Predicting protein-peptide interaction sites using distant protein complexes as structural templates

PepCA: Unveiling protein-peptide interaction sites with a multi-input neural network model

The protein-peptide interaction plays a pivotal role in fields such as drug development, yet remains underexplored experimentally and challenging to model computationally. Herein, we introduce PepCA, a sequence-based approach for predicting peptide-binding sites on proteins. A primary obstacle in predicting peptide-protein interactions is the difficulty in acquiring precise protein structures, coupled with the uncertainty of polypeptide configurations. To address this, we first encode protein sequences using the Evolutionary Scale Modeling 2 (ESM-2) pre-trained model to extract latent structural information. Additionally, we have developed a multi-input coattention mechanism to concurrently update the encoding of both peptide and protein residues. PepCA integrates this module within an encoder-decoder structure. This model's high precision in identifying binding sites significantly advances the field of computational biology, offering vital insights for peptide drug development and protein science.

The role and future prospects of artificial intelligence algorithms in peptide drug development

Peptide medications have been more well-known in recent years due to their many benefits, including low side effects, high biological activity, specificity, effectiveness, and so on. Over 100 peptide medications have been introduced to the market to treat a variety of illnesses. Most of these peptide medications are developed on the basis of endogenous peptides or natural peptides, which frequently required expensive, time-consuming, and extensive tests to confirm. As artificial intelligence advances quickly, it is now possible to build machine learning or deep learning models that screen a large number of candidate sequences for therapeutic peptides. Therapeutic peptides, such as those with antibacterial or anticancer properties, have been developed by the application of artificial intelligence algorithms.The process of finding and developing peptide drugs is outlined in this review, along with a few related cases that were helped by AI and conventional methods. These resources will open up new avenues for peptide drug development and discovery, helping to meet the pressing needs of clinical patients for disease treatment. Although peptide drugs are a new class of biopharmaceuticals that distinguish them from chemical and small molecule drugs, their clinical purpose and value cannot be ignored. However, the traditional peptide drug research and development has a long development cycle and high investment, and the creation of peptide medications will be substantially hastened by the AI-assisted (AI+) mode, offering a new boost for combating diseases.

GAPS: a geometric attention-based network for peptide binding site identification by the transfer learning approach

Protein-peptide interactions (PPepIs) are vital to understanding cellular functions, which can facilitate the design of novel drugs. As an essential component in forming a PPepI, protein-peptide binding sites are the basis for understanding the mechanisms involved in PPepIs. Therefore, accurately identifying protein-peptide binding sites becomes a critical task. The traditional experimental methods for researching these binding sites are labor-intensive and time-consuming, and some computational tools have been invented to supplement it. However, these computational tools have limitations in generality or accuracy due to the need for ligand information, complex feature construction, or their reliance on modeling based on amino acid residues. To deal with the drawbacks of these computational algorithms, we describe a geometric attention-based network for peptide binding site identification (GAPS) in this work. The proposed model utilizes geometric feature engineering to construct atom representations and incorporates multiple attention mechanisms to update relevant biological features. In addition, the transfer learning strategy is implemented for leveraging the protein-protein binding sites information to enhance the protein-peptide binding sites recognition capability, taking into account the common structure and biological bias between proteins and peptides. Consequently, GAPS demonstrates the state-of-the-art performance and excellent robustness in this task. Moreover, our model exhibits exceptional performance across several extended experiments including predicting the apo protein-peptide, protein-cyclic peptide and the AlphaFold-predicted protein-peptide binding sites. These results confirm that the GAPS model is a powerful, versatile, stable method suitable for diverse binding site predictions.

Leveraging machine learning models for peptide-protein interaction prediction

Peptides play a pivotal role in a wide range of biological activities through participating in up to 40% protein-protein interactions in cellular processes. They also demonstrate remarkable specificity and efficacy, making them promising candidates for drug development. However, predicting peptide-protein complexes by traditional computational approaches, such as docking and molecular dynamics simulations, still remains a challenge due to high computational cost, flexible nature of peptides, and limited structural information of peptide-protein complexes. In recent years, the surge of available biological data has given rise to the development of an increasing number of machine learning models for predicting peptide-protein interactions. These models offer efficient solutions to address the challenges associated with traditional computational approaches. Furthermore, they offer enhanced accuracy, robustness, and interpretability in their predictive outcomes. This review presents a comprehensive overview of machine learning and deep learning models that have emerged in recent years for the prediction of peptide-protein interactions.

