Interaction Of Peptides Research Articles

Insights into Molecular Interactions between a GAPDH-Related Fish Antimicrobial Peptide, Analogs Thereof, and Bacterial Membranes.

Interactions between SJGAP (skipjack tuna GAPDH-related antimicrobial peptide) and four analogs thereof with model bacterial membranes were studied using Fourier-transform infrared spectroscopy (FTIR) and molecular dynamics (MD) simulations. MD trajectory analyses showed that the N-terminal segment of the peptide analogs has many contacts with the polar heads of membrane phospholipids, while the central α helix interacts strongly with the hydrophobic core of the membranes. The peptides also had a marked influence on the wave numbers associated with the phase transition of phospholipids organized as liposomes in both the interface and aliphatic chain regions of the infrared spectra, supporting the interactions observed in the MD trajectories. In addition, interesting links were found between peptide interactions with the aliphatic chains of membrane phospholipids, as determined by FTIR and from the MD trajectories, and the membrane permeabilization capacity of these peptide analogs, as previously demonstrated. To summarize, the combined experimental and computational efforts have provided insights into crucial aspects of the interactions between the investigated peptides and bacterial membranes. This work thus makes an original contribution to our understanding of the molecular interactions underlying the antimicrobial activity of these GAPDH-related antimicrobial peptides from Scombridae.

ProtTrans and multi-window scanning convolutional neural networks for the prediction of protein-peptide interaction sites

This study delves into the prediction of protein-peptide interactions using advanced machine learning techniques, comparing models such as sequence-based, standard CNNs, and traditional classifiers. Leveraging pre-trained language models and multi-view window scanning CNNs, our approach yields significant improvements, with ProtTrans standing out based on 2.1 billion protein sequences and 393 billion amino acids. The integrated model demonstrates remarkable performance, achieving an AUC of 0.856 and 0.823 on the PepBCL Set_1 and Set_2 datasets, respectively. Additionally, it attains a Precision of 0.564 in PepBCL Set 1 and 0.527 in PepBCL Set 2, surpassing the performance of previous methods. Beyond this, we explore the application of this model in cancer therapy, particularly in identifying peptide interactions for selective targeting of cancer cells, and other fields. The findings of this study contribute to bioinformatics, providing valuable insights for drug discovery and therapeutic development.

Unlocking the specificity of antimicrobial peptide interactions for membrane-targeted therapies

Antimicrobial peptides (AMPs) are increasingly recognized as potent therapeutic agents, with their selective affinity for pathological membranes, low toxicity profile, and minimal resistance development making them particularly attractive in the pharmaceutical landscape. This study offers a comprehensive analysis of the interaction between specific AMPs, including magainin-2, pleurocidin, CM15, LL37, and clavanin, with lipid bilayer models of very different compositions that have been ordinarily used as biological membrane models of healthy mammal, cancerous, and bacterial cells. Employing unbiased molecular dynamics simulations and metadynamics techniques, we have deciphered the intricate mechanisms by which these peptides recognize pathogenic and pathologic lipid patterns and integrate into lipid assemblies. Our findings reveal that the transverse component of the peptide's hydrophobic dipole moment is critical for membrane interaction, decisively influencing the molecule's orientation and expected therapeutic efficacy. Our approach also provides insight on the kinetic and dynamic dependence on the peptide orientation in the axial and azimuthal angles when coming close to the membrane. The aim is to establish a robust framework for the rational design of peptide-based, membrane-targeted therapies, as well as effective quantitative descriptors that can facilitate the automated design of novel AMPs for these therapies using machine learning methods.

HighFold: accurately predicting structures of cyclic peptides and complexes with head-to-tail and disulfide bridge constraints.

In recent years, cyclic peptides have emerged as a promising therapeutic modality due to their diverse biological activities. Understanding the structures of these cyclic peptides and their complexes is crucial for unlocking invaluable insights about protein target-cyclic peptide interaction, which can facilitate the development of novel-related drugs. However, conducting experimental observations is time-consuming and expensive. Computer-aided drug design methods are not practical enough in real-world applications. To tackles this challenge, we introduce HighFold, an AlphaFold-derived model in this study. By integrating specific details about the head-to-tail circle and disulfide bridge structures, the HighFold model can accurately predict the structures of cyclic peptides and their complexes. Our model demonstrates superior predictive performance compared to other existing approaches, representing a significant advancement in structure-activity research. The HighFold model is openly accessible at https://github.com/hongliangduan/HighFold.

