Exceptional stability of ultrasmall cubic copper metal nanoclusters - a molecular dynamics study.

The fabrication of shape-selective coinage metal nanoclusters (MNCs) has promising applications due to their exceptional physical and chemical molecule-like properties. However, the stability of the specific geometry of the nanoclusters, such as their cubic shapes, is unclear and has been unraveled by assessing the nanoclusters' interactions with different environments. In this work, we investigate the morphological stability of cubic structured, coinage metal nanoclusters of varying sizes ranging from 14 to 1099 atoms. The impact of solvent environments like water and the presence of ionic liquids (IL) on the stabilization of the MNCs were assessed using molecular dynamics (MD) simulations. In general, smaller MNCs composed of less than 256 atoms encountered structural distortion easily compared to the larger ones, which preserved their cubic morphology with minimal surface aberrations in water. However, in the presence of 4M 1-butyl-1,1,1-trimethyl ammonium methane sulfonate [N1114][C1SO3] IL solution, the overall cubic shape of the MNCs was successfully preserved. Strikingly, it is observed that in contrast to the noble MNCs like Au and Ag, the cubic morphology for Cu MNCs with sizes less than 256 atoms exhibited significant stability even in the absence of IL.

Structural Rearrangements of Pigeon Cryptochrome 4 Undergoing a Complete Redox Cycle.

Cryptochrome is currently the major contender of a protein to underpin magnetoreception, the ability to sense the Earth's magnetic field. Among various types of cryptochromes, cryptochrome 4 has been identified as the likely magnetoreceptor in migratory birds. All-atom molecular dynamics (MD) studies have offered first insights into the structural dynamics of cryptochrome but are limited to a short time scale due to large computational demands. Here, we employ coarse-grained MD simulations to investigate the emergence of long-lived states and conformational changes in pigeon cryptochrome 4. Our coarse-grained simulations complete the picture by permitting observation on a significantly longer time scale. We observe conformational transitions in the phosphate-binding loop of pigeon cryptochrome 4 upon activation and identify prominent motions in residues 440-460, suggesting a possible role as a signaling state of the protein or as a gated interaction site for forming protein complexes that might facilitate downstream processes. The findings highlight the importance of considering longer time scales in studying cryptochrome dynamics and magnetoreception. Coarse-grained MD simulations offer a valuable tool to unravel the complex behavior of cryptochrome proteins and shed new light on the mechanisms underlying their role in magnetoreception. Further exploration of these conformational changes and their functional implications may contribute to a deeper understanding of the molecular mechanisms of magnetoreception in birds.

Studying the Collective Functional Response of a Receptor in Alchemical Ligand Binding Free Energy Simulations with Accelerated Solvation Layer Dynamics.

Ligand binding free energy simulations (LB-FES) that involve sampling of protein functional conformations have been longstanding challenges in research on molecular recognition. Particularly, modeling of the conformational transition pathway and design of the heuristic biasing mechanism are severe bottlenecks for the existing enhanced configurational sampling (ECS) methods. Inspired by the key role of hydration in regulating conformational dynamics of macromolecules, this report proposes a novel ECS approach that facilitates binding-associated structural dynamics by accelerated hydration transitions in combination with the λ-exchange of free energy perturbation (FEP). Two challenging protein-ligand binding processes involving large configurational transitions of the receptor are studied, with hydration transitions at binding sites accelerated by Hamiltonian-simulated annealing of the hydration layer. Without the need for pathway analysis or ad hoc barrier flattening potential, LB-FES were performed with FEP/λ-exchange molecular dynamics simulation at a minor overhead for annealing of the hydration layer. The LB-FES studies showed that the accelerated rehydration significantly enhances the collective conformational transitions of the receptor, and convergence of binding affinity calculations is obtained at a sweet-spot simulation time scale. Alchemical LB-FES with the proposed ECS strategy is free from the effort of trial and error for the setup and realizes efficient on-the-fly sampling for the collective functional response of the receptor and bound water and therefore presents a practical approach to high-throughput screening in drug discovery.

