LIGSITE: automatic and efficient detection of potential small molecule-binding sites in proteins

Databases of ligand-binding pockets and protein-ligand interactions

Many research groups and institutions have created a variety of databases curating experimental and predicted data related to protein-ligand binding. The landscape of available databases is dynamic, with new databases emerging and established databases becoming defunct. Here, we review the current state of databases that contain binding pockets and protein-ligand binding interactions. We have compiled a list of such databases, fifty-three of which are currently available for use. We discuss variation in how binding pockets are defined and summarize pocket-finding methods. We organize the fifty-three databases into subgroups based on goals and contents, and describe standard use cases. We also illustrate that pockets within the same protein are characterized differently across different databases. Finally, we assess critical issues of sustainability, accessibility and redundancy.

A Point Cloud Graph Neural Network for Protein-Ligand Binding Site Prediction

Predicting protein-ligand binding sites is an integral part of structural biology and drug design. A comprehensive understanding of these binding sites is essential for advancing drug innovation, elucidating mechanisms of biological function, and exploring the nature of disease. However, accurately identifying protein-ligand binding sites remains a challenging task. To address this, we propose PGpocket, a geometric deep learning-based framework to improve protein-ligand binding site prediction. Initially, the protein surface is converted into a point cloud, and then the geometric and chemical properties of each point are calculated. Subsequently, the point cloud graph is constructed based on the inter-point distances, and the point cloud graph neural network (GNN) is applied to extract and analyze the protein surface information to predict potential binding sites. PGpocket is trained on the scPDB dataset, and its performance is verified on two independent test sets, Coach420 and HOLO4K. The results show that PGpocket achieves a 58% success rate on the Coach420 dataset and a 56% success rate on the HOLO4K dataset. These results surpass competing algorithms, demonstrating PGpocket's advancement and practicality for protein-ligand binding site prediction.

Biochemical and structural elucidation of the L-carnitine degradation pathway of the human pathogen Acinetobacter baumannii

Acinetobacter baumannii is an opportunistic human pathogen which can use host-derived L-carnitine as sole carbon and energy source. Recently, an L-carnitine transporter (Aci1347) and a specific monooxygense (CntA/CntB) for the intracellular cleavage of L-carnitine have been characterized. Subsequent conversion of the resulting malic semialdehyde into the central metabolite L-malate was hypothesized. Alternatively, L-carnitine degradation via D-malate with subsequent oxidation into pyruvate was proposed. Here we describe the in vitro and in vivo reconstitution of the entire pathway, starting from the as yet uncharacterized gene products of the carnitine degradation gene operon. Using recombinantly purified enzymes, enantiomer-specific formation of D-malate by the NAD(P)+-dependent malic semialdehyde dehydrogenase (MSA-DH) is demonstrated. The solved X-ray crystal structure of tetrameric MSA-DH reveals the key catalytic residues Cys290 and Glu256, accessible through opposing substrate and cofactor funnels. Specific substrate binding is enabled by Arg166, Arg284 and Ser447 while dual cofactor specificity for NAD+ and NADP+ is mediated by Asn184. The subsequent conversion of the unusual D-malate reaction product by an uncharacterized NAD+-dependent malate dehydrogenase (MDH) is shown. Tetrameric MDH is a β-decarboxylating dehydrogenase that synthesizes pyruvate. MDH experiments with alternative substrates showed a high degree of substrate specificity. Finally, the entire A. baumannni pathway was heterologously reconstituted, allowing E. coli to grow on L-carnitine as a carbon and energy source. Overall, the metabolic conversion of L-carnitine via malic semialdehyde and D-malate into pyruvate, CO2 and trimethylamine was demonstrated. Trimethylamine is also an important gut microbiota-dependent metabolite that is associated with an increased risk of cardiovascular disease. The pathway reconstitution experiments allowed us to assess the TMA forming capacity of gut microbes which is related to human cardiovascular health.

Comparative in silico analysis of CNS-active molecules targeting the blood-brain barrier choline transporter for Alzheimer's disease therapy

This study investigated the blood‒brain barrier (BBB) permeability of the central nervous system (CNS)-active compounds donepezil (DON), methionine (MET), and memantine (MEM) by employing a comprehensive in silico approach. These compounds are of particular interest for Alzheimer's disease (AD) therapy. Rigid-flexible molecular docking simulations indicated favorable binding affinities of all the compounds with BBB-ChT, with DON exhibiting the highest binding affinity (ΔGbind = -10.26 kcal/mol), predominantly mediated by significant hydrophobic interactions. In silico kinetic profiling suggested the stability of the DON/BBB-ChT complex, with ligand release prompted by conformational changes. 3D molecular alignment corroborated a minor conformational shift for DON in its minimal binding energy pose. Predictions indicated that active transport mechanisms notably enhance the brain distribution of donepezil compared to that of MET and MEM. Additionally, DON and MEM exhibited low mutagenic probabilities, while MET was identified as highly mutagenic. Overall, these findings highlight the potential of donepezil for superior BBB penetration, primarily through active transport mechanisms, underscoring the need for further validation through in vitro and in vivo studies for effective AD treatment.

Graphical Abstract

Supplementary Information

The online version contains supplementary material available at 10.1007/s40203-024-00245-w.

CavitOmiX Drug Discovery: Engineering Antivirals with Enhanced Spectrum and Reduced Side Effects for Arboviral Diseases

Advancing climate change increases the risk of future infectious disease outbreaks, particularly of zoonotic diseases, by affecting the abundance and spread of viral vectors. Concerningly, there are currently no approved drugs for some relevant diseases, such as the arboviral diseases chikungunya, dengue or zika. The development of novel inhibitors takes 10-15 years to reach the market and faces critical challenges in preclinical and clinical trials, with approximately 30% of trials failing due to side effects. As an early response to emerging infectious diseases, CavitOmiX allows for a rapid computational screening of databases containing 3D point-clouds representing binding sites of approved drugs to identify candidates for off-label use. This process, known as drug repurposing, reduces the time and cost of regulatory approval. Here, we present potential approved drug candidates for off-label use, targeting the ADP-ribose binding site of Alphavirus chikungunya non-structural protein 3. Additionally, we demonstrate a novel in silico drug design approach, considering potential side effects at the earliest stages of drug development. We use a genetic algorithm to iteratively refine potential inhibitors for (i) reduced off-target activity and (ii) improved binding to different viral variants or across related viral species, to provide broad-spectrum and safe antivirals for the future.

Systematic computational strategies for identifying protein targets and lead discovery

Computational algorithms and tools have retrenched the drug discovery and development timeline. The applicability of computational approaches has gained immense relevance owing to the dramatic surge in the structural information of biomacromolecules and their heteromolecular complexes. Computational methods are now extensively used in identifying new protein targets, druggability assessment, pharmacophore mapping, molecular docking, the virtual screening of lead molecules, bioactivity prediction, molecular dynamics of protein-ligand complexes, affinity prediction, and for designing better ligands. Herein, we provide an overview of salient components of recently reported computational drug-discovery workflows that includes algorithms, tools, and databases for protein target identification and optimized ligand selection.