Peptriever: a Bi-Encoder approach for large-scale protein-peptide binding search

MOTIVATION

Peptide therapeutics hinge on the precise interaction between a tailored peptide and its designated receptor while mitigating interactions with alternate receptors is equally indispensable. Existing methods primarily estimate the binding score between protein and peptide pairs. However, for a specific peptide without a corresponding protein, it is challenging to identify the proteins it could bind due to the sheer number of potential candidates.

RESULTS

We propose a transformers-based protein embedding scheme in this study that can quickly identify and rank millions of interacting proteins. Furthermore, the proposed approach outperforms existing sequence- and structure-based methods, with a mean AUC-ROC and AUC-PR of 0.73.

AVAILABILITY AND IMPLEMENTATION

Training data, scripts, and fine-tuned parameters are available at https://github.com/RoniGurvich/Peptriever. The proposed method is linked with a web application available for customized prediction at https://peptriever.app/.

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.

Leveraging Machine Learning Models for Peptide-Protein Interaction Prediction

Peptides play a pivotal role in a wide range of biological activities through participating in up to 40% protein-protein interactions in cellular processes. They also demonstrate remarkable specificity and efficacy, making them promising candidates for drug development. However, predicting peptide-protein complexes by traditional computational approaches, such as Docking and Molecular Dynamics simulations, still remains a challenge due to high computational cost, flexible nature of peptides, and limited structural information of peptide-protein complexes. In recent years, the surge of available biological data has given rise to the development of an increasing number of machine learning models for predicting peptide-protein interactions. These models offer efficient solutions to address the challenges associated with traditional computational approaches. Furthermore, they offer enhanced accuracy, robustness, and interpretability in their predictive outcomes. This review presents a comprehensive overview of machine learning and deep learning models that have emerged in recent years for the prediction of peptide-protein interactions.

An Update on the Application of CRISPR Technology in Clinical Practice

The CRISPR/Cas system, an innovative gene-editing tool, is emerging as a promising technique for genome modifications. This straightforward technique was created based on the prokaryotic adaptive immune defense mechanism and employed in the studies on human diseases that proved enormous therapeutic potential. A genetically unique patient mutation in the process of gene therapy can be corrected by the CRISPR method to treat diseases that traditional methods were unable to cure. However, introduction of CRISPR/Cas9 into the clinic will be challenging because we still need to improve the technology's effectiveness, precision, and applications. In this review, we first describe the function and applications of the CRISPR-Cas9 system. We next delineate how this technology could be utilized for gene therapy of various human disorders, including cancer and infectious diseases and highlight the promising examples in the field. Finally, we document current challenges and the potential solutions to overcome these obstacles for the effective use of CRISPR-Cas9 in clinical practice.

Use of tree-based machine learning methods to screen affinitive peptides based on docking data

Screening peptides with good affinity is an important step in peptide-drug discovery. Recent advancement in computer and data science have made machine learning a useful tool in accurately affinitive-peptide screening. In current study, four different tree-based algorithms, including Classification and regression trees (CART), C5.0 decision tree (C50), Bagged CART (BAG) and Random Forest (RF), were employed to explore the relationship between experimental peptide affinities and virtual docking data, and the performance of each model was also compared in parallel. All four algorithms showed better performances on dataset pre-scaled, -centered and -PCA than other pre-processed dataset. After model re-built and hyperparameter optimization, the optimal C50 model (C50O) showed the best performances in terms of Accuracy, Kappa, Sensitivity, Specificity, F1, MCC and AUC when validated on test data and an unknown PEDV datasets evaluation (Accuracy=80.4 %). BAG and RFO (the optimal RF), as two best models during training process, did not performed as expecting during in testing and unknown dataset validations. Furthermore, the high correlation of the predictions of RFO and BAG to C50O implied the high stability and robustness of their prediction. Whereas although the good performance on unknown dataset, the poor performance in test data validation and correlation analysis indicated CARTO could not be used for future data prediction. To accurately evaluate the peptide affinity, the current study firstly gave a tree-model competition on affinitive peptide prediction by using virtual docking data, which would expand the application of machine learning algorithms in studying PepPIs and benefit the development of peptide therapeutics.