Identifying the temporal contributors and their interactions during dynamic formation of black tea cream.

Tea cream formed in hot and strong tea infusion while cooling deteriorates quality and health benefits of tea. However, the interactions among temporal contributors during dynamic formation of tea cream are still elusive. Here, by deletional recombination experiments and molecular dynamics simulation, it was found that proteins, caffeine (CAF), and phenolics played a dominant role throughout the cream formation, and the contribution of amino acids was highlighted in the early stage. Furthermore, CAF was prominent due to its extensive binding capacity and the filling complex voids property, and caffeine-theaflavins (TFs) complexation may be the core skeleton of the growing particles in black tea infusion. In addition to TFs, the unidentified phenolic oxidation-derived products (PODP) were confirmed to contribute greatly to the cream formation.

The antimicrobial fibupeptide lugdunin forms water-filled channel structures in lipid membranes.

Recently, a novel cyclo-heptapeptide composed of alternating D,L-amino acids and a unique thiazolidine heterocycle, called lugdunin, was discovered, which is produced by the nasal and skin commensal Staphylococcus lugdunensis. Lugdunin displays potent antimicrobial activity against a broad spectrum of Gram-positive bacteria, including challenging-to-treat methicillin-resistant Staphylococcus aureus (MRSA). Lugdunin specifically inhibits target bacteria by dissipating their membrane potential. However, the precise mode of action of this new class of fibupeptides remains largely elusive. Here, we disclose the mechanism by which lugdunin rapidly destabilizes the bacterial membrane potential using an in vitro approach. The peptide strongly partitions into lipid compositions resembling Gram-positive bacterial membranes but less in those harboring the eukaryotic membrane component cholesterol. Upon insertion, lugdunin forms hydrogen-bonded antiparallel ß-sheets by the formation of peptide nanotubes, as demonstrated by ATR-FTIR spectroscopy and molecular dynamics simulations. These hydrophilic nanotubes filled with a water wire facilitate not only the translocation of protons but also of monovalent cations as demonstrated by voltage-clamp experiments on black lipid membranes. Collectively, our results provide evidence that the natural fibupeptide lugdunin acts as a peptidic channel that is spontaneously formed by an intricate stacking mechanism, leading to the dissipation of a bacterial cell's membrane potential.

Mechanisms of the Drug Penetration Enhancer Propylene Glycol Interacting with Skin Lipid Membranes.

Very few drugs have the necessary physicochemical properties to cross the skin's main permeability barrier, the stratum corneum (SC), in sufficient amounts. Propylene glycol (PG) is a chemical penetration enhancer that could be included in topical formulations in order to overcome the barrier properties of the skin and facilitate the transport of drugs across it. Experiments have demonstrated that PG increases the mobility and disorder of SC lipids and may extract cholesterol from the SC, but little is known about the molecular mechanisms of drug permeation enhancement by PG. In this work, we have performed molecular dynamics (MD) simulations to investigate the molecular-level effects of PG on the structure and properties of model SC lipid bilayers. The model bilayers were simulated in the presence of PG concentrations over the range of 0-100% w/w PG, using both an all-atom and a united atom force field. PG was found to localize in the hydrophilic headgroup regions at the bilayer interface, to occupy the lipid-water hydrogen-bonding sites, and to slightly increase lipid tail disorder in a concentration-dependent manner. We showed with MD simulation that PG enhances the permeation of small molecules such as water by interacting with the bilayer interface; the results of our study may be used to guide the design of formulations for transdermal drug delivery with enhanced skin permeation, as well as topical formulations and cosmetic products.

GENESIS CGDYN: large-scale coarse-grained MD simulation with dynamic load balancing for heterogeneous biomolecular systems.

