graphLambda: Fusion Graph Neural Networks for Binding Affinity Prediction.

Predicting the binding affinity of protein-ligand complexes is crucial for computer-aided drug discovery (CADD) and the identification of potential drug candidates. The deep learning-based scoring functions have emerged as promising predictors of binding constants. Building on recent advancements in graph neural networks, we present graphLambda for protein-ligand binding affinity prediction, which utilizes graph convolutional, attention, and isomorphism blocks to enhance the predictive capabilities. The graphLambda model exhibits superior performance across CASF16 and CSAR HiQ NRC benchmarks and demonstrates robustness with respect to different types of train-validation set partitions. The development of graphLambda underscores the potential of graph neural networks in advancing binding affinity prediction models, contributing to more effective CADD methodologies.

Characterizing conformational states in GPCR structures using machine learning.

G protein-coupled receptors (GPCRs) play a pivotal role in signal transduction and represent attractive targets for drug development. Recent advances in structural biology have provided insights into GPCR conformational states, which are critical for understanding their signaling pathways and facilitating structure-based drug discovery. In this study, we introduce a machine learning approach for conformational state annotation of GPCRs. We represent GPCR conformations as high-dimensional feature vectors, incorporating information about amino acid residue pairs involved in the activation pathway. Using a dataset of GPCR conformations in inactive and active states obtained through molecular dynamics simulations, we trained machine learning models to distinguish between inactive-like and active-like conformations. The developed model provides interpretable predictions and can be used for the large-scale analysis of molecular dynamics trajectories of GPCRs.

A community effort in SARS-CoV-2 drug discovery.

The COVID-19 pandemic continues to pose a substantial threat to human lives and is likely to do so for years to come. Despite the availability of vaccines, searching for efficient small-molecule drugs that are widely available, including in low- and middle-income countries, is an ongoing challenge. In this work, we report the results of an open science community effort, the "Billion molecules against COVID-19 challenge", to identify small-molecule inhibitors against SARS-CoV-2 or relevant human receptors. Participating teams used a wide variety of computational methods to screen a minimum of 1 billion virtual molecules against 6 protein targets. Overall, 31 teams participated, and they suggested a total of 639,024 molecules, which were subsequently ranked to find 'consensus compounds'. The organizing team coordinated with various contract research organizations (CROs) and collaborating institutions to synthesize and test 878 compounds for biological activity against proteases (Nsp5, Nsp3, TMPRSS2), nucleocapsid N, RdRP (only the Nsp12 domain), and (alpha) spike protein S. Overall, 27 compounds with weak inhibition/binding were experimentally identified by binding-, cleavage-, and/or viral suppression assays and are presented here. Open science approaches such as the one presented here contribute to the knowledge base of future drug discovery efforts in finding better SARS-CoV-2 treatments.

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.

Evaluating and comparing success rates for inpatient treatment of alcohol addiction in the Czech Republic.

OBJECTIVES: This systematic review seeks to present and compare data from studies evaluating the success of medium-term inpatient treatment of alcohol-dependent patients in the Czech Republic. Another aim was to identify the problems that make such comparisons difficult. No previous review comparing the efficiency of various therapeutic programmes has been published in the Czech Republic. METHODS: Bibliographia medica Cechoslovaca and PubMed were used to find studies published in professional medical journals since 1970 evaluating the abstinence of patients who voluntarily completed medium-term inpatient treatment of alcohol dependence. RESULTS: Medium-term inpatient treatment of alcohol addiction leads to one year of abstinence in 34% to 76% of patients. Such variance in value is largely caused by selection bias, differences in the definition of abstinence, and differences in data collection methods. CONCLUSION: The comparison of studies presented many challenges. Further steps should be taken to help compare treatment programmes in the future, as the programmes provide different therapeutic interventions of different intensities and lengths to different patients. Adequate demographic and other pretreatment characteristics data collection, detailed descriptions of therapeutic interventions, and identification of effective components of the therapeutic programme could support further research in this area, optimize existing programmes, and increase the overall treatment efficiency.

Mirror proteorhodopsins.