Exploring binding positions and backbone conformations of peptide ligands of proteins with a backbone-centred statistical energy function

When designing peptide ligands based on the structure of a protein receptor, it can be very useful to narrow down the possible binding positions and bound conformations of the ligand without the need to choose its amino acid sequence in advance. Here, we construct and benchmark a tool for this purpose based on a recently reported statistical energy model named SCUBA (Sidechain-Unknown Backbone Arrangement) for designing protein backbones without considering specific amino acid sequences. With this tool, backbone fragments of different local conformation types are generated and optimized with SCUBA-driven stochastic simulations and simulated annealing, and then ranked and clustered to obtain representative backbone fragment poses of strong SCUBA interaction energies with the receptor. We computationally benchmarked the tool on 111 known protein-peptide complex structures. When the bound ligands are in the strand conformation, the method is able to generate backbone fragments of both low SCUBA energies and low root mean square deviations from experimental structures of peptide ligands. When the bound ligands are helices or coils, low-energy backbone fragments with binding poses similar to experimental structures have been generated for approximately 50% of benchmark cases. We have examined a number of predicted ligand-receptor complexes by atomistic molecular dynamics simulations, in which the peptide ligands have been found to stay at the predicted binding sites and to maintain their local conformations. These results suggest that promising backbone structures of peptides bound to protein receptors can be designed by identifying outstanding minima on the SCUBA-modeled backbone energy landscape.

A Random Forest Model for Peptide Classification Based on Virtual Docking Data

The affinity of peptides is a crucial factor in studying peptide-protein interactions. Despite the development of various techniques to evaluate peptide-receptor affinity, the results may not always reflect the actual affinity of the peptides accurately. The current study provides a free tool to assess the actual peptide affinity based on virtual docking data. This study employed a dataset that combined actual peptide affinity information (active and inactive) and virtual peptide-receptor docking data, and different machine learning algorithms were utilized. Compared with the other algorithms, the random forest (RF) algorithm showed the best performance and was used in building three RF models using different numbers of significant features (four, three, and two). Further analysis revealed that the four-feature RF model achieved the highest Accuracy of 0.714 in classifying an independent unknown peptide dataset designed with the PEDV spike protein, and it also revealed overfitting problems in the other models. This four-feature RF model was used to evaluate peptide affinity by constructing the relationship between the actual affinity and the virtual docking scores of peptides to their receptors.

Roles of Different β-Defensins in the Human Reproductive System: A Review Study

Human β-defensins (hBDs) are cationic peptides with an amphipathic spatial shape and a high cysteine content. The members of this peptide family have been found in the human body with various functions, including the human reproductive system. Of among β-defensins in the human body, β-defensin 1, β-defensin 2, and β-defensin 126 are known in the human reproductive system. Human β-defensin 1 interacts with chemokine receptor 6 (CCR6) in the male reproductive system to prevent bacterial infections. This peptide has a positive function in antitumor immunity by recruiting dendritic cells and memory T cells in prostate cancer. It is necessary for fertilization via facilitating capacitation and acrosome reaction in the female reproductive system. Human β-defensin 2 is another peptide with antibacterial action which can minimize infection in different parts of the female reproductive system such as the vagina by interacting with CCR6. Human β-defensin 2 could play a role in preventing cervical cancer via interactions with dendritic cells. Human β-defensin 126 is required for sperm motility and protecting the sperm against immune system factors. This study attempted to review the updated knowledge about the roles of β-defensin 1, β-defensin 2, and β-defensin 126 in both the male and female reproductive systems.