Exploring a Solvent Dependent Strategy to Control Self-Assembling Behavior and Cellular Interaction in Laminin-Mimetic Short Peptide based Supramolecular Hydrogels.

Self-assembled hydrogels, fabricated through diverse non-covalent interactions, have been extensively studied in regenerative medicines. Inspired from bioactive functional motifs of ECM protein, short peptide sequences have shown remarkable abilities to replicate the intrinsic features of the natural extracellular milieu. In this direction, we have fabricated two short hydrophobic bioactive sequences derived from the laminin protein i. e., IKVAV and YIGSR. Based on the substantial hydrophobicity of these peptides, we selected a co-solvent approach as a suitable gelation technique that included different concentrations of DMSO as an organic phase along with an aqueous solution containing 0.1 % TFA. These hydrophobic laminin-based bioactive peptides with limited solubility in aqueous physiological environment showed significantly enhanced solubility with higher DMSO content in water. The enhanced solubility resulted in extensive intermolecular interactions that led to the formation of hydrogels with a higher-order entangled network along with improved mechanical properties. Interestingly, by simply modulating DMSO content, highly tunable gels were accessed in the same gelator domain that displayed differential physicochemical properties. Further, the cellular studies substantiated the potential of these laminin-derived hydrogels in enhancing cell-matrix interactions, thereby reinforcing their applications in tissue engineering.

From Molecular Dynamics to Taste Sensory Perception: A Comprehensive Study on the Interaction of Umami Peptides with the T1R1/T1R3-VFT Receptor.

The research on the umami receptor-ligand interaction is crucial for understanding umami perception. This study integrated molecular simulations, sensory evaluation, and biosensor technology to analyze the interaction between umami peptides and the umami receptor T1R1/T1R3-VFT. Molecular dynamics simulations were used to investigate the dissociation process of seven umami peptides with the umami receptor T1R1/T1R3-VFT, and by calculating the potential mean force curve using the Jarzynski equation, it was found that the binding free energy of umami peptide is between -58.80 and -12.17 kcal/mol, which had a strong correlation with the umami intensity obtained by time intensity sensory evaluation. Through correlation analysis, the dissociation rate constants (0.0126-0.394 1/s) of umami peptides were found to have a great impact on umami perception. The faster the dissociation rate of umami peptides from receptors, the stronger the perceived intensity of the umami taste. This research aims to elucidate the relationship between the umami peptide-receptor interaction and umami perception, providing theoretical support for the exploration of umami perception mechanisms.

Construction of Reconfigurable and Polymorphic DNA Origami Assemblies with Coiled-Coil Patches and Patterns.

DNA origami nanodevices achieve programmable structure and tunable mechanical and dynamic properties by leveraging the sequence-specific interactions of nucleic acids. Previous advances have also established DNA origami as a useful building block to make well-defined micron-scale structures through hierarchical self-assembly, but these efforts have largely leveraged the structural features of DNA origami. The tunable dynamic and mechanical properties also provide an opportunity to make assemblies with adaptive structures and properties. Here the integration of DNA origami hinge nanodevices and coiled-coil peptides are reported into hybrid reconfigurable assemblies. With the same dynamic device and peptide interaction, it is made multiple higher-order assemblies (i.e., polymorphic assembly) by organizing clusters of peptides into patches or arranging single peptides into patterns on the surfaces of DNA origami to control the relative orientation of devices. The coiled-coil interactions are used to construct circular and linear assemblies whose structure and mechanical properties can be modulated with DNA-based reconfiguration. Reconfiguration of linear assemblies leads to micron scale motions and ≈2.5-10-fold increase in bending stiffness. The results provide a foundation for stimulus-responsive hybrid assemblies that can adapt their structure and properties in response to nucleic acid, peptide, protein, or other triggers.