Intracellular Allosteric Antagonist of the Olfactory Receptor OR51E2

Olfactory receptors are members of class A (rhodopsin-like) family of G protein-coupled receptors (GPCRs). Their expression and function have been increasingly studied in nonolfactory tissues, and many have been identified as potential therapeutic targets. In this manuscript, we focus on the discovery of novel ligands for the olfactory receptor family 51 subfamily E2 (OR51E2). We performed an artificial intelligence-based virtual drug screen of a ∼2.2 million small molecule library. Cell-based functional assay identified compound 80 (C80) as an antagonist and inverse agonist, and detailed pharmacological analysis revealed C80 acts as a negative allosteric modulator by significantly decreasing the agonist efficacy, while having a minimal effect on receptor affinity for agonist. C80 binds to an allosteric binding site formed by a network of nine residues localized in the intracellular parts of transmembrane domains 3, 5, 6, 7, and H8, which also partially overlaps with a G protein binding site. Mutational experiments of residues involved in C80 binding uncovered the significance of the C2406.37 position in blocking the activation-related conformational change and keeping the receptor in the inactive form. Our study provides a mechanistic understanding of the negative allosteric action of C80 on agonist-ctivated OR51E2. We believe the identification of the antagonist of OR51E2 will enable a multitude of studies aiming to determine the functional role of this receptor in specific biologic processes. SIGNIFICANCE STATEMENT: OR51E2 has been implicated in various biological processes, and its antagonists that can effectively modulate its activity have therapeutic potential. Here we report the discovery of a negative allosteric modulator of OR51E2 and provide a mechanistic understanding of its action. We demonstrate that this modulator has an inhibitory effect on the efficacy of the agonist for the receptor and reveal a network of nine residues that constitute its binding pocket, which also partially overlaps with the G protein binding site.

Subpocket-Based Analysis Approach for the Protein Pocket Dynamics

Structural and dynamic characteristics of protein pockets remarkably influence their biological functions and are also important for enzyme engineering and new drug research and development. To date, several softwares have been developed to analyze the dynamic properties of protein pockets. However, due to the complexity and diversity of the pocket information during the kinetic relaxation, further improvement and capacity expansion of current tools are required. Here, we developed a platform software AlphaTraj in which a computational strategy that divides the whole protein pocket into subpockets and examines various properties of the subpockets such as survival time, stability, and correlation was proposed and implemented. We also proposed a scoring function for the subpockets as well as the whole pocket to visualize the quality of the pocket. Furthermore, we implemented automated conformational search functions for ligand docking and ligand optimization. These functions may help us to gain a deep understanding of the dynamic properties of protein pockets and accelerate the protein engineering and the design of inhibitors and small-molecule drugs. The software is freely available at https://github.com/dooo12332/AlphaTraj.git under the GNU GPL license.

PUResNetV2.0: a deep learning model leveraging sparse representation for improved ligand binding site prediction

Accurate ligand binding site prediction (LBSP) within proteins is essential for drug discovery. We developed ProteinUNetResNetV2.0 (PUResNetV2.0), leveraging sparse representation of protein structures to improve LBSP accuracy. Our training dataset included protein complexes from 4729 protein families. Evaluations on benchmark datasets showed that PUResNetV2.0 achieved an 85.4% Distance Center Atom (DCA) success rate and a 74.7% F1 Score on the Holo801 dataset, outperforming existing methods. However, its performance in specific cases, such as RNA, DNA, peptide-like ligand, and ion binding site prediction, was limited due to constraints in our training data. Our findings underscore the potential of sparse representation in LBSP, especially for oligomeric structures, suggesting PUResNetV2.0 as a promising tool for computational drug discovery.

Diverse models of cavity engineering in enzyme modification: Creation, filling, and reshaping

Most enzyme modification strategies focus on designing the active sites or their surrounding structures. Interestingly, a large portion of the enzymes (60%) feature active sites located within spacious cavities. Despite recent discoveries, cavity-mediated enzyme engineering remains crucial for enhancing enzyme properties and unraveling folding-unfolding mechanisms. Cavity engineering influences enzyme stability, catalytic activity, specificity, substrate recognition, and docking. This article provides a comprehensive review of various cavity engineering models for enzyme modification, including cavity creation, filling, and reshaping. Additionally, it also discusses feasible tools for geometric analysis, functional assessment, and modification of cavities, and explores potential future research directions in this field. Furthermore, a promising universal modification strategy for cavity engineering that leverages state-of-the-art technologies and methodologies to tailor cavities according to the specific requirements of industrial production conditions is proposed.

ANN based prediction of ligand binding sites outside deep cavities to facilitate drug designing

The ever-changing environmental conditions and pollution are the prime reasons for the onset of several emerging and re-merging diseases. This demands the faster designing of new drugs to curb the deadly diseases in less waiting time to cure the animals and humans. Drug molecules interact with only protein surface on specific locations termed as ligand binding sites (LBS). Therefore, the knowledge of LBS is required for rational drug designing. Existing geometrical LBS prediction methods rely on search of cavities based on the fact that 83% of the LBS found in deep cavities, however, these methods usually fail where LBS localize outside deep cavities. To overcome this challenge, the present work provides an artificial neural network (ANN) based method to predict LBS outside deep cavities in animal proteins including human to facilitate drug designing. In the present work a feed-forward backpropagation neural network was trained by utilizing 38 structural, atomic, physiochemical, and evolutionary discriminant features of LBS and non-LBS residues localized in the extracted roughest patch on protein surface. The performance of this ANN based prediction method was found 76% better for those proteins where cavity subspace (extracted by MetaPocket 2.0, a consensus method) failed to predict LBS due to their localization outside the deep cavities. The prediction of LBS outside deep cavities will facilitate in drug designing for the proteins where it is not possible due to lack of LBS information as the geometrical LBS prediction methods rely on extraction of deep cavities.

Graph Attention Site Prediction (GrASP): Identifying Druggable Binding Sites Using Graph Neural Networks with Attention

Identifying and discovering druggable protein binding sites is an important early step in computer-aided drug discovery, but it remains a difficult task where most campaigns rely on a priori knowledge of binding sites from experiments. Here, we present a binding site prediction method called Graph Attention Site Prediction (GrASP) and re-evaluate assumptions in nearly every step in the site prediction workflow from data set preparation to model evaluation. GrASP is able to achieve state-of-the-art performance at recovering binding sites in PDB structures while maintaining a high degree of precision which will minimize wasted computation in downstream tasks such as docking and free energy perturbation.

Colabind: A Cloud-Based Approach for Prediction of Binding Sites Using Coarse-Grained Simulations with Molecular Probes

Binding site prediction is a crucial step in understanding protein-ligand and protein-protein interactions (PPIs) with broad implications in drug discovery and bioinformatics. This study introduces Colabind, a robust, versatile, and user-friendly cloud-based approach that employs coarse-grained molecular dynamics simulations in the presence of molecular probes, mimicking fragments of drug-like compounds. Our method has demonstrated high effectiveness when validated across a diverse range of biological targets spanning various protein classes, successfully identifying orthosteric binding sites, as well as known druggable allosteric or PPI sites, in both experimentally determined and AI-predicted protein structures, consistently placing them among the top-ranked sites. Furthermore, we suggest that careful inspection of the identified regions with a high affinity for specific probes can provide valuable insights for the development of pharmacophore hypotheses. The approach is available at https://github.com/porekhov/CG_probeMD.

