific biological processes, assessing the co-localization of interaction partners, comparing the PPI networks obtained through computational experiments versus homology transfer, and creating PPI benchmarks with desired properties.

Categorization of hotspots into three types - weak, moderate and strong to distinguish protein-protein versus protein-peptide interactions

Protein-protein and protein-peptide interactions (PPI and PPepI) belong to a similar category of interactions, yet seemingly subtle differences exist among them. To characterize differences between protein-protein (PP) and protein-peptide (PPep) interactions, we have focussed on two important classes of residues-hotspot and anchor residues. Using implicit solvation-based free energy calculations, a very large-scale alanine scanning has been performed on benchmark datasets, consisting of over 5700 interface residues. The differences in the two categories are more pronounced, if the data were divided into three distinct types, namely - weak hotspots (having binding free energy loss upon Ala mutation, ΔΔG, ∼2-10 kcal/mol), moderate hotspots (ΔΔG, ∼10-20 kcal/mol) and strong hotspots (ΔΔG ≥ ∼20 kcal/mol). The analysis suggests that for PPI, weak hotspots are predominantly populated by polar and hydrophobic residues. The distribution shifts towards charged and polar residues for moderate hotspot and charged residues (principally Arg) are overwhelmingly present in the strong hotspot. On the other hand, in the PPepI dataset, the distribution shifts from predominantly hydrophobic and polar (in the weak type) to almost similar preference for polar, hydrophobic and charged residues (in moderate type) and finally the charged residue (Arg) and Trp are mostly occupied in the strong type. The preferred anchor residues in both categories are Arg, Tyr and Leu, possessing bulky side chain and which also strike a delicate balance between side chain flexibility and rigidity. The present knowledge should aid in effective design of biologics, by augmentation or disruption of PPIs with peptides or peptidomimetics.Communicated by Ramaswamy H. Sarma.

Integration of pre-trained protein language models into geometric deep learning networks

Geometric deep learning has recently achieved great success in non-Euclidean domains, and learning on 3D structures of large biomolecules is emerging as a distinct research area. However, its efficacy is largely constrained due to the limited quantity of structural data. Meanwhile, protein language models trained on substantial 1D sequences have shown burgeoning capabilities with scale in a broad range of applications. Several preceding studies consider combining these different protein modalities to promote the representation power of geometric neural networks but fail to present a comprehensive understanding of their benefits. In this work, we integrate the knowledge learned by well-trained protein language models into several state-of-the-art geometric networks and evaluate a variety of protein representation learning benchmarks, including protein-protein interface prediction, model quality assessment, protein-protein rigid-body docking, and binding affinity prediction. Our findings show an overall improvement of 20% over baselines. Strong evidence indicates that the incorporation of protein language models' knowledge enhances geometric networks' capacity by a significant margin and can be generalized to complex tasks.

DIPS-Plus: The enhanced database of interacting protein structures for interface prediction

In this work, we expand on a dataset recently introduced for protein interface prediction (PIP), the Database of Interacting Protein Structures (DIPS), to present DIPS-Plus, an enhanced, feature-rich dataset of 42,112 complexes for machine learning of protein interfaces. While the original DIPS dataset contains only the Cartesian coordinates for atoms contained in the protein complex along with their types, DIPS-Plus contains multiple residue-level features including surface proximities, half-sphere amino acid compositions, and new profile hidden Markov model (HMM)-based sequence features for each amino acid, providing researchers a curated feature bank for training protein interface prediction methods. We demonstrate through rigorous benchmarks that training an existing state-of-the-art (SOTA) model for PIP on DIPS-Plus yields new SOTA results, surpassing the performance of some of the latest models trained on residue-level and atom-level encodings of protein complexes to date.

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.