Residue-level coarse-grained (CG) molecular dynamics (MD) simulation is widely used to investigate slow biological processes that involve multiple proteins, nucleic acids, and their complexes. Biomolecules in a large simulation system are distributed non-uniformly, limiting computational efficiency with conventional methods. Here, we develop a hierarchical domain decomposition scheme with dynamic load balancing for heterogeneous biomolecular systems to keep computational efficiency even after drastic changes in particle distribution. These schemes are applied to the dynamics of intrinsically disordered protein (IDP) droplets. During the fusion of two droplets, we find that the changes in droplet shape correlate with the mixing of IDP chains. Additionally, we simulate large systems with multiple IDP droplets, achieving simulation sizes comparable to those observed in microscopy. In our MD simulations, we directly observe Ostwald ripening, a phenomenon where small droplets dissolve and their molecules redeposit into larger droplets. These methods have been implemented in CGDYN of the GENESIS software, offering a tool for investigating mesoscopic biological processes using the residue-level CG models.

Dehydration of Lipid Membranes Drives Redistribution of Cholesterol Between Lateral Domains.

Cholesterol-rich lipid rafts are found to facilitate membrane fusion, central to processes like viral entry, fertilization, and neurotransmitter release. While the fusion process involves local, transient membrane dehydration, the impact of reduced hydration on cholesterol's structural organization in biological membranes remains unclear. Here, we employ confocal fluorescence microscopy and atomistic molecular dynamics simulations to investigate cholesterol behavior in phase-separated lipid bilayers under controlled hydration. We unveiled that dehydration prompts cholesterol release from raft-like domains into the surrounding fluid phase. Unsaturated phospholipids undergo more significant dehydration-induced structural changes and lose more hydrogen bonds with water than sphingomyelin. The results suggest that cholesterol redistribution is driven by the equalization of biophysical properties between phases and the need to satisfy lipid hydrogen bonds. This underscores the role of cholesterol-phospholipid-water interplay in governing cholesterol affinity for a specific lipid type, providing a new perspective on the regulatory role of cell membrane heterogeneity during membrane fusion.

PaCS-Toolkit: Optimized Software Utilities for Parallel Cascade Selection Molecular Dynamics (PaCS-MD) Simulations and Subsequent Analyses.

Parallel cascade selection molecular dynamics (PaCS-MD) is an enhanced conformational sampling method conducted as a "repetition of time leaps in parallel worlds", comprising cycles of multiple molecular dynamics (MD) simulations performed in parallel and selection of the initial structures of MDs for the next cycle. We developed PaCS-Toolkit, an optimized software utility enabling the use of different MD software and trajectory analysis tools to facilitate the execution of the PaCS-MD simulation and analyze the obtained trajectories, including the preparation for the subsequent construction of the Markov state model. PaCS-Toolkit is coded with Python, is compatible with various computing environments, and allows for easy customization by editing the configuration file and specifying the MD software and analysis tools to be used. We present the software design of PaCS-Toolkit and demonstrate applications of PaCS-MD variations: original targeted PaCS-MD to peptide folding; rmsdPaCS-MD to protein domain motion; and dissociation PaCS-MD to ligand dissociation from adenosine A2A receptor.

Modeling Zinc Complexes Using Neural Networks.

Understanding the energetic landscapes of large molecules is necessary for the study of chemical and biological systems. Recently, deep learning has greatly accelerated the development of models based on quantum chemistry, making it possible to build potential energy surfaces and explore chemical space. However, most of this work has focused on organic molecules due to the simplicity of their electronic structures as well as the availability of data sets. In this work, we build a deep learning architecture to model the energetics of zinc organometallic complexes. To achieve this, we have compiled a configurationally and conformationally diverse data set of zinc complexes using metadynamics to overcome the limitations of traditional sampling methods. In terms of the neural network potentials, our results indicate that for zinc complexes, partial charges play an important role in modeling the long-range interactions with a neural network. Our developed model outperforms semiempirical methods in predicting the relative energy of zinc conformers, yielding a mean absolute error (MAE) of 1.32 kcal/mol with reference to the double-hybrid PWPB95 method.

