Search

The interpretation of ligand–target interactions at atomistic resolution is central to most efforts in computational drug discovery and optimization. However, the highly dynamic nature of protein targets, as well as possible induced fit effects, makes difficult to sample many interactions effectively with docking studies or even with large-scale molecular dynamics (MD) simulations. We propose a novel application of Self-Organizing Maps (SOMs) to address the sampling and dynamic mapping tasks, particularly in cases involving ligand flexibility and induced fit. The SOM approach offers a data-driven strategy to create a map of the interaction process and pathways based on unbiased MD. Furthermore, we show how the preliminary SOM mapping is complementary to kinetic analysis, with the employment of both network-based approaches and Markov state models. We demonstrate the method by comprehensively mapping a large dataset of 640 μs of unbiased trajectories sampling the recognition process between the phosphorylated YEEI peptide and its high-specificity target lck-SH2. The integration of SOM into unbiased simulation protocols significantly advances our understanding of the ligand binding mechanism. This approach serves as a potent tool for mapping intricate ligand–target interactions with unprecedented detail, thereby enhancing the characterization of kinetic properties crucial to drug design. [ABSTRACT FROM AUTHOR]

The Structure and TOpology Replica Molecular Mechanics (STORMM) code is a next-generation molecular simulation engine and associated libraries optimized for performance on fast, vectorized central processor units and graphics processing units (GPUs) with independent memory and tens of thousands of threads. STORMM is built to run thousands of independent molecular mechanical calculations on a single GPU with novel implementations that tune numerical precision, mathematical operations, and scarce on-chip memory resources to optimize throughput. The libraries are built around accessible classes with detailed documentation, supporting fine-grained parallelism and algorithm development as well as copying or swapping groups of systems on and off of the GPU. A primary intention of the STORMM libraries is to provide developers of atomic simulation methods with access to a high-performance molecular mechanics engine with extensive facilities to prototype and develop bespoke tools aimed toward drug discovery applications. In its present state, STORMM delivers molecular dynamics simulations of small molecules and small proteins in implicit solvent with tens to hundreds of times the throughput of conventional codes. The engineering paradigm transforms two of the most memory bandwidth-intensive aspects of condensed-phase dynamics, particle–mesh mapping, and valence interactions, into compute-bound problems for several times the scalability of existing programs. Numerical methods for compressing and streamlining the information present in stored coordinates and lookup tables are also presented, delivering improved accuracy over methods implemented in other molecular dynamics engines. The open-source code is released under the MIT license. [ABSTRACT FROM AUTHOR]

Datafication is driving organizations to invest in data commons, not only to share the costs of data generation, analysis, and curation but, more importantly, to realize synergies in precompetitive research collaborations where private and public motives interact (i.e., semicommons). The fanfare surrounding datafication often hails the sophisticated algorithms used to develop large quantities of data toward greater insight, naively assuming that more data equals better data. Yet for datafication in general and precompetitive research specifically, less attention is awarded to what actually constitutes data and evidence in the first place—that is, to its genesis, construction, and interpretation by heterogeneous scientific and commercial entities. We present the case of Open Targets, a precompetitive collaboration in the life sciences, where publicly funded research, nonprofit foundations, and for-profit pharma collaborate to generate and share data in genomics, proteomics, and bioinformatics. We theorize about the process of data commoning, a political activity in the semicommons where data are created, evidential value is assembled, and scientific meaning converges as data travels, or journeys, across creators, validators, and users. Our findings highlight the effects of relational dynamics and the political nature of data journeys: why these dynamics form, how they are manifested in a precompetitive semicommons, and what implications this can have for the mobility of data as a shared, public good. [ABSTRACT FROM AUTHOR]

In pharmaceutical discovery, the "magic methyl" effect describes a substantial improvement in the pharmacological properties of a drug candidate with the incorporation of methyl groups. Therefore, to expedite the synthesis of methylated drug analogs, late-stage, undirected methylations of C(sp3)-H bonds in complex molecules would be valuable. However, current methods for site-selective methylations are limited to activated C(sp3)-H bonds. Here we describe a site-selective, undirected methylation of unactivated C(sp3)-H bonds, enabled by photochemically activated peroxides and a nickel(II) complex whose turnover is enhanced by an ancillary ligand. The methodology displays compatibility with a wide range of functional groups and a high selectivity for tertiary C-H bonds, making it suitable for the late-stage methylation of complex organic compounds that contain multiple alkyl C-H bonds, such as terpene natural products, peptides, and active pharmaceutical ingredients. Overall, this method provides a synthetic tool to explore the "magic methyl" effect in drug discovery.