In Silico and In Vivo Studies of a Tumor-Penetrating and Interfering Peptide with Antitumoral Effect on Xenograft Models of Breast Cancer

The combination of a tumor-penetrating peptide (TPP) with a peptide able to interfere with a given protein-protein interaction (IP) is a promising strategy with potential clinical application. Little is known about the impact of fusing a TPP with an IP, both in terms of internalization and functional effect. Here, we analyze these aspects in the context of breast cancer, targeting PP2A/SET interaction, using both in silico and in vivo approaches. Our results support the fact that state-of-the-art deep learning approaches developed for protein-peptide interaction modeling can reliably identify good candidate poses for the IP-TPP in interaction with the Neuropilin-1 receptor. The association of the IP with the TPP does not seem to affect the ability of the TPP to bind to Neuropilin-1. Molecular simulation results suggest that peptide IP-GG-LinTT1 in a cleaved form interacts with Neuropilin-1 in a more stable manner and has a more helical secondary structure than the cleaved IP-GG-iRGD. Surprisingly, in silico investigations also suggest that the non-cleaved TPPs can bind the Neuropilin-1 in a stable manner. The in vivo results using xenografts models show that both bifunctional peptides resulting from the combination of the IP and either LinTT1 or iRGD are effective against tumoral growth. The peptide iRGD-IP shows the highest stability to serum proteases degradation while having the same antitumoral effect as Lin TT1-IP, which is more sensitive to proteases degradation. Our results support the development of the TPP-IP strategy as therapeutic peptides against cancer.

An Integrated Machine Learning Model To Spot Peptide Binding Pockets in 3D Protein Screening

The prediction of peptide-protein binding sites is of utmost importance to tackle the onset of severe neurodegenerative diseases and cancer. In this work, we detail a novel machine learning model based on Linear Discriminant Analysis (LDA) demonstrating to be highly predictive in detecting the putative protein binding regions of small peptides. Starting from 439 high-quality pockets derived from peptide-protein crystallographic complexes, three sets of well-established peptide-binding regions were first selected through a Partitioning Around Medoids (PAM) clustering algorithm based on morphological and energetic 3D GRID-MIF molecular descriptors. Next, the best combination between all the putative interacting peptide pockets and related GRID-MIF scores was automatically explored by using the LDA-based protocol implemented in BioGPS. This approach proved successful to recognize the actual interacting peptide regions (that is, AUC = 0.86 and partial ROC enrichment at 5% of 0.48) from all the other pockets of the protein. Validated on two external collections sets, including 445 and 347 crystallographic peptide-protein complexes, our LDA-based model could be effective to further run peptide-protein virtual screening campaigns.

Comparative analysis of human and bovine thyroglobulin structures

In biology, evolutionary conserved protein sequences show homologous physiological phenotypes in their structures and functions. If the protein has a vital function, its sequence is usually conserved across the species. However, in highly conserved protein there still remains small differences across the species. Upon protein–protein interaction (PPI), it is observed that the conserved proteins can have different binding partners that are considered to be caused by the small sequence variations in a specific domain. Thyroglobulin (TG) is the most commonly found protein in the thyroid gland of vertebrates and serves as the precursor of the thyroid hormones, tetraiodothyronine and triiodothyronine that are critical for growth, development and metabolism in vertebrates. In this study, we comparatively analyzed the sequences and structures of the highly conserved regions of TG from two different species in relation to their PPIs. In order to do so, we employed SIM for sequence alignment, STRING for PPI analysis and cryo-electron microscopy for 3D structural analysis. Our Cryo-EM model for TG of Bos taurus determined at 7.1 Å resolution fitted well with the previously published Cryo-EM model for Homo sapiens TG. By demonstrating overall structural homology between TGs from different species, we address that local amino acid sequence variation is sufficient to alter PPIs specific for the organism. We predict that our result will contribute to a deeper understanding in the evolutionary pattern applicable to many other proteins.

Overview of methods for characterization and visualization of a protein-protein interaction network in a multi-omics integration context

At the heart of the cellular machinery through the regulation of cellular functions, protein-protein interactions (PPIs) have a significant role. PPIs can be analyzed with network approaches. Construction of a PPI network requires prediction of the interactions. All PPIs form a network. Different biases such as lack of data, recurrence of information, and false interactions make the network unstable. Integrated strategies allow solving these different challenges. These approaches have shown encouraging results for the understanding of molecular mechanisms, drug action mechanisms, and identification of target genes. In order to give more importance to an interaction, it is evaluated by different confidence scores. These scores allow the filtration of the network and thus facilitate the representation of the network, essential steps to the identification and understanding of molecular mechanisms. In this review, we will discuss the main computational methods for predicting PPI, including ones confirming an interaction as well as the integration of PPIs into a network, and we will discuss visualization of these complex data.