Molecular investigation of antimicrobial peptides against Helicobacter pylori proteins using a peptide-protein docking approach

The impact of H. pylori resistance on patient's treatment failure is a major concern. Therefore, the development of novel or alternative therapies for H. pylori is urgently needed. The purpose of this study was to investigate the molecular interactions of various antimicrobial peptides (AMPs) to H. pylori proteins. We performed the peptide-protein molecular docking using HADDOCK 2.4 webserver. Fourteen AMPs were tested for their binding efficacy against four H. pylori proteins. Simulation of the peptide-protein complex was performed using molecular dynamic software package AMBER20. From molecular docking analysis, five peptides (LL-37, Tilapia piscidin 4, napin, snakin-1 and EcAMP1) showed strong binding interactions against H. pylori proteins. The strongest binding affinity was observed in the interactions between Snakin-1 and PBP2, TP4 and type I HopQ and EcAMP1 and type I HopQ with −11.1, −13.6 and −13.8 kcal/mol, respectively. The dynamic simulation was performed for two complexes (snakin1-PBP2 and EcAMP1-HopQ). Results of the dynamics simulation showed that EcAMP1 had stable interaction and binding to type I HopQ protein without significant structural changes. In conclusion, both results of docking and simulation showed that EcAMP1 might be useful as a potential therapeutic agent for H. pylori treatment. This molecular approach provides deep understanding of the interaction insights between AMPs and H. pylori proteins. It paves the way for the development of novel anti-H. pylori using antimicrobial peptides.

Bee Venom: Composition and Anticancer Properties.

Among the various natural compounds used in alternative and Oriental medicine, toxins isolated from different organisms have had their application for many years, and Apis mellifera venom has been studied the most extensively. Numerous studies dealing with the positive assets of bee venom (BV) indicated its beneficial properties. The usage of bee products to prevent the occurrence of diseases and for their treatment is often referred to as apitherapy and is based mainly on the experience of the traditional system of medical practice in diverse ethnic communities. Today, a large number of studies are focused on the antitumor effects of BV, which are mainly attributed to its basic polypeptide melittin (MEL). Previous studies have indicated that BV and its major constituent MEL cause a strong toxic effect on different cancer cells, such as liver, lung, bladder, kidney, prostate, breast, and leukemia cells, while a less pronounced effect was observed in normal non-target cells. Their proposed mechanisms of action, such as the effect on proliferation and growth inhibition, cell cycle alterations, and induction of cell death through several cancer cell death mechanisms, are associated with the activation of phospholipase A2 (PLA2), caspases, and matrix metalloproteinases that destroy cancer cells. Numerous cellular effects of BV and MEL need to be elucidated on the molecular level, while the key issue has to do with the trigger of the apoptotic cascade. Apoptosis could be either a consequence of the plasmatic membrane fenestration or the result of the direct interaction of the BV components with pro-apoptotic and anti-apoptotic factors. The interaction of BV peptides and enzymes with the plasma membrane is a crucial step in the whole process. However, before its possible application as a remedy, it is crucial to identify the correct route of exposure and dosage of BV and MEL for potential therapeutic use as well as potential side effects on normal cells and tissues to avoid any possible adverse event.

Interaction of KLAKLAK-NH2 and Analogs with Biomimetic Membrane Models.

Specifically designed peptide mimetics offer higher selectivity regarding their toxicity to mammalian cells. In addition to the α-helix conformation, the specific activity is related to the peptide's ability to penetrate the cell membrane. The alterations in lipid membrane properties were addressed in the presence of the peptide KLAKLAK-NH2 and analogs containing β-alanine, strengthening the antibacterial activity and/or naphtalimide with proven anticancer properties. The molecular interactions of the peptide mimetics with POPC bilayers were studied using FTIR-ATR spectroscopy. The thermal shape fluctuation analysis of quasispherical unilamellar vesicles was applied to probe the membrane bending elasticity. The impedance characteristics of bilayer lipid membranes were measured using fast Fourier-transform electrochemical impedance spectroscopy. A lateral peptide association with the membrane is reported for β-alanine-containing peptides. The most pronounced membrane softening is found for the NphtG-KLβAKLβAK-NH2 analog containing both active groups that corroborate with the indications for 1,8-naphthalimide penetration in the lipid hydrophobic area obtained from the FTIR-ATR spectra analysis. The β-alanine substitution induces strong membrane-rigidifying properties even at very low concentrations of both β-alanine-containing peptides. The reported results are expected to advance the progress in tailoring the pharmacokinetic properties of antimicrobial peptides with strengthened stability towards enzymatic degradation. The investigation of the nonspecific interactions of peptides with model lipid membranes is featured as a useful tool to assess the antitumor and antimicrobial potential of new peptide mimetics.