Proteorhodopsins (PRs), bacterial light-driven outward proton pumps comprise the first discovered and largest family of rhodopsins, they play a significant role in life on the Earth. A big remaining mystery was that up-to-date there was no described bacterial rhodopsins pumping protons at acidic pH despite the fact that bacteria live in different pH environment. Here we describe conceptually new bacterial rhodopsins which are operating as outward proton pumps at acidic pH. A comprehensive function-structure study of a representative of a new clade of proton pumping rhodopsins which we name "mirror proteorhodopsins", from Sphingomonas paucimobilis (SpaR) shows cavity/gate architecture of the proton translocation pathway rather resembling channelrhodopsins than the known rhodopsin proton pumps. Another unique property of mirror proteorhodopsins is that proton pumping is inhibited by a millimolar concentration of zinc. We also show that mirror proteorhodopsins are extensively represented in opportunistic multidrug resistant human pathogens, plant growth-promoting and zinc solubilizing bacteria. They may be of optogenetic interest.

Molecular mechanism of biased signaling at the kappa opioid receptor.

The κ-opioid receptor (KOR) has emerged as an attractive drug target for pain management without addiction, and biased signaling through particular pathways of KOR may be key to maintaining this benefit while minimizing side-effect liabilities. As for most G protein-coupled receptors (GPCRs), however, the molecular mechanisms of ligand-specific signaling at KOR have remained unclear. To better understand the molecular determinants of KOR signaling bias, we apply structure determination, atomic-level molecular dynamics (MD) simulations, and functional assays. We determine a crystal structure of KOR bound to the G protein-biased agonist nalfurafine, the first approved KOR-targeting drug. We also identify an arrestin-biased KOR agonist, WMS-X600. Using MD simulations of KOR bound to nalfurafine, WMS-X600, and a balanced agonist U50,488, we identify three active-state receptor conformations, including one that appears to favor arrestin signaling over G protein signaling and another that appears to favor G protein signaling over arrestin signaling. These results, combined with mutagenesis validation, provide a molecular explanation of how agonists achieve biased signaling at KOR.

Structural basis for receptor selectivity and inverse agonism in S1P5 receptors.

The bioactive lysophospholipid sphingosine-1-phosphate (S1P) acts via five different subtypes of S1P receptors (S1PRs) - S1P1-5. S1P5 is predominantly expressed in nervous and immune systems, regulating the egress of natural killer cells from lymph nodes and playing a role in immune and neurodegenerative disorders, as well as carcinogenesis. Several S1PR therapeutic drugs have been developed to treat these diseases; however, they lack receptor subtype selectivity, which leads to side effects. In this article, we describe a 2.2 Å resolution room temperature crystal structure of the human S1P5 receptor in complex with a selective inverse agonist determined by serial femtosecond crystallography (SFX) at the Pohang Accelerator Laboratory X-Ray Free Electron Laser (PAL-XFEL) and analyze its structure-activity relationship data. The structure demonstrates a unique ligand-binding mode, involving an allosteric sub-pocket, which clarifies the receptor subtype selectivity and provides a template for structure-based drug design. Together with previously published S1PR structures in complex with antagonists and agonists, our structure with S1P5-inverse agonist sheds light on the activation mechanism and reveals structural determinants of the inverse agonism in the S1PR family.

Biphenyl scaffold for the design of NMDA-receptor negative modulators: molecular modeling, synthesis, and biological activity.

NMDA (N-methyl-d-aspartate) receptor antagonists are promising tools for the treatment of a wide variety of central nervous system impairments including major depressive disorder. We present here the activity optimization process of a biphenyl-based NMDA negative allosteric modulator (NAM) guided by free energy calculations, which led to a 100 times activity improvement (IC50 = 50 nM) compared to a hit compound identified in virtual screening. Preliminary calculation results suggest a low affinity for the human ether-a-go-go-related gene ion channel (hERG), a high affinity for which was earlier one of the main obstacles for the development of first-generation NMDA-receptor negative allosteric modulators. The docking study and the molecular dynamics calculations suggest a completely different binding mode (ifenprodil-like) compared to another biaryl-based NMDA NAM EVT-101.

Inactive and active state structures template selective tools for the human 5-HT5A receptor.

Serotonin receptors are important targets for established therapeutics and drug development as they are expressed throughout the human body and play key roles in cell signaling. There are 12 serotonergic G protein-coupled receptor members encoded in the human genome, of which the 5-hydroxytryptamine (5-HT)5A receptor (5-HT5AR) is the least understood and lacks selective tool compounds. Here, we report four high-resolution (2.73-2.80 Å) structures of human 5-HT5ARs, including an inactive state structure bound to an antagonist AS2674723 by crystallization and active state structures bound to a partial agonist lisuride and two full agonists, 5-carboxamidotryptamine (5-CT) and methylergometrine, by cryo-EM. Leveraging the new structures, we developed a highly selective and potent antagonist for 5-HT5AR. Collectively, these findings both enhance our understanding of this enigmatic receptor and provide a roadmap for structure-based drug discovery for 5-HT5AR.