Evaluation of AlphaFold Antibody-Antigen Modeling with Implications for Improving Predictive Accuracy

High resolution antibody-antigen structures provide critical insights into immune recognition and can inform therapeutic design. The challenges of experimental structural determination and the diversity of the immune repertoire underscore the necessity of accurate computational tools for modeling antibody-antigen complexes. Initial benchmarking showed that despite overall success in modeling protein-protein complexes, AlphaFold and AlphaFold-Multimer have limited success in modeling antibody-antigen interactions. In this study, we performed a thorough analysis of AlphaFold's antibody-antigen modeling performance on 429 nonredundant antibody-antigen complex structures, identifying useful confidence metrics for predicting model quality, and features of complexes associated with improved modeling success. We show the importance of bound-like component modeling in complex assembly accuracy, and that the current version of AlphaFold improves near-native modeling success to over 30%, versus approximately 20% for a previous version. With this improved success, AlphaFold can generate accurate antibody-antigen models in many cases, while additional training may further improve its performance.

Discovering functionally important sites in proteins

Proteins play important roles in biology, biotechnology and pharmacology, and missense variants are a common cause of disease. Discovering functionally important sites in proteins is a central but difficult problem because of the lack of large, systematic data sets. Sequence conservation can highlight residues that are functionally important but is often convoluted with a signal for preserving structural stability. We here present a machine learning method to predict functional sites by combining statistical models for protein sequences with biophysical models of stability. We train the model using multiplexed experimental data on variant effects and validate it broadly. We show how the model can be used to discover active sites, as well as regulatory and binding sites. We illustrate the utility of the model by prospective prediction and subsequent experimental validation on the functional consequences of missense variants in HPRT1 which may cause Lesch-Nyhan syndrome, and pinpoint the molecular mechanisms by which they cause disease.

EnCPdock: a web-interface for direct conjoint comparative analyses of complementarity and binding energetics in inter-protein associations

CONTEXT

Protein-protein interaction (PPI) is a key component linked to virtually all cellular processes. Be it an enzyme catalysis ('classic type functions' of proteins) or a signal transduction ('non-classic'), proteins generally function involving stable or quasi-stable multi-protein associations. The physical basis for such associations is inherent in the combined effect of shape and electrostatic complementarities (Sc, EC) of the interacting protein partners at their interface, which provides indirect probabilistic estimates of the stability and affinity of the interaction. While Sc is a necessary criterion for inter-protein associations, EC can be favorable as well as disfavored (e.g., in transient interactions). Estimating equilibrium thermodynamic parameters (∆G_binding, K_d) by experimental means is costly and time consuming, thereby opening windows for computational structural interventions. Attempts to empirically probe ∆G_binding from coarse-grain structural descriptors (primarily, surface area based terms) have lately been overtaken by physics-based, knowledge-based and their hybrid approaches (MM/PBSA, FoldX, etc.) that directly compute ∆G_binding without involving intermediate structural descriptors.

METHODS

Here, we present EnCPdock ( https://www.scinetmol.in/EnCPdock/ ), a user-friendly web-interface for the direct conjoint comparative analyses of complementarity and binding energetics in proteins. EnCPdock returns an AI-predicted ∆G_binding computed by combining complementarity (Sc, EC) and other high-level structural descriptors (input feature vectors), and renders a prediction accuracy comparable to the state-of-the-art. EnCPdock further locates a PPI complex in terms of its {Sc, EC} values (taken as an ordered pair) in the two-dimensional complementarity plot (CP). In addition, it also generates mobile molecular graphics of the interfacial atomic contact network for further analyses. EnCPdock also furnishes individual feature trends along with the relative probability estimates (Pr_fmax) of the obtained feature-scores with respect to the events of their highest observed frequencies. Together, these functionalities are of real practical use for structural tinkering and intervention as might be relevant in the design of targeted protein-interfaces. Combining all its features and applications, EnCPdock presents a unique online tool that should be beneficial to structural biologists and researchers across related fraternities.