Insights into protein-DNA interactions from hydrogen bond energy-based comparative protein-ligand analyses

Hydrogen bonds play important roles in protein folding and protein-ligand interactions, particularly in specific protein-DNA recognition. However, the distributions of hydrogen bonds, especially hydrogen bond energy (HBE) in different types of protein-ligand complexes, is unknown. Here we performed a comparative analysis of hydrogen bonds among three non-redundant datasets of protein-protein, protein-peptide, and protein-DNA complexes. Besides comparing the number of hydrogen bonds in terms of types and locations, we investigated the distributions of HBE. Our results indicate that while there is no significant difference of hydrogen bonds within protein chains among the three types of complexes, interfacial hydrogen bonds are significantly more prevalent in protein-DNA complexes. More importantly, the interfacial hydrogen bonds in protein-DNA complexes displayed a unique energy distribution of strong and weak hydrogen bonds whereas majority of the interfacial hydrogen bonds in protein-protein and protein-peptide complexes are of predominantly high strength with low energy. Moreover, there is a significant difference in the energy distributions of minor groove hydrogen bonds between protein-DNA complexes with different binding specificity. Highly specific protein-DNA complexes contain more strong hydrogen bonds in the minor groove than multi-specific complexes, suggesting important role of minor groove in specific protein-DNA recognition. These results can help better understand protein-DNA interactions and have important implications in improving quality assessments of protein-DNA complex models.

PepNN: a deep attention model for the identification of peptide binding sites

Protein-peptide interactions play a fundamental role in many cellular processes, but remain underexplored experimentally and difficult to model computationally. Here, we present PepNN-Struct and PepNN-Seq, structure and sequence-based approaches for the prediction of peptide binding sites on a protein. A main difficulty for the prediction of peptide-protein interactions is the flexibility of peptides and their tendency to undergo conformational changes upon binding. Motivated by this, we developed reciprocal attention to simultaneously update the encodings of peptide and protein residues while enforcing symmetry, allowing for information flow between the two inputs. PepNN integrates this module with modern graph neural network layers and a series of transfer learning steps are used during training to compensate for the scarcity of peptide-protein complex information. We show that PepNN-Struct achieves consistently high performance across different benchmark datasets. We also show that PepNN makes reasonable peptide-agnostic predictions, allowing for the identification of novel peptide binding proteins.

Machine learning assessment of the binding region as a tool for more efficient computational receptor-ligand docking

We present a combined computational approach to protein-ligand binding, which consists of two steps: (1) a deep neural network is used to locate a binding region on a target protein, and (2) molecular docking of a ligand is performed within the specified region to obtain the best pose using Autodock Vina. Our in-house designed neural network was trained using the PepBDB dataset. Although the training dataset consisted of protein-peptide complexes, we show that the approach is not limited to peptides, but also works remarkably well for a large class of non-peptide ligands. The results are compared with those in which the binding region (first step) was provided by Accluster. In cases where no prior experimental data on the binding region are available, our deep neural network provides a fast and effective alternative to classical software for its localization. Our code is available at https://github.com/mksmd/NNforDocking.

Strategies to overcome the main challenges of the use of CRISPR/Cas9 as a replacement for cancer therapy

CRISPR/Cas9 (clustered regularly interspaced short palindromic repeats-associated protein 9) shows the opportunity to treat a diverse array of untreated various genetic and complicated disorders. Therapeutic genome editing processes that target disease-causing genes or mutant genes have been greatly accelerated in recent years as a consequence of improvements in sequence-specific nuclease technology. However, the therapeutic promise of genome editing has yet to be explored entirely, many challenges persist that increase the risk of further mutations. Here, we highlighted the main challenges facing CRISPR/Cas9-based treatments and proposed strategies to overcome these limitations, for further enhancing this revolutionary novel therapeutics to improve long-term treatment outcome human health.