Structure-based deep learning for binding site detection in nucleic acid macromolecules.

Structure-based drug design (SBDD) targeting nucleic acid macromolecules, particularly RNA, is a gaining momentum research direction that already resulted in several FDA-approved compounds. Similar to proteins, one of the critical components in SBDD for RNA is the correct identification of the binding sites for putative drug candidates. RNAs share a common structural organization that, together with the dynamic nature of these molecules, makes it challenging to recognize binding sites for small molecules. Moreover, there is a need for structure-based approaches, as sequence information only does not consider conformation plasticity of nucleic acid macromolecules. Deep learning holds a great promise to resolve binding site detection problem, but requires a large amount of structural data, which is very limited for nucleic acids, compared to proteins. In this study we composed a set of ∼2000 nucleic acid-small molecule structures comprising ∼2500 binding sites, which is ∼40-times larger than previously used one, and demonstrated the first structure-based deep learning approach, BiteNet N , to detect binding sites in nucleic acid structures. BiteNet N operates with arbitrary nucleic acid complexes, shows the state-of-the-art performance, and can be helpful in the analysis of different conformations and mutant variants, as we demonstrated for HIV-1 TAR RNA and ATP-aptamer case studies.

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

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

Structural insights on ligand recognition at the human leukotriene B4 receptor 1.

The leukotriene B4 receptor 1 (BLT1) regulates the recruitment and chemotaxis of different cell types and plays a role in the pathophysiology of infectious, allergic, metabolic, and tumorigenic human diseases. Here we present a crystal structure of human BLT1 (hBLT1) in complex with a selective antagonist MK-D-046, developed for the treatment of type 2 diabetes and other inflammatory conditions. Comprehensive analysis of the structure and structure-activity relationship data, reinforced by site-directed mutagenesis and docking studies, reveals molecular determinants of ligand binding and selectivity toward different BLT receptor subtypes and across species. The structure helps to identify a putative membrane-buried ligand access channel as well as potential receptor binding modes of endogenous agonists. These structural insights of hBLT1 enrich our understanding of its ligand recognition and open up future avenues in structure-based drug design.

Mutations in transmembrane proteins: diseases, evolutionary insights, prediction and comparison with globular proteins.

Membrane proteins are unique in that they interact with lipid bilayers, making them indispensable for transporting molecules and relaying signals between and across cells. Due to the significance of the protein's functions, mutations often have profound effects on the fitness of the host. This is apparent both from experimental studies, which implicated numerous missense variants in diseases, as well as from evolutionary signals that allow elucidating the physicochemical constraints that intermembrane and aqueous environments bring. In this review, we report on the current state of knowledge acquired on missense variants (referred to as to single amino acid variants) affecting membrane proteins as well as the insights that can be extrapolated from data already available. This includes an overview of the annotations for membrane protein variants that have been collated within databases dedicated to the topic, bioinformatics approaches that leverage evolutionary information in order to shed light on previously uncharacterized membrane protein structures or interaction interfaces, tools for predicting the effects of mutations tailored specifically towards the characteristics of membrane proteins as well as two clinically relevant case studies explaining the implications of mutated membrane proteins in cancer and cardiomyopathy.

Spatiotemporal identification of druggable binding sites using deep learning.

Identification of novel protein binding sites expands druggable genome and opens new opportunities for drug discovery. Generally, presence or absence of a binding site depends on the three-dimensional conformation of a protein, making binding site identification resemble the object detection problem in computer vision. Here we introduce a computational approach for the large-scale detection of protein binding sites, that considers protein conformations as 3D-images, binding sites as objects on these images to detect, and conformational ensembles of proteins as 3D-videos to analyze. BiteNet is suitable for spatiotemporal detection of hard-to-spot allosteric binding sites, as we showed for conformation-specific binding site of the epidermal growth factor receptor, oligomer-specific binding site of the ion channel, and binding site in G protein-coupled receptor. BiteNet outperforms state-of-the-art methods both in terms of accuracy and speed, taking about 1.5 minutes to analyze 1000 conformations of a protein with ~2000 atoms.

graphDelta: MPNN Scoring Function for the Affinity Prediction of Protein-Ligand Complexes.