Flexible Protein-Protein Docking with a Multi-Track Iterative Transformer

Conventional protein-protein docking algorithms usually rely on heavy candidate sampling and re-ranking, but these steps are time-consuming and hinder applications that require high-throughput complex structure prediction, e.g., structure-based virtual screening. Existing deep learning methods for protein-protein docking, despite being much faster, suffer from low docking success rates. In addition, they simplify the problem to assume no conformational changes within any protein upon binding (rigid docking). This assumption precludes applications when binding-induced conformational changes play a role, such as allosteric inhibition or docking from uncertain unbound model structures. To address these limitations, we present GeoDock, a multi-track iterative transformer network to predict a docked structure from separate docking partners. Unlike deep learning models for protein structure prediction that input multiple sequence alignments (MSAs), GeoDock inputs just the sequences and structures of the docking partners, which suits the tasks when the individual structures are given. GeoDock is flexible at the protein residue level, allowing the prediction of conformational changes upon binding. For a benchmark set of rigid targets, GeoDock obtains a 41% success rate, outperforming all the other tested methods. For a more challenging benchmark set of flexible targets, GeoDock achieves a similar number of top-model successes as the traditional method ClusPro [1], but fewer than ReplicaDock2 [2]. GeoDock attains an average inference speed of under one second on a single GPU, enabling its application in large-scale structure screening. Although binding-induced conformational changes are still a challenge owing to limited training and evaluation data, our architecture sets up the foundation to capture this backbone flexibility. Code and a demonstration Jupyter notebook are available at https://github.com/Graylab/GeoDock.

Deep Local Analysis deconstructs protein-protein interfaces and accurately estimates binding affinity changes upon mutation

MOTIVATION

The spectacular recent advances in protein and protein complex structure prediction hold promise for reconstructing interactomes at large-scale and residue resolution. Beyond determining the 3D arrangement of interacting partners, modeling approaches should be able to unravel the impact of sequence variations on the strength of the association.

RESULTS

In this work, we report on Deep Local Analysis, a novel and efficient deep learning framework that relies on a strikingly simple deconstruction of protein interfaces into small locally oriented residue-centered cubes and on 3D convolutions recognizing patterns within cubes. Merely based on the two cubes associated with the wild-type and the mutant residues, DLA accurately estimates the binding affinity change for the associated complexes. It achieves a Pearson correlation coefficient of 0.735 on about 400 mutations on unseen complexes. Its generalization capability on blind datasets of complexes is higher than the state-of-the-art methods. We show that taking into account the evolutionary constraints on residues contributes to predictions. We also discuss the influence of conformational variability on performance. Beyond the predictive power on the effects of mutations, DLA is a general framework for transferring the knowledge gained from the available non-redundant set of complex protein structures to various tasks. For instance, given a single partially masked cube, it recovers the identity and physicochemical class of the central residue. Given an ensemble of cubes representing an interface, it predicts the function of the complex.

AVAILABILITY AND IMPLEMENTATION

Source code and models are available at http://gitlab.lcqb.upmc.fr/DLA/DLA.git.

A gated graph transformer for protein complex structure quality assessment and its performance in CASP15

MOTIVATION

Proteins interact to form complexes to carry out essential biological functions. Computational methods such as AlphaFold-multimer have been developed to predict the quaternary structures of protein complexes. An important yet largely unsolved challenge in protein complex structure prediction is to accurately estimate the quality of predicted protein complex structures without any knowledge of the corresponding native structures. Such estimations can then be used to select high-quality predicted complex structures to facilitate biomedical research such as protein function analysis and drug discovery.

RESULTS

In this work, we introduce a new gated neighborhood-modulating graph transformer to predict the quality of 3D protein complex structures. It incorporates node and edge gates within a graph transformer framework to control information flow during graph message passing. We trained, evaluated and tested the method (called DProQA) on newly-curated protein complex datasets before the 15th Critical Assessment of Techniques for Protein Structure Prediction (CASP15) and then blindly tested it in the 2022 CASP15 experiment. The method was ranked 3rd among the single-model quality assessment methods in CASP15 in terms of the ranking loss of TM-score on 36 complex targets. The rigorous internal and external experiments demonstrate that DProQA is effective in ranking protein complex structures.