Structural insights into the distinct substrate preferences of two bacterial epoxide hydrolases

Epoxide hydrolases (EHs), which catalyze the transformation of epoxides to diols, are present in many eukaryotic and prokaryotic organisms. They have recently drawn considerable attention from organic chemists owing to their application in the semisynthesis of enantiospecific diol compounds. Here, we report the crystal structures of BoEH from Bosea sp. PAMC 26642 and CaEH from Caballeronia sordidicola PAMC 26510 at 1.95 and 2.43 Å resolution, respectively. Structural analysis showed that the overall structures of BoEH and CaEH commonly possess typical α/β hydrolase fold with the same ring-opening residues (Tyr-Tyr) and conserved catalytic triad residues (Asp-Asp-His). However, the two enzymes were found to have significantly different sequence compositions in the cap domain region, which is involved in the formation of the substrate-binding site in both enzymes. Enzyme activity assay results showed that BoEH had the strongest activity toward the linear aliphatic substrates, whereas CaEH had a higher preference for aromatic- and cycloaliphatic substrates. Computational docking simulations and tunnel identification revealed important residues with different substrate-binding preferences. Collectively, structure comparison studies, together with ligand docking simulation results, suggested that the differences in substrate-binding site residues were highly correlated with substrate specificity.

Equilibrium landscape of ingress/egress channels and gating residues of the Cytochrome P450 3A4

The Cytochrome P450 (CYP) enzymes metabolize a variety of drugs, which may potentially lead to toxicity or reduced efficacy when drugs are co-administered. These drug-drug interactions are often manifested by CYP3A4, the most prevalent of all CYP isozymes. We carried out multiple MD simulations employing CAVER to quantify the channels, and Hidden Markov Models (HMM) to characterize the behavior of the gating residues. We discuss channel properties, bottleneck residues with respect to their likelihood to deem the respective channel ingress or egress, gating residues regarding their open or closed states, and channel location relative to the membrane. Channels do not display coordinated motion and randomly transition between different conformations. Gateway residues also behave in a random fashion. Our findings shed light on the equilibrium behavior of the gating residues and channels in the apo state.

Thirty years with three-dimensional structures of lipoxygenases

The X-ray crystal structures of soybean lipoxygenase (LOX) and rabbit 15-LOX were reported in the 1990s. Subsequent 3D structures demonstrated a conserved U-like shape of the substrate cavities as reviewed here. The 8-LOX:arachidonic acid (AA) complex showed AA bound to the substrate cavity carboxylate-out with C10 at 3.4 Å from the iron metal center. A recent cryo-electron microscopy (EM) analysis of the 12-LOX:AA complex illustrated AA in the same position as in the 8-LOX:AA complex. The 15- and 12-LOX complexes with isoenzyme-specific inhibitors/substrate mimics confirmed the U-fold. 5-LOX oxidizes AA to leukotriene A4, the first step in biosynthesis of mediators of asthma. The X-ray structure showed that the entrance to the substrate cavity was closed to AA by Phe and Tyr residues of a partly unfolded α2-helix. Recent X-ray analysis revealed that soaking with inhibitors shifted the short α2-helix to a long and continuous, which opened the substrate cavity. The α2-helix also adopted two conformations in 15-LOX. 12-LOX dimers consisted of one closed and one open subunit with an elongated α2-helix. 13C-ENDOR-MD computations of the 9-MnLOX:linoleate complex showed carboxylate-out position with C11 placed 3.4 ± 0.1 Å from the catalytic water. 3D structures have provided a solid ground for future research.

CRAFT: a web-integrated cavity prediction tool based on flow transfer algorithm

Numerous computational methods, including evolutionary-based, energy-based, and geometrical-based methods, are utilized to identify cavities inside proteins. Cavity information aids protein function annotation, drug design, poly-pharmacology, and allosteric site investigation. This article introduces "flow transfer algorithm" for rapid and effective identification of diverse protein cavities through multidimensional cavity scan. Initially, it identifies delimiter and susceptible tetrahedra to establish boundary regions and provide seed tetrahedra. Seed tetrahedron faces are precisely scanned using the maximum circle radius to transfer seed flow to neighboring tetrahedra. Seed flow continues until terminated by boundaries or forbidden faces, where a face is forbidden if the estimated maximum circle radius is less or equal to the user-defined maximum circle radius. After a seed scanning, tetrahedra involved in the flow are clustered to locate the cavity. The CRAFT web interface integrates this algorithm for protein cavity identification with enhanced user control. It supports proteins with cofactors, hydrogens, and ligands and provides comprehensive features such as 3D visualization, cavity physicochemical properties, percentage contribution graphs, and highlighted residues for each cavity. CRAFT can be accessed through its web interface at http://pitools.niper.ac.in/CRAFT , complemented by the command version available at https://github.com/PGlab-NIPER/CRAFT/ .Scientific contribution: Flow transfer algorithm is a novel geometric approach for accurate and reliable prediction of diverse protein cavities. This algorithm employs a distinct concept involving maximum circle radius within the 3D Delaunay triangulation to address diverse van der Waals radii while existing methods overlook atom specific van der Waals radii or rely on complex weighted geometric techniques.

E2EATP: Fast and High-Accuracy Protein-ATP Binding Residue Prediction via Protein Language Model Embedding

Identifying the ATP-binding sites of proteins is fundamentally important to uncover the mechanisms of protein functions and explore drug discovery. Many computational methods are proposed to predict ATP-binding sites. However, due to the limitation of the quality of feature representation, the prediction performance still has a big room for improvement. In this study, we propose an end-to-end deep learning model, E2EATP, to dig out more discriminative information from a protein sequence for improving the ATP-binding site prediction performance. Concretely, we employ a pretrained deep learning-based protein language model (ESM2) to automatically extract high-latent discriminative representations of protein sequences relevant for protein functions. Based on ESM2, we design a residual convolutional neural network to train a protein-ATP binding site prediction model. Furthermore, a weighted focal loss function is used to reduce the negative impact of imbalanced data on the model training stage. Experimental results on the two independent testing data sets demonstrate that E2EATP could achieve higher Matthew's correlation coefficient and AUC values than most existing state-of-the-art prediction methods. The speed (about 0.05 s per protein) of E2EATP is much faster than the other existing prediction methods. Detailed data analyses show that the major advantage of E2EATP lies at the utilization of the pretrained protein language model that extracts more discriminative information from the protein sequence only. The standalone package of E2EATP is freely available for academic at https://github.com/jun-csbio/e2eatp/.