In this work, we present graph-convolutional neural networks for the prediction of binding constants of protein-ligand complexes. We derived the model using multi task learning, where the target variables are the dissociation constant (K d), inhibition constant (K i), and half maximal inhibitory concentration (IC50). Being rigorously trained on the PDBbind dataset, the model achieves the Pearson correlation coefficient of 0.87 and the RMSE value of 1.05 in pK units, outperforming recently developed 3D convolutional neural network model K deep.

The structural study of mutation-induced inactivation of human muscarinic receptor M4.

Human muscarinic receptor M4 belongs to the class A subfamily of the G-protein-coupled receptors (GPCRs). M4 has emerged as an attractive drug target for the treatment of Alzheimer's disease and schizophrenia. Recent results showed that M4-mediated cholinergic transmission is related to motor symptoms in Parkinson's disease. Selective ligand design for the five muscarinic acetylcholine receptor (mAchR) subtypes currently remains challenging owing to the high sequence and structural similarity of their orthosteric binding pockets. In order to obtain M4-selective antagonists, a new approach was tried to lock M4 into an inactive form by rationally designing an N4497.49R mutation, which mimics the allosteric sodium binding in the conserved sodium site usually found in class A GPCRs. In addition, the crystal structure of the mutation-induced inactive M4 was determined. By comparative analysis with other mAchR structures, followed by functional assays, the N4497.49R mutation was shown to stabilize M4 into an inactive state. Virtual screening of a focused ligand library using the crystal structure showed that the inactive M4 prefers antagonists much more than agonists. This study provides a powerful mutation strategy to stabilize GPCRs in inactive states and facilitate their structure determination.

Structural basis of ligand selectivity and disease mutations in cysteinyl leukotriene receptors.

Cysteinyl leukotriene G protein-coupled receptors CysLT1 and CysLT2 regulate pro-inflammatory responses associated with allergic disorders. While selective inhibition of CysLT1R has been used for treating asthma and associated diseases for over two decades, CysLT2R has recently started to emerge as a potential drug target against atopic asthma, brain injury and central nervous system disorders, as well as several types of cancer. Here, we describe four crystal structures of CysLT2R in complex with three dual CysLT1R/CysLT2R antagonists. The reported structures together with the results of comprehensive mutagenesis and computer modeling studies shed light on molecular determinants of CysLTR ligand selectivity and specific effects of disease-related single nucleotide variants.

Structure-based mechanism of cysteinyl leukotriene receptor inhibition by antiasthmatic drugs.

The G protein-coupled cysteinyl leukotriene receptor CysLT1R mediates inflammatory processes and plays a major role in numerous disorders, including asthma, allergic rhinitis, cardiovascular disease, and cancer. Selective CysLT1R antagonists are widely prescribed as antiasthmatic drugs; however, these drugs demonstrate low effectiveness in some patients and exhibit a variety of side effects. To gain deeper understanding into the functional mechanisms of CysLTRs, we determined the crystal structures of CysLT1R bound to two chemically distinct antagonists, zafirlukast and pranlukast. The structures reveal unique ligand-binding modes and signaling mechanisms, including lateral ligand access to the orthosteric pocket between transmembrane helices TM4 and TM5, an atypical pattern of microswitches, and a distinct four-residue-coordinated sodium site. These results provide important insights and structural templates for rational discovery of safer and more effective drugs.

Prediction of disease-associated mutations in the transmembrane regions of proteins with known 3D structure.

Being able to assess the phenotypic effects of mutations is a much required capability in precision medicine. However, most of the currently available structure-based methods actually predict stability changes caused by mutations rather than their pathogenic potential. There are also no dedicated methods to predict damaging mutations specifically in transmembrane proteins. In this study we developed and applied a machine-learning approach to discriminate between disease-associated and benign point mutations in the transmembrane regions of proteins with known 3D structure. The method, called BorodaTM (BOosted RegressiOn trees for Disease-Associated mutations in TransMembrane proteins), was trained on sequence-, structure-, and energy-derived descriptors. When compared with the state-of-the-art methods, BorodaTM is superior in classifying point mutations in transmembrane regions. Using BorodaTM we have conducted a large-scale mutation analysis in the transmembrane regions of human proteins with known 3D structures. For each protein we generated structural models for all point mutations by replacing each residue to 19 possible residue types. We classified ~1.8 millions point mutations as benign or diseased-associated and made all predictions available as a Web-server at https://www.iitm.ac.in/bioinfo/MutHTP/boroda.php.