Interaction of antimicrobial peptides with model membranes: a perspective towards new antibiotics

Interaction of antimicrobial peptides with model membranes: a perspective towards new antibiotics

Structure and ligand binding in the putative anti-microbial peptide transporter protein, YejA.

YejABEF is an ATP-binding cassette transporter that is implicated in the sensitivity of Escherichia coli to anti-microbial peptides, the best-characterized example being microcin C, a peptide-nucleotide antibiotic that targets aspartyl-tRNA synthetase. Here the structure of the extracellular solute binding protein, YejA, has been determined, revealing an oligopeptide-binding protein fold enclosing a ligand-binding pocket larger than those of other peptide-binding proteins of known structure. Prominent electron density in this cavity defines an undecapeptide sequence LGEPRYAFNFN, an observation that is confirmed by mass spectrometry. In the structure, the peptide interactions with the protein are mediated by main chain hydrogen bonds with the exception of Arg5 whose guanidinium side chain makes a set of defining polar interactions with four YejA residues. More detailed characterization of purified recombinant YejA, by a combination of ESI and MALDI-mass spectrometry as well as thermal shift assays, reveals a set of YejA complexes containing overlapping peptides 10-19 residues in length. All contain the sequence LGEPRYAFN. Curiously, these peptides correspond to residues 8-26 of the mature YejA protein, which belong to a unique N-terminal extension that distinguishes YejA from other cluster C oligopeptide binding proteins of known structure. This 35-residue extension is well-ordered and packs across the surface of the protein. The undecapeptide ligand occupies only a fraction of the enclosed pocket volume suggesting the possibility that much larger peptides or peptide conjugates could be accommodated, though thermal shift assays of YejA binding to antimicrobial peptides and peptides unrelated to LGEPRYAFNFN have not provided evidence of binding. While the physiological significance of this 'auto-binding' is not clear, the experimental data suggest that it is not an artefact of the crystallization process and that it may have a function in the sensing of periplasmic or membrane stress.

Characterization of antimicrobial peptide interactions with a realistically phase separated model of the E. coli outer membrane via large-scale simulations

Characterization of antimicrobial peptide interactions with a realistically phase separated model of the E. coli outer membrane via large-scale simulations

Computational exploration of the genomic assignments, molecular structure, and dynamics of the ccdABXn 2 toxin-antitoxin homolog with its bacterial target, the DNA gyrase, in the entomopathogen Xenorhabdus nematophila

Toxin-antitoxin (TA) modules, initially discovered on bacterial plasmids and subsequently identified within chromosomal contexts, hold a pivotal role in the realm of bacterial physiology. Among these, the pioneering TA system, ccd (Control of Cell Death), primarily localized on the F-plasmid, is known for its orchestration of plasmid replication with cellular division. Nonetheless, the precise functions of such systems within bacterial chromosomal settings remain a compelling subject that demands deeper investigation. To bridge this knowledge gap, our study focuses on exploring ccdABXn2 , a chromosomally encoded TA module originating from the entomopathogenic bacterium Xenorhabdus nematophila. We meticulously delved into the system’s genomic assignments, structural attributes, and functional interplay. Our findings uncovered intriguing patterns—CcdB toxin homologs exhibited higher conservation levels compared to their CcdA antitoxin counterparts. Moreover, we constructed secondary as well as tertiary models for both the CcdB toxin and CcdA antitoxin using threading techniques and subsequently validated their structural integrity. Our exploration extended to the identification of key interactions, including the peptide interaction with gyrase for the CcdB homolog and CcdB toxin interactions for the CcdA homolog, highlighting the intricate TA interaction network. Through docking and simulation analyses, we unequivocally demonstrated the inhibition of replication via binding the CcdB toxin to its target, DNA gyrase. These insights provide valuable knowledge about the metabolic and physiological roles of the chromosomally encoded ccdABXn 2 TA module within the context of X. nematophila, significantly enhancing our comprehension of its functional significance within the intricate ecosystem of the bacterial host. Communicated by Ramaswamy H. Sarma

Emerging role of carbonyl-carbonyl interactions in the classification of beta turns.