AVAILABILITY AND IMPLEMENTATION

The source code, data, and pre-trained models are available at https://github.com/jianlin-cheng/DProQA.

GDockScore: a graph-based protein-protein docking scoring function

Summary

Protein complexes play vital roles in a variety of biological processes, such as mediating biochemical reactions, the immune response and cell signalling, with 3D structure specifying function. Computational docking methods provide a means to determine the interface between two complexed polypeptide chains without using time-consuming experimental techniques. The docking process requires the optimal solution to be selected with a scoring function. Here, we propose a novel graph-based deep learning model that utilizes mathematical graph representations of proteins to learn a scoring function (GDockScore). GDockScore was pre-trained on docking outputs generated with the Protein Data Bank biounits and the RosettaDock protocol, and then fine-tuned on HADDOCK decoys generated on the ZDOCK Protein Docking Benchmark. GDockScore performs similarly to the Rosetta scoring function on docking decoys generated using the RosettaDock protocol. Furthermore, state-of-the-art is achieved on the CAPRI score set, a challenging dataset for developing docking scoring functions.

Availability and implementation

The model implementation is available at https://gitlab.com/mcfeemat/gdockscore.

Supplementary information

Supplementary data are available at Bioinformatics Advances online.

The ClusPro AbEMap web server for the prediction of antibody epitopes

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

AREA-AFFINITY: A Web Server for Machine Learning-Based Prediction of Protein-Protein and Antibody-Protein Antigen Binding Affinities

Protein-Protein binding affinity reflects the binding strength between the binding partners. The prediction of protein-protein binding affinity is important for elucidating protein functions and also for designing protein-based therapeutics. The geometric characteristics such as area (both interface and surface areas) in the structure of a protein-protein complex play an important role in determining protein-protein interactions and their binding affinity. Here, we present a free web server for academic use, AREA-AFFINITY, for prediction of protein-protein or antibody-protein antigen binding affinity based on interface and surface areas in the structure of a protein-protein complex. AREA-AFFINITY implements 60 effective area-based protein-protein affinity predictive models and 37 effective area-based models specific for antibody-protein antigen binding affinity prediction developed in our recent studies. These models take into consideration the roles of interface and surface areas in binding affinity by using areas classified according to different amino acid types with different biophysical nature. The models with the best performances integrate machine learning methods such as neural network or random forest. These newly developed models have superior or comparable performance compared to the commonly used existing methods. AREA-AFFINITY is available for free at: https://affinity.cuhk.edu.cn/.

What Makes a Good Protein-Protein Interaction Stabilizer: Analysis and Application of the Dual-Binding Mechanism

Protein-protein interactions (PPIs) are essential for biological processes including immune reactions and diseases. Inhibition of PPIs by drug-like compounds is a common basis for therapeutic approaches. In many cases the flat interface of PP complexes prevents discovery of specific compound binding to cavities on one partner and PPI inhibition. However, frequently new pockets are formed at the PP interface that allow accommodation of stabilizers which is often as desirable as inhibition but a much less explored alternative strategy. Herein, we employ molecular dynamics simulations and pocket detection to investigate 18 known stabilizers and associated PP complexes. For most cases, we find that a dual-binding mechanism, a similar stabilizer interaction strength with each protein partner, is an important prerequisite for effective stabilization. A few stabilizers follow an allosteric mechanism by stabilizing the protein bound structure and/or increase the PPI indirectly. On 226 protein-protein complexes, we find in >75% of the cases interface cavities suitable for binding of drug-like compounds. We propose a computational compound identification workflow that exploits new PP interface cavities and optimizes the dual-binding mechanism and apply it to 5 PP complexes. Our study demonstrates a great potential for in silico PPI stabilizers discovery with a wide range of therapeutic applications.