Getting Insights into Structural and Energetic Properties of Reciprocal Peptide-Protein Interactions

Peptide-protein interactions play a key role for many cellular and metabolic processes involved in the onset of largely spread diseases such as cancer and neurodegenerative pathologies. Despite the progress in the structural characterization of peptide-protein interfaces, the in-depth knowledge of the molecular details behind their interactions is still a daunting task. Here, we present the first comprehensive in silico morphological and energetic study of peptide binding sites by focusing on both peptide and protein standpoints. Starting from the PixelDB database, a nonredundant benchmark collection of high-quality 3D crystallographic structures of peptide-protein complexes, a classification analysis of the most representative categories based on the nature of each cocrystallized peptide has been carried out. Several interpretable geometrical and energetic descriptors have been computed both from peptide and target protein sides in the attempt to unveil physicochemical and structural causative correlations. Finally, we investigated the most frequent peptide-protein residue pairs at the binding interface and made extensive energetic analyses, based on GRID MIFs, with the aim to study the peptide affinity-enhancing interactions to be further exploited in rational drug design strategies.

Harnessing protein folding neural networks for peptide-protein docking

Highly accurate protein structure predictions by deep neural networks such as AlphaFold2 and RoseTTAFold have tremendous impact on structural biology and beyond. Here, we show that, although these deep learning approaches have originally been developed for the in silico folding of protein monomers, AlphaFold2 also enables quick and accurate modeling of peptide-protein interactions. Our simple implementation of AlphaFold2 generates peptide-protein complex models without requiring multiple sequence alignment information for the peptide partner, and can handle binding-induced conformational changes of the receptor. We explore what AlphaFold2 has memorized and learned, and describe specific examples that highlight differences compared to state-of-the-art peptide docking protocol PIPER-FlexPepDock. These results show that AlphaFold2 holds great promise for providing structural insight into a wide range of peptide-protein complexes, serving as a starting point for the detailed characterization and manipulation of these interactions.

A novel multi-objective metaheuristic algorithm for protein-peptide docking and benchmarking on the LEADS-PEP dataset

Protein-peptide interactions have attracted the attention of many drug discovery scientists due to their possible druggability features on most key biological activities such as regulating disease-related signaling pathways and enhancing the immune system's responses. Different studies have utilized some protein-peptide-specific docking algorithms/methods to predict protein-peptide interactions. However, the existing algorithms/methods suffer from two serious limitations which make them unsuitable for protein-peptide docking problems. First, it seems that the prevalent approaches require to be modified and remodeled for weighting the unbounded forces between a protein and a peptide. Second, they do not employ state-of-the-art search algorithms for detecting the 3D pose of a peptide relative to a protein. To address these restrictions, the present study aims to introduce a novel multi-objective algorithm, which first generates some potential 3D poses of a peptide, and then, improves them through its operators. The candidate solutions are further evaluated using Multi-Objective Pareto Front (MOPF) optimization concepts. To this end, van der Waals, electrostatic, solvation, and hydrogen bond energies between the atoms of a protein and designated peptide are computed. To evaluate the algorithm, it is first applied to the LEADS-PEP dataset containing 53 protein-peptide complexes with up to 53 rotatable branches/bonds and then compared with three popular/efficient algorithms. The obtained results indicate that the MOPF-based approaches which reduce the backbone RMSD between the original and predicted states, achieve significantly better results in terms of the success rate in predicting the near-native conditions. Besides, a comparison between the different types of search algorithms reveals that efficient ones like the multi-objective Trader/differential evolution algorithm can predict protein-peptide interactions better than the popular algorithms such as the multi-objective genetic/particle swarm optimization algorithms.