In Vivo and In Silico Studies of the Hepatoprotective Activity of Tert-Butylhydroquinone

Tert-butylhydroquinone (TBHQ) is a synthetic food antioxidant with biological activities, but little is known about its pharmacological benefits in liver disease. Therefore, this work aimed to evaluate TBHQ during acute liver damage induced by CCl4 (24 h) or BDL (48 h) in Wistar rats. It was found that pretreatment with TBHQ prevents 50% of mortality induced by a lethal dose of CCl4 (4 g/kg, i.p.), and 80% of BDL+TBHQ rats survived, while only 50% of the BDL group survived. Serum markers of liver damage and macroscopic and microscopic (H&E staining) observations suggest that TBHQ protects from both hepatocellular necrosis caused by the sublethal dose of CCl4 (1.6 g/kg, i.p.), as well as necrosis/ductal proliferation caused by BDL. Additionally, online databases identified 49 potential protein targets for TBHQ. Finally, a biological target candidate (Keap1) was evaluated in a proof-of-concept in silico molecular docking assay, resulting in an interaction energy of -5.5491 kcal/mol, which was higher than RA839 and lower than monoethyl fumarate (compounds known to bind to Keap1). These findings suggest that TBHQ increases the survival of animals subjected to CCl4 intoxication or BDL, presumably by reducing hepatocellular damage, probably due to the interaction of TBHQ with Keap1.

AI-assisted structural consensus-proteome prediction of human monkeypox viruses isolated within a year after the 2022 multi-country outbreak

IMPORTANCE

The 2022 outbreak of the monkeypox virus already involves, by April 2023, 110 countries with 86,956 confirmed cases and 119 deaths. Understanding an emerging disease on a molecular level is essential to study infection processes and eventually guide drug discovery at an early stage. To support this, we provide the so far most comprehensive structural proteome of the monkeypox virus, which includes 210 structural models, each computed with three state-of-the-art structure prediction methods. Instead of building on a single-genome sequence, we generated our models from a consensus of 3,713 high-quality genome sequences sampled from patients within 1 year of the outbreak. Therefore, we present an average structural proteome of the currently isolated viruses, including mutational analyses with a special focus on drug-binding sites. Continuing dynamic mutation monitoring within the structural proteome presented here is essential to timely predict possible physiological changes in the evolving virus.

Unraveling viral drug targets: a deep learning-based approach for the identification of potential binding sites

The coronavirus disease 2019 (COVID-19) pandemic has spurred a wide range of approaches to control and combat the disease. However, selecting an effective antiviral drug target remains a time-consuming challenge. Computational methods offer a promising solution by efficiently reducing the number of candidates. In this study, we propose a structure- and deep learning-based approach that identifies vulnerable regions in viral proteins corresponding to drug binding sites. Our approach takes into account the protein dynamics, accessibility and mutability of the binding site and the putative mechanism of action of the drug. We applied this technique to validate drug targeting toward severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) spike glycoprotein S. Our findings reveal a conformation- and oligomer-specific glycan-free binding site proximal to the receptor binding domain. This site comprises topologically important amino acid residues. Molecular dynamics simulations of Spike in complex with candidate drug molecules bound to the potential binding sites indicate an equilibrium shifted toward the inactive conformation compared with drug-free simulations. Small molecules targeting this binding site have the potential to prevent the closed-to-open conformational transition of Spike, thereby allosterically inhibiting its interaction with human angiotensin-converting enzyme 2 receptor. Using a pseudotyped virus-based assay with a SARS-CoV-2 neutralizing antibody, we identified a set of hit compounds that exhibited inhibition at micromolar concentrations.

Integrated Molecular Modeling and Machine Learning for Drug Design

Modern therapeutic development often involves several stages that are interconnected, and multiple iterations are usually required to bring a new drug to the market. Computational approaches have increasingly become an indispensable part of helping reduce the time and cost of the research and development of new drugs. In this Perspective, we summarize our recent efforts on integrating molecular modeling and machine learning to develop computational tools for modulator design, including a pocket-guided rational design approach based on AlphaSpace to target protein-protein interactions, delta machine learning scoring functions for protein-ligand docking as well as virtual screening, and state-of-the-art deep learning models to predict calculated and experimental molecular properties based on molecular mechanics optimized geometries. Meanwhile, we discuss remaining challenges and promising directions for further development and use a retrospective example of FDA approved kinase inhibitor Erlotinib to demonstrate the use of these newly developed computational tools.

Vibrio cholerae's ToxRS bile sensing system

The seventh pandemic of the diarrheal cholera disease, which began in 1960, is caused by the Gram-negative bacterium Vibrio cholerae. Its environmental persistence provoking recurring sudden outbreaks is enabled by V. cholerae's rapid adaption to changing environments involving sensory proteins like ToxR and ToxS. Located at the inner membrane, ToxR and ToxS react to environmental stimuli like bile acid, thereby inducing survival strategies for example bile resistance and virulence regulation. The presented crystal structure of the sensory domains of ToxR and ToxS in combination with multiple bile acid interaction studies, reveals that a bile binding pocket of ToxS is only properly folded upon binding to ToxR. Our data proposes an interdependent functionality between ToxR transcriptional activity and ToxS sensory function. These findings support the previously suggested link between ToxRS and VtrAC-like co-component systems. Besides VtrAC, ToxRS is now the only experimentally determined structure within this recently defined superfamily, further emphasizing its significance. In-depth analysis of the ToxRS complex reveals its remarkable conservation across various Vibrio species, underlining the significance of conserved residues in the ToxS barrel and the more diverse ToxR sensory domain. Unravelling the intricate mechanisms governing ToxRS's environmental sensing capabilities, provides a promising tool for disruption of this vital interaction, ultimately inhibiting Vibrio's survival and virulence. Our findings hold far-reaching implications for all Vibrio strains that rely on the ToxRS system as a shared sensory cornerstone for adapting to their surroundings.

RefinePocket: An Attention-Enhanced and Mask-Guided Deep Learning Approach for Protein Binding Site Prediction

Protein binding site prediction is an important prerequisite task of drug discovery and design. While binding sites are very small, irregular and varied in shape, making the prediction very challenging. Standard 3D U-Net has been adopted to predict binding sites but got stuck with unsatisfactory prediction results, incomplete, out-of-bounds, or even failed. The reason is that this scheme is less capable of extracting the chemical interactions of the entire region and hardly takes into account the difficulty of segmenting complex shapes. In this paper, we propose a refined U-Net architecture, called RefinePocket, consisting of an attention-enhanced encoder and a mask-guided decoder. During encoding, taking binding site proposal as input, we employ Dual Attention Block (DAB) hierarchically to capture rich global information, exploring residue relationship and chemical correlations in spatial and channel dimensions respectively. Then, based on the enhanced representation extracted by the encoder, we devise Refine Block (RB) in the decoder to enable self-guided refinement of uncertain regions gradually, resulting in more precise segmentation. Experiments show that DAB and RB complement and promote each other, making RefinePocket has an average improvement of 10.02% on DCC and 4.26% on DVO compared with the state-of-the-art method on four test sets.

SiteFerret: Beyond Simple Pocket Identification in Proteins

We present a novel method for the automatic detection of pockets on protein molecular surfaces. The algorithm is based on an ad hoc hierarchical clustering of virtual probe spheres obtained from the geometrical primitives used by the NanoShaper software to build the solvent-excluded molecular surface. The final ranking of putative pockets is based on the Isolation Forest method, an unsupervised learning approach originally developed for anomaly detection. A detailed importance analysis of pocket features provides insight into which geometrical (clustering) and chemical (amino acidic composition) properties characterize a good binding site. The method also provides a segmentation of pockets into smaller subpockets. We prove that subpockets are a convenient representation to pinpoint the binding site with great precision. SiteFerret is outstanding in its versatility, accurately predicting a wide range of binding sites, from those binding small molecules to those binding peptides, including difficult shallow sites.