Carbonyl-carbonyl interactions in peptides and proteins attracted considerable interest in recent years. Here, we report a survey of carbonyl-carbonyl interactions in cyclic peptides, depsipeptides, peptoids and discuss the relationship between backbone torsion angles and CO∙∙∙CO distances. In general, φ values in the range between -40° and -90° and between 40° and 90° correspond to CO∙∙∙CO distances below 3.22 Å. By extending the analysis of carbonyl-carbonyl interactions in different types of beta turns in proteins, we also highlight the role of direct or reciprocal carbonyl-carbonyl interactions in stabilizing the beta turn conformation for each specific type. We confirmed the new type II beta turn, detected by Dunbrack and coworkers, and named Pa, and detect the presence of a direct carbonyl-carbonyl interaction between the second and third residues of the turn. We also evidenced the existence of another new type II beta turn, named pA (following Dunbrack's notation), which represents the alternative conformation of Pa with opposite φ and ψ values and is characterized by a direct carbonyl-carbonyl interaction between the second and third residues of the turn. Finally, we show that the occurrence of CO∙∙∙CO interactions could be also advocated to explain from a chemical point of view the diversity of turn types.

Structural insights into the unique recognition module between α-synuclein peptide and nanobody.

Nanobodies are single-domain fragments of antibodies with comparable specificity and affinity to antibodies. They are emerging as versatile tools in biology due to their relatively small size. Here, we report the crystal structure of a specific nanobody Nbα-syn01, bound to a 14 amino acid long peptide of α-synuclein (αSyn), a 140-residue protein whose aggregation is associated with Parkinson's disease. The complex structure exhibits a unique binding pattern where the αSyn peptide replaces the N-terminal region of nanobody. Recognition is mediated principally by extended main chain interaction of the αSyn peptide and specificity of the interaction lies in the central 48-52 region of αSyn peptide. Structure-guided truncation of Nbα-syn01 shows tighter binding to αSyn peptide and improved inhibition of α-synuclein aggregation. The structure of the truncated complex was subsequently determined and was indistinguishable to full length complex as the full-length form had no visible electron density for the N-terminal end. These findings reveal the molecular basis for a previously unobserved binding mode for nanobody recognition of α-synuclein, providing an explanation for the enhanced binding, and potential for an alternate framework for structure-based protein engineering of nanobodies to develop better diagnostic and therapeutic tools.

Interplay of Angiotensin Peptides, Vasopressin, and Insulin in the Heart: Experimental and Clinical Evidence of Altered Interactions in Obesity and Diabetes Mellitus.

The present review draws attention to the specific role of angiotensin peptides [angiotensin II (Ang II), angiotensin-(1-7) (Ang-(1-7)], vasopressin (AVP), and insulin in the regulation of the coronary blood flow and cardiac contractions. The interactions of angiotensin peptides, AVP, and insulin in the heart and in the brain are also discussed. The intracardiac production and the supply of angiotensin peptides and AVP from the systemic circulation enable their easy access to the coronary vessels and the cardiomyocytes. Coronary vessels and cardiomyocytes are furnished with AT1 receptors, AT2 receptors, Ang (1-7) receptors, vasopressin V1 receptors, and insulin receptor substrates. The presence of some of these molecules in the same cells creates good conditions for their interaction at the signaling level. The broad spectrum of actions allows for the engagement of angiotensin peptides, AVP, and insulin in the regulation of the most vital cardiac processes, including (1) cardiac tissue oxygenation, energy production, and metabolism; (2) the generation of the other cardiovascular compounds, such as nitric oxide, bradykinin (Bk), and endothelin; and (3) the regulation of cardiac work by the autonomic nervous system and the cardiovascular neurons of the brain. Multiple experimental studies and clinical observations show that the interactions of Ang II, Ang(1-7), AVP, and insulin in the heart and in the brain are markedly altered during heart failure, hypertension, obesity, and diabetes mellitus, especially when these diseases coexist. A survey of the literature presented in the review provides evidence for the belief that very individualized treatment, including interactions of angiotensins and vasopressin with insulin, should be applied in patients suffering from both the cardiovascular and metabolic diseases.

Functional Classification and Interaction Selectivity Landscape of the Human SH3 Domain Superfamily.