High Charge Density in Peptide Dendrimers is Required to Destabilize Membranes: Insights into Endosome Evasion.

Peptide dendrimers are a type of branched, symmetric, and topologically well-defined molecule that have already been used as delivery systems for nucleic acid transfection. Several of the most promising sequences showed high efficiency in many key steps of transfection, namely, binding siRNA, entering cells, and evading the endosome. However, small changes to the peptide dendrimers, such as in the hydrophobic core, the amino acid chirality, or the total available charges, led to significantly different experimental results with unclear mechanistic insights. In this work, we built a computational model of several of those peptide dendrimers (MH18, MH13, and MH47) and some of their variants to study the molecular details of the structure and function of these molecules. We performed CpHMD simulations in the aqueous phase and in interaction with a lipid bilayer to assess how conformation and protonation are affected by pH in different environments. We found that while the different peptide dendrimer sequences lead to no substantial structural differences in the aqueous phase, the total charge and, more importantly, the total charge density are key for the capacity of the dendrimer to interact and destabilize the membrane. These dendrimers become highly charged when the pH changes from 7.5 to 4.5, and the presence of a high charge density, which is decreased for MH47 that has four fewer titratable lysines, is essential to trigger membrane destabilization. These findings are in excellent agreement with the experimental data and help us to understand the high efficiency of some dendrimers and why the dendrimer MH47 is unable to complete the transfection process. This evidence provides further understanding of the mode of action of these peptide dendrimers and will be pivotal for the future design of new sequences with improved transfection capabilities.

Molecular mechanism underlying SNARE-mediated membrane fusion enlightened by all-atom molecular dynamics simulations.

The SNAP receptor (SNARE) proteins syntaxin-1, SNAP-25, and synaptobrevin mediate neurotransmitter release by forming tight SNARE complexes that fuse synaptic vesicles with the plasma membranes in microseconds. Membrane fusion is generally explained by the action of proteins on macroscopic membrane properties such as curvature, elastic modulus, and tension, and a widespread model envisions that the SNARE motifs, juxtamembrane linkers, and C-terminal transmembrane regions of synaptobrevin and syntaxin-1 form continuous helices that act mechanically as semirigid rods, squeezing the membranes together as they assemble ("zipper") from the N to the C termini. However, the mechanism underlying fast SNARE-induced membrane fusion remains unknown. We have used all-atom molecular dynamics simulations to investigate this mechanism. Our results need to be interpreted with caution because of the limited number and length of the simulations, but they suggest a model of membrane fusion that has a natural physicochemical basis, emphasizes local molecular events over general membrane properties, and explains extensive experimental data. In this model, the central event that initiates fast (microsecond scale) membrane fusion occurs when the SNARE helices zipper into the juxtamembrane linkers which, together with the adjacent transmembrane regions, promote encounters of acyl chains from both bilayers at the polar interface. The resulting hydrophobic nucleus rapidly expands into stalk-like structures that gradually progress to form a fusion pore, aided by the SNARE transmembrane regions and without clearly discernible intermediates. The propensity of polyunsaturated lipids to participate in encounters that initiate fusion suggests that these lipids may be important for the high speed of neurotransmitter release.

Molecular dynamics of structural effects of reactive carbonyl species derivate of lipid peroxidation on bovine serum albumin.