Graph Attention Site Prediction (GrASP): Identifying Druggable Binding Sites Using Graph Neural Networks with Attention

Identifying and discovering druggable protein binding sites is an important early step in computer-aided drug discovery but remains a difficult task where most campaigns rely on a priori knowledge of binding sites from experiments. Here we present a novel binding site prediction method called Graph Attention Site Prediction (GrASP) and re-evaluate assumptions in nearly every step in the site prediction workflow from dataset preparation to model evaluation. GrASP is able to achieve state-of-the-art performance at recovering binding sites in PDB structures while maintaining a high degree of precision which will minimize wasted computation in downstream tasks such as docking and free energy perturbation.

Discovery of a Hidden Trypanosoma cruzi Spermidine Synthase Binding Site and Inhibitors through In Silico, In Vitro, and X-ray Crystallography

In drug discovery research, the selection of promising binding sites and understanding the binding mode of compounds are crucial fundamental studies. The current understanding of the proteins-ligand binding model extends beyond the simple lock and key model to include the induced-fit model, which alters the conformation to match the shape of the ligand, and the pre-existing equilibrium model, selectively binding structures with high binding affinity from a diverse ensemble of proteins. Although methods for detecting target protein binding sites and virtual screening techniques using docking simulation are well-established, with numerous studies reported, they only consider a very limited number of structures in the diverse ensemble of proteins, as these methods are applied to a single structure. Molecular dynamics (MD) simulation is a method for predicting protein dynamics and can detect potential ensembles of protein binding sites and hidden sites unobservable in a single-point structure. In this study, to demonstrate the utility of virtual screening with protein dynamics, MD simulations were performed on Trypanosoma cruzi spermidine synthase to obtain an ensemble of dominant binding sites with a high probability of existence. The structure of the binding site obtained through MD simulation revealed pockets in addition to the active site that was present in the initial structure. Using the obtained binding site structures, virtual screening of 4.8 million compounds by docking simulation, in vitro assays, and X-ray analysis was conducted, successfully identifying two hit compounds.

Identification and validation of fusidic acid and flufenamic acid as inhibitors of SARS-CoV-2 replication using DrugSolver CavitomiX

In this work, we present DrugSolver CavitomiX, a novel computational pipeline for drug repurposing and identifying ligands and inhibitors of target enzymes. The pipeline is based on cavity point clouds representing physico-chemical properties of the cavity induced solely by the protein. To test the pipeline's ability to identify inhibitors, we chose enzymes essential for SARS-CoV-2 replication as a test system. The active-site cavities of the viral enzymes main protease (M^pro) and papain-like protease (Pl^pro), as well as of the human transmembrane serine protease 2 (TMPRSS2), were selected as target cavities. Using active-site point-cloud comparisons, it was possible to identify two compounds-flufenamic acid and fusidic acid-which show strong inhibition of viral replication. The complexes from which fusidic acid and flufenamic acid were derived would not have been identified using classical sequence- and structure-based methods as they show very little structural (TM-score: 0.1 and 0.09, respectively) and very low sequence (~ 5%) identity to M^pro and TMPRSS2, respectively. Furthermore, a cavity-based off-target screening was performed using acetylcholinesterase (AChE) as an example. Using cavity comparisons, the human carboxylesterase was successfully identified, which is a described off-target for AChE inhibitors.

Analyzing the Geometry and Dynamics of Viral Structures: A Review of Computational Approaches Based on Alpha Shape Theory, Normal Mode Analysis, and Poisson-Boltzmann Theories

The current SARS-CoV-2 pandemic highlights our fragility when we are exposed to emergent viruses either directly or through zoonotic diseases. Fortunately, our knowledge of the biology of those viruses is improving. In particular, we have more and more structural information on virions, i.e., the infective form of a virus that includes its genomic material and surrounding protective capsid, and on their gene products. It is important to have methods that enable the analyses of structural information on such large macromolecular systems. We review some of those methods in this paper. We focus on understanding the geometry of virions and viral structural proteins, their dynamics, and their energetics, with the ambition that this understanding can help design antiviral agents. We discuss those methods in light of the specificities of those structures, mainly that they are huge. We focus on three of our own methods based on the alpha shape theory for computing geometry, normal mode analyses to study dynamics, and modified Poisson-Boltzmann theories to study the organization of ions and co-solvent and solvent molecules around biomacromolecules. The corresponding software has computing times that are compatible with the use of regular desktop computers. We show examples of their applications on some outer shells and structural proteins of the West Nile Virus.

Binding Site Detection Remastered: Enabling Fast, Robust, and Reliable Binding Site Detection and Descriptor Calculation with DoGSite3

Binding site prediction on protein structures is a crucial step in early phase drug discovery whenever experimental or predicted structure models are involved. DoGSite belongs to the widely used tools for this task. It is a grid-based method that uses a Difference-of-Gaussian filter to detect cavities on the protein surface. We recently reimplemented the first version of this method, released in 2010, focusing on improved binding site detection in the presence of ligands and optimized parameters for more robust, reliable, and fast predictions and binding site descriptor calculations. Here, we introduce the new version, DoGSite3, compare it to its predecessor, and re-evaluate DoGSite on published data sets for a large-scale comparative performance evaluation.

Bioinformatics approaches and big data analytics opportunities in improving fisheries and aquaculture

Aquaculture has witnessed an excellent growth rate during the last two decades and offers huge potential to provide nutritional as well as livelihood security. Genomic research has contributed significantly toward the development of beneficial technologies for aquaculture. The existing high throughput technologies like next-generation technologies generate oceanic data which requires extensive analysis using appropriate tools. Bioinformatics is a rapidly evolving science that involves integrating gene based information and computational technology to produce new knowledge for the benefit of aquaculture. Bioinformatics provides new opportunities as well as challenges for information and data processing in new generation aquaculture. Rapid technical advancements have opened up a world of possibilities for using current genomics to improve aquaculture performance. Understanding the genes that govern economically relevant characteristics, necessitates a significant amount of additional research. The various dimensions of data sources includes next-generation DNA sequencing, protein sequencing, RNA sequencing gene expression profiles, metabolic pathways, molecular markers, and so on. Appropriate bioinformatics tools are developed to mine the biologically relevant and commercially useful results. The purpose of this scoping review is to present various arms of diverse bioinformatics tools with special emphasis on practical translation to the aquaculture industry.