A deep-learning framework for multi-level peptide-protein interaction prediction

Peptide-protein interactions are involved in various fundamental cellular functions and their identification is crucial for designing efficacious peptide therapeutics. Recently, a number of computational methods have been developed to predict peptide-protein interactions. However, most of the existing prediction approaches heavily depend on high-resolution structure data. Here, we present a deep learning framework for multi-level peptide-protein interaction prediction, called CAMP, including binary peptide-protein interaction prediction and corresponding peptide binding residue identification. Comprehensive evaluation demonstrated that CAMP can successfully capture the binary interactions between peptides and proteins and identify the binding residues along the peptides involved in the interactions. In addition, CAMP outperformed other state-of-the-art methods on binary peptide-protein interaction prediction. CAMP can serve as a useful tool in peptide-protein interaction prediction and identification of important binding residues in the peptides, which can thus facilitate the peptide drug discovery process.

Experimental and Computational Models for Side Chain Discrimination in Peptide-Protein Interactions

A bis(18‐crown‐6) Tröger's base receptor and 4‐substituted hepta‐1,7‐diyl bisammonium salt ligands have been used as a model system to study the interactions between non‐polar side chains of peptides and an aromatic cavity of a protein. NMR titrations and NOESY/ROESY NMR spectroscopy were used to analyze the discrimination of the ligands by the receptor based on the substituent of the ligand, both quantitatively (free binding energies) and qualitatively (conformations). The analysis showed that an all‐anti conformation of the heptane chain was preferred for most of the ligands, both free and when bound to the receptor, and that for all of the receptor‐ligand complexes, the substituent was located inside or partly inside of the aromatic cavity of the receptor. We estimated the free binding energy of a methyl‐ and a phenyl group to an aromatic cavity, via CH‐π, and combined aromatic CH‐π and π‐π interactions to be −1.7 and −3.3 kJ mol⁻¹, respectively. The experimental results were used to assess the accuracy of different computational methods, including molecular mechanics (MM) and density functional theory (DFT) methods, showing that MM was superior.

Protein-Peptide Binding Site Detection Using 3D Convolutional Neural Networks

Peptides and peptide-based molecules represent a promising therapeutic modality targeting intracellular protein-protein interactions, potentially combining the beneficial properties of biologics and small-molecule drugs. Protein-peptide complexes occupy a unique niche of interaction interfaces with respect to protein-protein and protein-small molecule complexes. Protein-peptide binding site identification resembles image object detection, a field that had been revolutionalized with computer vision techniques. We present a new protein-peptide binding site detection method called BiteNet_Pp by harnessing the power of 3D convolutional neural network. Our method employs a tensor-based representation of spatial protein structures, which is fed to 3D convolutional neural network, resulting in probability scores and coordinates of the binding "hot spots" in the input structures. We used the domain adaptation technique to fine-tune model trained on protein-small molecule complexes using a manually curated set of protein-peptide structures. BiteNet_Pp consistently outperforms existing state-of-the-art methods in the independent test benchmark. It takes less than a second to analyze a single-protein structure, making BiteNet_Pp suitable for the large-scale analysis of protein-peptide binding sites.

Binding characteristics of staphylococcal protein A and streptococcal protein G for fragment crystallizable portion of human immunoglobulin G

In the wide array of physiological processes, protein-protein interactions and their binding are the most basal activities for achieving adequate biological metabolism. Among the studies on binding proteins, the examination of interactions between immunoglobulin G (IgG) and natural immunoglobulin-binding ligands, such as staphylococcal protein A (spA) and streptococcal protein G (spG), is essential in the development of pharmaceutical science, biotechnology, and affinity chromatography. The widespread utilization of IgG-spA/spG binding characteristics has allowed researchers to investigate these molecular interactions. However, the detailed binding strength of each ligand and the corresponding binding mechanisms have yet to be fully investigated. In this study, the authors analyzed the binding strengths of IgG-spA and IgG-spG complexes and identified the mechanisms enabling these bindings using molecular dynamics simulation, steered molecular dynamics, and advanced Poisson-Boltzmann Solver simulations. Based on the presented data, the binding strength of the spA ligand was found to significantly exceed that of the spG ligand. To find out which non-covalent interactions or amino acid sites have a dominant role in the tight binding of these ligands, further detailed analyses of electrostatic interactions, hydrophobic bonding, and binding free energies have been performed. In investigating their binding affinity, a relatively independent and different unbinding mechanism was found in each ligand. These distinctly different mechanisms were observed to be highly correlated to the protein secondary and tertiary structures of spA and spG ligands, as explicated from the perspective of hydrogen bonding.