BACKGROUND: Serum albumin is the most abundant protein in the Mammalia blood plasma at where plays a decisive role in the transport wide variety of hydrophobic ligands. BSA undergoes oxidative modifications like the carbonylation by the reactive carbonyl species (RCSs) 4-hydroxy-2-nonenal (HNE), 4 hydroxy-2-hexenal (HHE), malondialdehyde (MDA) and 4-oxo-2-nonenal (ONE), among others. The structural and functional changes induced by protein carbonylation have been associated with the advancement of neurodegenerative, cardiovascular, metabolic and cancer diseases. METHODS: To elucidate structural effects of protein carbonylation with RCSs on BSA, parameters for six new non-standard amino acids were designated and molecular dynamics simulations of its mono­carbonylated-BSA systems were conducted in the AMBER force field. Trajectories were evaluated by RMSD, RMSF, PCA, RoG and SASA analysis. RESULTS: An increase in the conformational instability for all proteins modified with local changes were observed, without significant changes on the BSA global three-dimensional folding. A more relaxed compaction level and major solvent accessible surface area for modified systems was found. Four regions of high molecular fluctuation were identified in all modified systems, being the subdomains IA and IIIB those with the most remarkable local conformational changes. Regarding essential modes of domain movements, it was evidenced that the most representatives were those related to IA subdomain, while IIIB subdomain presented discrete changes. CONCLUSIONS: RCSs induces local structural changes on mono­carbonylated BSA. Also, this study extends our knowledge on how carbonylation by RCSs induce structural effects on proteins.

Structural basis of selective TRPM7 inhibition by the anticancer agent CCT128930.

TRP channels are implicated in various diseases, but high structural similarity between them makes selective pharmacological modulation challenging. Here, we study the molecular mechanism underlying specific inhibition of the TRPM7 channel, which is essential for cancer cell proliferation, by the anticancer agent CCT128930 (CCT). Using cryo-EM, functional analysis, and MD simulations, we show that CCT binds to a vanilloid-like (VL) site, stabilizing TRPM7 in the closed non-conducting state. Similar to other allosteric inhibitors of TRPM7, NS8593 and VER155008, binding of CCT is accompanied by displacement of a lipid that resides in the VL site in the apo condition. Moreover, we demonstrate the principal role of several residues in the VL site enabling CCT to inhibit TRPM7 without impacting the homologous TRPM6 channel. Hence, our results uncover the central role of the VL site for the selective interaction of TRPM7 with small molecules that can be explored in future drug design.

Electron videography of a lipid-protein tango.

Biological phenomena, from enzymatic catalysis to synaptic transmission, originate in the structural transformations of biomolecules and biomolecular assemblies in liquid water. However, directly imaging these nanoscopic dynamics without probes or labels has been a fundamental methodological challenge. Here, we developed an approach for "electron videography"-combining liquid phase electron microscopy with molecular modeling-with which we filmed the nanoscale structural fluctuations of individual, suspended, and unlabeled membrane protein nanodiscs in liquid. Systematic comparisons with biochemical data and simulation indicate the graphene encapsulation involved can afford sufficiently reduced effects of the illuminating electron beam for these observations to yield quantitative fingerprints of nanoscale lipid-protein interactions. Our results suggest that lipid-protein interactions delineate dynamically modified membrane domains across unexpectedly long ranges. Moreover, they contribute to the molecular mechanics of the nanodisc as a whole in a manner specific to the protein within. Overall, this work illustrates an experimental approach to film, quantify, and understand biomolecular dynamics at the nanometer scale.

Statine-based peptidomimetic compounds as inhibitors for SARS-CoV-2 main protease (SARS-CoV­2 Mpro).

COVID-19 is a multisystemic disease caused by the SARS-CoV-2 airborne virus, a member of the Coronaviridae family. It has a positive sense single-stranded RNA genome and encodes two non-structural proteins through viral cysteine-proteases processing. Blocking this step is crucial to control virus replication. In this work, we reported the synthesis of 23 statine-based peptidomimetics to determine their ability to inhibit the main protease (Mpro) activity of SARS-CoV-2. Among the 23 peptidomimetics, 15 compounds effectively inhibited Mpro activity by 50% or more, while three compounds (7d, 8e, and 9g) exhibited maximum inhibition above 70% and IC50 < 1 µM. Compounds 7d, 8e, and 9g inhibited roughly 80% of SARS-CoV-2 replication and proved no cytotoxicity. Molecular docking simulations show putative hydrogen bond and hydrophobic interactions between specific amino acids and these inhibitors. Molecular dynamics simulations further confirmed the stability and persisting interactions in Mpro's subsites, exhibiting favorable free energy binding (ΔGbind) values. These findings suggest the statine-based peptidomimetics as potential therapeutic agents against SARS-CoV-2 by targeting Mpro.