Discovery of a cryptic pocket in the AI-predicted structure of PPM1D phosphatase explains the binding site and potency of its allosteric inhibitors

Virtual screening is a widely used tool for drug discovery, but its predictive power can vary dramatically depending on how much structural data is available. In the best case, crystal structures of a ligand-bound protein can help find more potent ligands. However, virtual screens tend to be less predictive when only ligand-free crystal structures are available, and even less predictive if a homology model or other predicted structure must be used. Here, we explore the possibility that this situation can be improved by better accounting for protein dynamics, as simulations started from a single structure have a reasonable chance of sampling nearby structures that are more compatible with ligand binding. As a specific example, we consider the cancer drug target PPM1D/Wip1 phosphatase, a protein that lacks crystal structures. High-throughput screens have led to the discovery of several allosteric inhibitors of PPM1D, but their binding mode remains unknown. To enable further drug discovery efforts, we assessed the predictive power of an AlphaFold-predicted structure of PPM1D and a Markov state model (MSM) built from molecular dynamics simulations initiated from that structure. Our simulations reveal a cryptic pocket at the interface between two important structural elements, the flap and hinge regions. Using deep learning to predict the pose quality of each docked compound for the active site and cryptic pocket suggests that the inhibitors strongly prefer binding to the cryptic pocket, consistent with their allosteric effect. The predicted affinities for the dynamically uncovered cryptic pocket also recapitulate the relative potencies of the compounds (τ _b =0.70) better than the predicted affinities for the static AlphaFold-predicted structure (τ _b =0.42). Taken together, these results suggest that targeting the cryptic pocket is a good strategy for drugging PPM1D and, more generally, that conformations selected from simulation can improve virtual screening when limited structural data is available.

Accelerating Cryptic Pocket Discovery Using AlphaFold

Cryptic pockets, or pockets absent in ligand-free, experimentally determined structures, hold great potential as drug targets. However, cryptic pocket openings are often beyond the reach of conventional biomolecular simulations because certain cryptic pocket openings involve slow motions. Here, we investigate whether AlphaFold can be used to accelerate cryptic pocket discovery either by generating structures with open pockets directly or generating structures with partially open pockets that can be used as starting points for simulations. We use AlphaFold to generate ensembles for 10 known cryptic pocket examples, including five that were deposited after AlphaFold's training data were extracted from the PDB. We find that in 6 out of 10 cases AlphaFold samples the open state. For plasmepsin II, an aspartic protease from the causative agent of malaria, AlphaFold only captures a partial pocket opening. As a result, we ran simulations from an ensemble of AlphaFold-generated structures and show that this strategy samples cryptic pocket opening, even though an equivalent amount of simulations launched from a ligand-free experimental structure fails to do so. Markov state models (MSMs) constructed from the AlphaFold-seeded simulations quickly yield a free energy landscape of cryptic pocket opening that is in good agreement with the same landscape generated with well-tempered metadynamics. Taken together, our results demonstrate that AlphaFold has a useful role to play in cryptic pocket discovery but that many cryptic pockets may remain difficult to sample using AlphaFold alone.

3-Amino-Substituted Analogues of Fusidic Acid as Membrane-Active Antibacterial Compounds

Fusidic acid (FA) is an antibiotic with high activity against Staphylococcus aureus; it has been used in clinical practice since the 1960s. However, the narrow antimicrobial spectrum of FA limits its application in the treatment of bacterial infections. In this regard, this work aims both at the study of the antimicrobial effect of a number of FA amines and at the identification of their potential biological targets. In this way, FA analogues containing aliphatic and aromatic amino groups and biogenic polyamine, spermine and spermidine, moieties at the C-3 atom, were synthesized (20 examples). Pyrazinecarboxamide-substituted analogues exhibit a high antibacterial activity against S. aureus (MRSA) with MIC ≤ 0.25 μg/mL. Spermine and spermidine derivatives, along with activity against S. aureus, also inhibit the growth and reproduction of Gram-negative bacteria Escherichia coli, Acinetobacter baumannii, and Pseudomonas aeruginosa, and have a high fungicidal effect against Candida albicans and Cryptococcus neoformans. The study of the membrane activity demonstrated that the spermidine- and spermine-containing compounds are able to immerse into membranes and disorder the lipidsleading to a detergent effect. Moreover, spermine-based compounds are also able to form ion-permeable pores in the lipid bilayers mimicking the bacterial membranes. Using molecular docking, inhibition of the protein synthesis elongation factor EF-G was proposed, and polyamine substituents were shown to make the greatest contribution to the stability of the complexes of fusidic acid derivatives with biological targets. This suggests that the antibacterial effect of the obtained compounds may be associated with both membrane activity and inhibition of the elongation factor EF-G.

Predicting locations of cryptic pockets from single protein structures using the PocketMiner graph neural network

Cryptic pockets expand the scope of drug discovery by enabling targeting of proteins currently considered undruggable because they lack pockets in their ground state structures. However, identifying cryptic pockets is labor-intensive and slow. The ability to accurately and rapidly predict if and where cryptic pockets are likely to form from a structure would greatly accelerate the search for druggable pockets. Here, we present PocketMiner, a graph neural network trained to predict where pockets are likely to open in molecular dynamics simulations. Applying PocketMiner to single structures from a newly curated dataset of 39 experimentally confirmed cryptic pockets demonstrates that it accurately identifies cryptic pockets (ROC-AUC: 0.87) >1,000-fold faster than existing methods. We apply PocketMiner across the human proteome and show that predicted pockets open in simulations, suggesting that over half of proteins thought to lack pockets based on available structures likely contain cryptic pockets, vastly expanding the potentially druggable proteome.

Artificial Intelligence-Based Computational Screening and Functional Assays Identify Candidate Small Molecule Antagonists of PTPmu-Dependent Adhesion

PTPmu (PTPµ) is a member of the receptor protein tyrosine phosphatase IIb family that participates in cell-cell adhesion and signaling. PTPmu is proteolytically downregulated in glioblastoma (glioma), and the resulting extracellular and intracellular fragments are believed to stimulate cancer cell growth and/or migration. Therefore, drugs targeting these fragments may have therapeutic potential. Here, we used the AtomNet^® platform, the first deep learning neural network for drug design and discovery, to screen a molecular library of several million compounds and identified 76 candidates predicted to interact with a groove between the MAM and Ig extracellular domains required for PTPmu-mediated cell adhesion. These candidates were screened in two cell-based assays: PTPmu-dependent aggregation of Sf9 cells and a tumor growth assay where glioma cells grow in three-dimensional spheres. Four compounds inhibited PTPmu-mediated aggregation of Sf9 cells, six compounds inhibited glioma sphere formation/growth, while two priority compounds were effective in both assays. The stronger of these two compounds inhibited PTPmu aggregation in Sf9 cells and inhibited glioma sphere formation down to 25 micromolar. Additionally, this compound was able to inhibit the aggregation of beads coated with an extracellular fragment of PTPmu, directly demonstrating an interaction. This compound presents an interesting starting point for the development of PTPmu-targeting agents for treating cancer including glioblastoma.

Application of Molecular Simulation Methods in Food Science: Status and Prospects

Molecular simulation methods, such as molecular docking, molecular dynamic (MD) simulation, and quantum chemical (QC) calculation, have become popular as characterization and/or virtual screening tools because they can visually display interaction details that in vitro experiments can not capture and quickly screen bioactive compounds from large databases with millions of molecules. Currently, interdisciplinary research has expanded molecular simulation technology from computer aided drug design (CADD) to food science. More food scientists are supporting their hypotheses/results with this technology. To understand better the use of molecular simulation methods, it is necessary to systematically summarize the latest applications and usage trends of molecular simulation methods in the research field of food science. However, this type of review article is rare. To bridge this gap, we have comprehensively summarized the principle, combination usage, and application of molecular simulation methods in food science. We also analyzed the limitations and future trends and offered valuable strategies with the latest technologies to help food scientists use molecular simulation methods.