Predicting Protein-Protein Interactions Using Symmetric Logistic Matrix Factorization

Accurate assessment of protein-protein interactions (PPIs) is critical to deciphering disease mechanisms and developing novel drugs, and with rapidly growing PPI data, the need for more efficient predictive methods is emerging. We propose here a symmetric logistic matrix factorization (symLMF)-based approach to predict PPIs, especially useful for large PPI networks. Benchmarked against two widely used datasets (Saccharomyces cerevisiae and Homo sapiens benchmarks) and their extended versions, the symLMF-based method proves to outperform most of the state-of-the-art data-driven methods applied to human PPIs, and it shows a performance comparable to those of deep learning methods despite its conceptual and technical simplicity and efficiency. Tests performed on humans, yeast, and tissue (brain and liver)- and disease (neurodegenerative and metabolic disorders)-specific datasets further demonstrate the high capability to capture the hidden interactions. Notably, many "de novo predictions" made by symLMF are verified to exist in PPI databases other than those used for training/testing the method, indicating that the method could be of broad utility as a simple, yet efficient and accurate, tool applicable to PPI datasets.

Comprehensive Prediction of Molecular Recognition in a Combinatorial Chemical Space Using Machine Learning

In combinatorial chemical approaches, optimizing the composition and arrangement of building blocks toward a particular function has been done using a number of methods, including high throughput molecular screening, molecular evolution, and computational prescreening. Here, a different approach is considered that uses sparse measurements of library molecules as the input to a machine learning algorithm which generates a comprehensive, quantitative relationship between covalent molecular structure and function that can then be used to predict the function of any molecule in the possible combinatorial space. To test the feasibility of the approach, a defined combinatorial chemical space consisting of ∼10¹² possible linear combinations of 16 different amino acids was used. The binding of a very sparse, but nearly random, sampling of this amino acid sequence space to 9 different protein targets is measured and used to generate a general relationship between peptide sequence and binding for each target. Surprisingly, measuring as little as a few hundred to a few thousand of the ∼10¹² possible molecules provides sufficient training to be highly predictive of the binding of the remaining molecules in the combinatorial space. Furthermore, measuring only amino acid sequences that bind weakly to a target allows the accurate prediction of which sequences will bind 10-100 times more strongly. Thus, the molecular recognition information contained in a tiny fraction of molecules in this combinatorial space is sufficient to characterize any set of molecules randomly selected from the entire space, a fact that potentially has significant implications for the design of new chemical function using combinatorial chemical libraries.

InterPep2: global peptide-protein docking using interaction surface templates

Motivation

Interactions between proteins and peptides or peptide-like intrinsically disordered regions are involved in many important biological processes, such as gene expression and cell life-cycle regulation. Experimentally determining the structure of such interactions is time-consuming and difficult because of the inherent flexibility of the peptide ligand. Although several prediction-methods exist, most are limited in performance or availability.

Results

InterPep2 is a freely available method for predicting the structure of peptide–protein interactions. Improved performance is obtained by using templates from both peptide–protein and regular protein–protein interactions, and by a random forest trained to predict the DockQ-score for a given template using sequence and structural features. When tested on 252 bound peptide–protein complexes from structures deposited after the complexes used in the construction of the training and templates sets of InterPep2, InterPep2-Refined correctly positioned 67 peptides within 4.0 Å LRMSD among top10, similar to another state-of-the-art template-based method which positioned 54 peptides correctly. However, InterPep2 displays a superior ability to evaluate the quality of its own predictions. On a previously established set of 27 non-redundant unbound-to-bound peptide–protein complexes, InterPep2 performs on-par with leading methods. The extended InterPep2-Refined protocol managed to correctly model 15 of these complexes within 4.0 Å LRMSD among top10, without using templates from homologs. In addition, combining the template-based predictions from InterPep2 with ab initio predictions from PIPER-FlexPepDock resulted in 22% more near-native predictions compared to the best single method (22 versus 18).

Availability and implementation

The program is available from: http://wallnerlab.org/InterPep2.

Supplementary information

Supplementary data are available at Bioinformatics online.