PURA syndrome-causing mutations impair PUR-domain integrity and affect P-body association.

Mutations in the human PURA gene cause the neurodevelopmental PURA syndrome. In contrast to several other monogenetic disorders, almost all reported mutations in this nucleic acid-binding protein result in the full disease penetrance. In this study, we observed that patient mutations across PURA impair its previously reported co-localization with processing bodies. These mutations either destroyed the folding integrity, RNA binding, or dimerization of PURA. We also solved the crystal structures of the N- and C-terminal PUR domains of human PURA and combined them with molecular dynamics simulations and nuclear magnetic resonance measurements. The observed unusually high dynamics and structural promiscuity of PURA indicated that this protein is particularly susceptible to mutations impairing its structural integrity. It offers an explanation why even conservative mutations across PURA result in the full penetrance of symptoms in patients with PURA syndrome.

PURA syndrome is a neurodevelopmental disorder that affects about 650 patients worldwide, resulting in a range of symptoms including neurodevelopmental delays, intellectual disability, muscle weakness, seizures, and eating difficulties. The condition is caused by a mutated gene that codes for a protein called PURA. PURA binds RNA ­ the molecule that carries genetic information so it can be translated into proteins ­ and has roles in regulating the production of new proteins. Contrary to other conditions that result from mutations in a single gene, PURA syndrome patients show 'high penetrance', meaning almost every reported mutation in the gene leads to symptoms. Proske, Janowski et al. wanted to understand the molecular basis for this high penetrance. To find out more, the researchers first examined how patient mutations affected the location of the PURA in the cell, using human cells grown in the laboratory. Normally, PURA travels to P-bodies, which are groupings of RNA and proteins involved in regulating which genes get translated into proteins. The researchers found that in cells carrying PURA syndrome mutations, PURA failed to move adequately to P-bodies. To find out how this 'mislocalization' might happen, Proske, Janowski et al. tested how different mutations affected the three-dimensional folding of PURA. These analyses showed that the mutations impair the protein's folding and thereby disrupt PURA's ability to bind RNA, which may explain why mutant PURA cannot localize correctly. Proske, Janowski et al. describe the molecular abnormalities of PURA underlying this disorder and show how molecular analysis of patient mutations can reveal the mechanisms of a disease at the cell level. The results show that the impact of mutations on the structural integrity of the protein, which affects its ability to bind RNA, are likely key to the symptoms of the syndrome. Additionally, their approach used establishes a way to predict and test mutations that will cause PURA syndrome. This may help to develop diagnostic tools for this condition.

Bacillus velezensis iturins inhibit the hemolytic activity of Staphylococcus aureus.

Bovine mastitis caused by S. aureus has a major economic impact on the dairy sector. With the crucial need for new therapies, anti-virulence strategies have gained attention as alternatives to antibiotics. Here we aimed to identify novel compounds that inhibit the production/activity of hemolysins, a virulence factor of S. aureus associated with mastitis severity. We screened Bacillus strains obtained from diverse sources for compounds showing anti-hemolytic activity. Our results demonstrate that lipopeptides produced by Bacillus spp. completely prevented the hemolytic activity of S. aureus at certain concentrations. Following purification, both iturins, fengycins, and surfactins were able to reduce hemolysis caused by S. aureus, with iturins showing the highest anti-hemolytic activity (up to 76% reduction). The lipopeptides showed an effect at the post-translational level. Molecular docking simulations demonstrated that these compounds can bind to hemolysin, possibly interfering with enzyme action. Lastly, molecular dynamics analysis indicated general stability of important residues for hemolysin activity as well as the presence of hydrogen bonds between iturins and these residues, with longevous interactions. Our data reveals, for the first time, an anti-hemolytic activity of lipopeptides and highlights the potential application of iturins as an anti-virulence therapy to control bovine mastitis caused by S. aureus.