Drug specificity and affinity are encoded in the probability of cryptic pocket opening in myosin motor domains

The design of compounds that can discriminate between closely related target proteins remains a central challenge in drug discovery. Specific therapeutics targeting the highly conserved myosin motor family are urgently needed as mutations in at least six of its members cause numerous diseases. Allosteric modulators, like the myosin-II inhibitor blebbistatin, are a promising means to achieve specificity. However, it remains unclear why blebbistatin inhibits myosin-II motors with different potencies given that it binds at a highly conserved pocket that is always closed in blebbistatin-free experimental structures. We hypothesized that the probability of pocket opening is an important determinant of the potency of compounds like blebbistatin. To test this hypothesis, we used Markov state models (MSMs) built from over 2 ms of aggregate molecular dynamics simulations with explicit solvent. We find that blebbistatin's binding pocket readily opens in simulations of blebbistatin-sensitive myosin isoforms. Comparing these conformational ensembles reveals that the probability of pocket opening correctly identifies which isoforms are most sensitive to blebbistatin inhibition and that docking against MSMs quantitatively predicts blebbistatin binding affinities (R2=0.82). In a blind prediction for an isoform (Myh7b) whose blebbistatin sensitivity was unknown, we find good agreement between predicted and measured IC50s (0.67 μM vs. 0.36 μM). Therefore, we expect this framework to be useful for the development of novel specific drugs across numerous protein targets.

Apo2ph4: A Versatile Workflow for the Generation of Receptor-based Pharmacophore Models for Virtual Screening

Pharmacophore models are widely used as efficient virtual screening (VS) filters for the target-directed enrichment of large compound libraries. However, the generation of pharmacophore models that have the power to discriminate between active and inactive molecules traditionally requires structural information about ligand-target complexes or at the very least knowledge of one active ligand. The fact that the discovery of the first known active ligand of a newly investigated target represents a major hurdle at the beginning of every drug discovery project underscores the need for methods that are able to derive high-quality pharmacophore models even without the prior knowledge of any active ligand structures. In this work, we introduce a novel workflow, called apo2ph4, that enables the rapid derivation of pharmacophore models solely from the three-dimensional structure of the target receptor. The utility of this workflow is demonstrated retrospectively for the generation of a pharmacophore model for the M2 muscarinic acetylcholine receptor. Furthermore, in order to show the general applicability of apo2ph4, the workflow was employed for all 15 targets of the recently published LIT-PCBA dataset. Pharmacophore-based VS runs using the apo2ph4-derived models achieved a significant enrichment of actives for 13 targets. In the last presented example, a pharmacophore model derived from the etomidate site of the α1β2γ2 GABAA receptor was used in VS campaigns. Subsequent in vitro testing of selected hits revealed that 19 out of 20 (95%) tested compounds were able to significantly enhance GABA currents, which impressively demonstrates the applicability of apo2ph4 for real-world drug design projects.

Prioritizing Small Sets of Molecules for Synthesis through in-silico Tools: A Comparison of Common Ranking Methods

Prioritizing molecules for synthesis is a key role of computational methods within medicinal chemistry. Multiple tools exist for ranking molecules, from the cheap and popular molecular docking methods to more computationally expensive molecular-dynamics (MD)-based methods. It is often questioned whether the accuracy of the more rigorous methods justifies the higher computational cost and associated calculation time. Here, we compared the performance on ranking the binding of small molecules for seven scoring functions from five docking programs, one end-point method (MM/GBSA), and two MD-based free energy methods (PMX, FEP+). We investigated 16 pharmaceutically relevant targets with a total of 423 known binders. The performance of docking methods for ligand ranking was strongly system dependent. We observed that MD-based methods predominantly outperformed docking algorithms and MM/GBSA calculations. Based on our results, we recommend the application of MD-based free energy methods for prioritization of molecules for synthesis in lead optimization, whenever feasible.

In Silico Structure-Based Vaccine Design

Structure-based vaccine design (SBVD) is an important technique in computational vaccine design that uses structural information on a targeted protein to design novel vaccine candidates. This increasing ability to rapidly model structural information on proteins and antibodies has provided the scientific community with many new vaccine targets and novel opportunities for future vaccine discovery. This chapter provides a comprehensive overview of the status of in silico SBVD and discusses the current challenges and limitations. Key strategies in the field of SBVD are exemplified by a case study on design of COVID-19 vaccines targeting SARS-CoV-2 spike protein.

Discovery of a cryptic pocket in the AI-predicted structure of PPM1D phosphatase explains the binding site and potency of its allosteric inhibitors

Virtual screening is a widely used tool for drug discovery, but its predictive power can vary dramatically depending on how much structural data is available. In the best case, crystal structures of a ligand-bound protein can help find more potent ligands. However, virtual screens tend to be less predictive when only ligand-free crystal structures are available, and even less predictive if a homology model or other predicted structure must be used. Here, we explore the possibility that this situation can be improved by better accounting for protein dynamics, as simulations started from a single structure have a reasonable chance of sampling nearby structures that are more compatible with ligand binding. As a specific example, we consider the cancer drug target PPM1D/Wip1 phosphatase, a protein that lacks crystal structures. High-throughput screens have led to the discovery of several allosteric inhibitors of PPM1D, but their binding mode remains unknown. To enable further drug discovery efforts, we assessed the predictive power of an AlphaFold-predicted structure of PPM1D and a Markov state model (MSM) built from molecular dynamics simulations initiated from that structure. Our simulations reveal a cryptic pocket at the interface between two important structural elements, the flap and hinge regions. Using deep learning to predict the pose quality of each docked compound for the active site and cryptic pocket suggests that the inhibitors strongly prefer binding to the cryptic pocket, consistent with their allosteric effect. The predicted affinities for the dynamically uncovered cryptic pocket also recapitulate the relative potencies of the compounds (τ_b = 0.70) better than the predicted affinities for the static AlphaFold-predicted structure (τ_b = 0.42). Taken together, these results suggest that targeting the cryptic pocket is a good strategy for drugging PPM1D and, more generally, that conformations selected from simulation can improve virtual screening when limited structural data is available.

Recent changes in the mutational dynamics of the SARS-CoV-2 main protease substantiate the danger of emerging resistance to antiviral drugs

Introduction

The current coronavirus pandemic is being combated worldwide by nontherapeutic measures and massive vaccination programs. Nevertheless, therapeutic options such as severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) main-protease (Mpro) inhibitors are essential due to the ongoing evolution toward escape from natural or induced immunity. While antiviral strategies are vulnerable to the effects of viral mutation, the relatively conserved Mpro makes an attractive drug target: Nirmatrelvir, an antiviral targeting its active site, has been authorized for conditional or emergency use in several countries since December 2021, and a number of other inhibitors are under clinical evaluation. We analyzed recent SARS-CoV-2 genomic data, since early detection of potential resistances supports a timely counteraction in drug development and deployment, and discovered accelerated mutational dynamics of Mpro since early December 2021.