SRC homology 3 (SH3) domains are critical interaction modules that orchestrate the assembly of protein complexes involved in diverse biological processes. They facilitate transient protein-protein interactions by selectively interacting with proline-rich motifs (PRMs). A database search revealed 298 SH3 domains in 221 human proteins. Multiple sequence alignment of human SH3 domains is useful for phylogenetic analysis and determination of their selectivity towards PRM-containing peptides (PRPs). However, a more precise functional classification of SH3 domains is achieved by constructing a phylogenetic tree only from PRM-binding residues and using existing SH3 domain-PRP structures and biochemical data to determine the specificity within each of the 10 families for particular PRPs. In addition, the C-terminal proline-rich domain of the RAS activator SOS1 covers 13 of the 14 recognized proline-rich consensus sequence motifs, encompassing differential PRP pattern selectivity among all SH3 families. To evaluate the binding capabilities and affinities, we conducted fluorescence dot blot and polarization experiments using 25 representative SH3 domains and various PRPs derived from SOS1. Our analysis has identified 45 interacting pairs, with binding affinities ranging from 0.2 to 125 micromolar, out of 300 tested and potential new SH3 domain-SOS1 interactions. Furthermore, it establishes a framework to bridge the gap between SH3 and PRP interactions and provides predictive insights into the potential interactions of SH3 domains with PRMs based on sequence specifications. This novel framework has the potential to enhance the understanding of protein networks mediated by SH3 domain-PRM interactions and be utilized as a general approach for other domain-peptide interactions.

PH-dependent interactions of coacervate-forming histidine-rich peptide with model lipid membranes

Peptide-based liquid droplets (coacervates) produced by spontaneous liquid-liquid phase separation (LLPS), have emerged as a promising class of drug delivery systems due to their high entrapping efficiency and the simplicity of their formulation. However, the detailed mechanisms governing their interaction with cell membranes and cellular uptake remain poorly understood. In this study, we investigated the interactions of peptide coacervates composed of HBpep—peptide derived from the histidine-rich beak proteins (HBPs) of the Humboldt squid—with model cellular membranes in the form of supported lipid bilayers (SLBs). We employed quartz crystal microbalance with dissipation monitoring (QCM-D), neutron reflectometry (NR) and atomistic molecular dynamics (MD) simulations to reveal the nature of these interactions in the absence of fluorescent labels or tags. HBpep forms small oligomers at pH 6 whereas it forms µm-sized coacervates at physiological pH. Our findings reveal that both HBpep oligomers and HBpep-coacervates adsorb onto SLBs at pH 6 and 7.4, respectively. At pH 6, when the peptide carries a net positive charge, HBpep oligomers insert into the SLB, facilitated by the peptide’s interactions with the charged lipids and cholesterol. Importantly, however, HBpep coacervate adsorption at physiological pH, when it is largely uncharged, is fully reversible, suggesting no significant lipid bilayer rearrangement. HBpep coacervates, previously identified as efficient drug delivery vehicles, do not interact with the lipid membrane in the same manner as traditional cationic drug delivery systems or cell-penetrating peptides. Based on our findings, HBpep coacervates at physiological pH cannot cross the cell membrane by a simple passive mechanism and are thus likely to adopt a non-canonical cell entry pathway.

Using explainable machine learning to uncover the kinase-substrate interaction landscape.

Phosphorylation, a post-translational modification regulated by protein kinase enzymes, plays an essential role in almost all cellular processes. Understanding how each of the nearly 500 human protein kinases selectively phosphorylates their substrates is a foundational challenge in bioinformatics and cell signaling. Although deep learning models have been a popular means to predict kinase-substrate relationships, existing models often lack interpretability and are trained on datasets skewed toward a subset of well-studied kinases. Here we leverage recent peptide library datasets generated to determine substrate specificity profiles of 300 serine/threonine kinases to develop an explainable Transformer model for kinase-peptide interaction prediction. The model, trained solely on primary sequences, achieved state-of-the-art performance. Its unique multitask learning paradigm built within the model enables predictions on virtually any kinase-peptide pair, including predictions on 139 kinases not used in peptide library screens. Furthermore, we employed explainable machine learning methods to elucidate the model's inner workings. Through analysis of learned embeddings at different training stages, we demonstrate that the model employs a unique strategy of substrate prediction considering both substrate motif patterns and kinase evolutionary features. SHapley Additive exPlanation (SHAP) analysis reveals key specificity determining residues in the peptide sequence. Finally, we provide a web interface for predicting kinase-substrate associations for user-defined sequences and a resource for visualizing the learned kinase-substrate associations. All code and data are available at https://github.com/esbgkannan/Phosformer-ST. Web server is available at https://phosformer.netlify.app.