Predicting FFAR4 agonists using structure-based machine learning approach based on molecular fingerprints.

Free Fatty Acid Receptor 4 (FFAR4), a G-protein-coupled receptor, is responsible for triggering intracellular signaling pathways that regulate various physiological processes. FFAR4 agonists are associated with enhancing insulin release and mitigating the atherogenic, obesogenic, pro-carcinogenic, and pro-diabetogenic effects, normally associated with the free fatty acids bound to FFAR4. In this research, molecular structure-based machine-learning techniques were employed to evaluate compounds as potential agonists for FFAR4. Molecular structures were encoded into bit arrays, serving as molecular fingerprints, which were subsequently analyzed using the Bayesian network algorithm to identify patterns for screening the data. The shortlisted hits obtained via machine learning protocols were further validated by Molecular Docking and via ADME and Toxicity predictions. The shortlisted compounds were then subjected to MD Simulations of the membrane-bound FFAR4-ligand complexes for 100 ns each. Molecular analyses, encompassing binding interactions, RMSD, RMSF, RoG, PCA, and FEL, were conducted to scrutinize the protein-ligand complexes at the inter-atomic level. The analyses revealed significant interactions of the shortlisted compounds with the crucial residues of FFAR4 previously documented. FFAR4 as part of the complexes demonstrated consistent RMSDs, ranging from 3.57 to 3.64, with minimal residue fluctuations 5.27 to 6.03 nm, suggesting stable complexes. The gyration values fluctuated between 22.8 to 23.5 nm, indicating structural compactness and orderliness across the studied systems. Additionally, distinct conformational motions were observed in each complex, with energy contours shifting to broader energy basins throughout the simulation, suggesting thermodynamically stable protein-ligand complexes. The two compounds CHEMBL2012662 and CHEMBL64616 are presented as potential FFAR4 agonists, based on these insights and in-depth analyses. Collectively, these findings advance our comprehension of FFAR4's functions and mechanisms, highlighting these compounds as potential FFAR4 agonists worthy of further exploration as innovative treatments for metabolic and immune-related conditions.

HBCVTr: an end-to-end transformer with a deep neural network hybrid model for anti-HBV and HCV activity predictor from SMILES.

Hepatitis B and C viruses (HBV and HCV) are significant causes of chronic liver diseases, with approximately 350 million infections globally. To accelerate the finding of effective treatment options, we introduce HBCVTr, a novel ligand-based drug design (LBDD) method for predicting the inhibitory activity of small molecules against HBV and HCV. HBCVTr employs a hybrid model consisting of double encoders of transformers and a deep neural network to learn the relationship between small molecules' simplified molecular-input line-entry system (SMILES) and their antiviral activity against HBV or HCV. The prediction accuracy of HBCVTr has surpassed baseline machine learning models and existing methods, with R-squared values of 0.641 and 0.721 for the HBV and HCV test sets, respectively. The trained models were successfully applied to virtual screening against 10 million compounds within 240 h, leading to the discovery of the top novel inhibitor candidates, including IJN04 for HBV and IJN12 and IJN19 for HCV. Molecular docking and dynamics simulations identified IJN04, IJN12, and IJN19 target proteins as the HBV core antigen, HCV NS5B RNA-dependent RNA polymerase, and HCV NS3/4A serine protease, respectively. Overall, HBCVTr offers a new and rapid drug discovery and development screening method targeting HBV and HCV.