Methods

We performed a comparative analysis of 10.5 million SARS-CoV-2 genome sequences available by June 2022 at GISAID to the NCBI reference genome sequence NC_045512.2. Amino-acid exchanges within high-quality regions in 69,878 unique Mpro sequences were identified and time- and in-depth sequence analyses including a structural representation of mutational dynamics were performed using in-house software.

Results

The analysis showed a significant recent event of mutational dynamics in Mpro. We report a remarkable increase in mutational variability in an eight-residue long consecutive region (R188-G195) near the active site since December 2021.

Discussion

The increased mutational variability in close proximity to an antiviral-drug binding site as described herein may suggest the onset of the development of antiviral resistance. This emerging diversity urgently needs to be further monitored and considered in ongoing drug development and lead optimization.

Classification of Protein-Binding Sites Using a Spherical Convolutional Neural Network

The analysis and comparison of protein-binding sites aid various applications in the drug discovery process, e.g., hit finding, drug repurposing, and polypharmacology. Classification of binding sites has been a hot topic for the past 30 years, and many different methods have been published. The rapid development of machine learning computational algorithms, coupled with the large volume of publicly available protein-ligand 3D structures, makes it possible to apply deep learning techniques in binding site comparison. Our method uses a cutting-edge spherical convolutional neural network based on the DeepSphere architecture to learn global representations of protein-binding sites. The model was trained on TOUGH-C1 and TOUGH-M1 data and validated with the ProSPECCTs datasets. Our results show that our model can (1) perform well in protein-binding site similarity and classification tasks and (2) learn and separate the physicochemical properties of binding sites. Lastly, we tested the model on a set of kinases, where the results show that it is able to cluster the different kinase subfamilies effectively. This example demonstrates the method's promise for lead hopping within or outside a protein target, directly based on binding site information.

A multilayer dynamic perturbation analysis method for predicting ligand-protein interactions

BACKGROUND

Ligand-protein interactions play a key role in defining protein function, and detecting natural ligands for a given protein is thus a very important bioengineering task. In particular, with the rapid development of AI-based structure prediction algorithms, batch structural models with high reliability and accuracy can be obtained at low cost, giving rise to the urgent requirement for the prediction of natural ligands based on protein structures. In recent years, although several structure-based methods have been developed to predict ligand-binding pockets and ligand-binding sites, accurate and rapid methods are still lacking, especially for the prediction of ligand-binding regions and the spatial extension of ligands in the pockets.

RESULTS

In this paper, we proposed a multilayer dynamics perturbation analysis (MDPA) method for predicting ligand-binding regions based solely on protein structure, which is an extended version of our previously developed fast dynamic perturbation analysis (FDPA) method. In MDPA/FDPA, ligand binding tends to occur in regions that cause large changes in protein conformational dynamics. MDPA, examined using a standard validation dataset of ligand-protein complexes, yielded an averaged ligand-binding site prediction Matthews coefficient of 0.40, with a prediction precision of at least 50% for 71% of the cases. In particular, for 80% of the cases, the predicted ligand-binding region overlaps the natural ligand by at least 50%. The method was also compared with other state-of-the-art structure-based methods.

CONCLUSIONS

MDPA is a structure-based method to detect ligand-binding regions on protein surface. Our calculations suggested that a range of spaces inside the protein pockets has subtle interactions with the protein, which can significantly impact on the overall dynamics of the protein. This work provides a valuable tool as a starting point upon which further docking and analysis methods can be used for natural ligand detection in protein functional annotation. The source code of MDPA method is freely available at: https://github.com/mingdengming/mdpa .

Estimating the Similarity between Protein Pockets

With the exponential increase in publicly available protein structures, the comparison of protein binding sites naturally emerged as a scientific topic to explain observations or generate hypotheses for ligand design, notably to predict ligand selectivity for on- and off-targets, explain polypharmacology, and design target-focused libraries. The current review summarizes the state-of-the-art computational methods applied to pocket detection and comparison as well as structural druggability estimates. The major strengths and weaknesses of current pocket descriptors, alignment methods, and similarity search algorithms are presented. Lastly, an exhaustive survey of both retrospective and prospective applications in diverse medicinal chemistry scenarios illustrates the capability of the existing methods and the hurdle that still needs to be overcome for more accurate predictions.

Genome interpretation using in silico predictors of variant impact

Estimating the effects of variants found in disease driver genes opens the door to personalized therapeutic opportunities. Clinical associations and laboratory experiments can only characterize a tiny fraction of all the available variants, leaving the majority as variants of unknown significance (VUS). In silico methods bridge this gap by providing instant estimates on a large scale, most often based on the numerous genetic differences between species. Despite concerns that these methods may lack reliability in individual subjects, their numerous practical applications over cohorts suggest they are already helpful and have a role to play in genome interpretation when used at the proper scale and context. In this review, we aim to gain insights into the training and validation of these variant effect predicting methods and illustrate representative types of experimental and clinical applications. Objective performance assessments using various datasets that are not yet published indicate the strengths and limitations of each method. These show that cautious use of in silico variant impact predictors is essential for addressing genome interpretation challenges.

The druggable genome: Twenty years later

The concept of the druggable genome has been with us for 20 years. During this time, researchers have developed several methods and resources to help assess a target's druggability. In parallel, evidence for target-disease associations has been collated at scale by Open Targets. More recently, the Protein Data Bank in Europe (PDBe) have built a knowledge base matching per-residue annotations with available protein structure. While each resource is useful in isolation, we believe there is enormous potential in bringing all relevant data into a single knowledge graph, from gene-level to protein residue. Automation is vital for the processing and assessment of all available structures. We have developed scalable, automated workflows that provide hotspot-based druggability assessments for all available structures across large numbers of targets. Ultimately, we will run our method at a proteome scale, an ambition made more realistic by the arrival of AlphaFold 2. Bringing together annotations from the residue up to the gene level and building connections within the graph to represent pathways or protein-protein interactions will create complexity that mirrors the biological systems they represent. Such complexity is difficult for the human mind to utilise effectively, particularly at scale. We believe that graph-based AI methods will be able to expertly navigate such a knowledge graph, selecting the targets of the future.

GraphSite: Ligand Binding Site Classification with Deep Graph Learning

The binding of small organic molecules to protein targets is fundamental to a wide array of cellular functions. It is also routinely exploited to develop new therapeutic strategies against a variety of diseases. On that account, the ability to effectively detect and classify ligand binding sites in proteins is of paramount importance to modern structure-based drug discovery. These complex and non-trivial tasks require sophisticated algorithms from the field of artificial intelligence to achieve a high prediction accuracy. In this communication, we describe GraphSite, a deep learning-based method utilizing a graph representation of local protein structures and a state-of-the-art graph neural network to classify ligand binding sites. Using neural weighted message passing layers to effectively capture the structural, physicochemical, and evolutionary characteristics of binding pockets mitigates model overfitting and improves the classification accuracy. Indeed, comprehensive cross-validation benchmarks against a large dataset of binding pockets belonging to 14 diverse functional classes demonstrate that GraphSite yields the class-weighted F1-score of 81.7%, outperforming other approaches such as molecular docking and binding site matching. Further, it also generalizes well to unseen data with the F1-score of 70.7%, which is the expected performance in real-world applications. We also discuss new directions to improve and extend GraphSite in the future.