Drug discovery relies on efficient identification of small-molecule leads and their interactions with macromolecular targets. However, understanding how chemotypes impact mechanistically important conformational states often remains secondary among high-throughput discovery methods. Here, we present a conformational discovery pipeline integrating time-resolved, high-throughput small-angle X-ray scattering (TR-HT-SAXS) and classic fragment screening applied to allosteric states of the mitochondrial import oxidoreductase apoptosis-inducing factor (AIF). By monitoring oxidized and X-ray-reduced AIF states, TR-HT-SAXS leverages structure and kinetics to generate a multidimensional screening dataset that identifies fragment chemotypes allosterically stimulating AIF dimerization. Fragment-induced dimerization rates, quantified with time-resolved SAXS similarity analysis (kVR), capture structure-activity relationships (SAR) across the top-ranked 4-aminoquinoline chemotype. Crystallized AIF-aminoquinoline complexes validate TR-SAXS-guided SAR, supporting this conformational chemotype for optimization. AIF-aminoquinoline structures and mutational analysis reveal active site F482 as an underappreciated allosteric stabilizer of AIF dimerization. This conformational discovery pipeline illustrates TR-HT-SAXS as an effective technology for targeting chemical leads to important macromolecular states.

The cystic fibrosis transmembrane conductance regulator (CFTR) is a crucial ion channel whose loss of function leads to cystic fibrosis, whereas its hyperactivation leads to secretory diarrhea. Small molecules that improve CFTR folding (correctors) or function (potentiators) are clinically available. However, the only potentiator, ivacaftor, has suboptimal pharmacokinetics and inhibitors have yet to be clinically developed. Here, we combine molecular docking, electrophysiology, cryo-EM, and medicinal chemistry to identify CFTR modulators. We docked ∼155 million molecules into the potentiator site on CFTR, synthesized 53 test ligands, and used structure-based optimization to identify candidate modulators. This approach uncovered mid-nanomolar potentiators, as well as inhibitors, that bind to the same allosteric site. These molecules represent potential leads for the development of more effective drugs for cystic fibrosis and secretory diarrhea, demonstrating the feasibility of large-scale docking for ion channel drug discovery.

In drug discovery, the in silico prediction of binding affinity is one of the major means to prioritize compounds for synthesis. Alchemical relative binding free energy (RBFE) calculations based on molecular dynamics (MD) simulations are nowadays a popular approach for the accurate affinity ranking of compounds. MD simulations rely on empirical force field parameters, which strongly influence the accuracy of the predicted affinities. Here, we evaluate the ability of six different small-molecule force fields to predict experimental protein-ligand binding affinities in RBFE calculations on a set of 598 ligands and 22 protein targets. The public force fields OpenFF Parsley and Sage, GAFF, and CGenFF show comparable accuracy, while OPLS3e is significantly more accurate. However, a consensus approach using Sage, GAFF, and CGenFF leads to accuracy comparable to OPLS3e. While Parsley and Sage are performing comparably based on aggregated statistics across the whole dataset, there are differences in terms of outliers. Analysis of the force field reveals that improved parameters lead to significant improvement in the accuracy of affinity predictions on subsets of the dataset involving those parameters. Lower accuracy can not only be attributed to the force field parameters but is also dependent on input preparation and sampling convergence of the calculations. Especially large perturbations and nonconverged simulations lead to less accurate predictions. The input structures, Gromacs force field files, as well as the analysis Python notebooks are available on GitHub.

BackgroundRestraining or slowing ageing hallmarks at the cellular level have been proposed as a route to increased organismal lifespan and healthspan. Consequently, there is great interest in anti-ageing drug discovery. However, this currently requires laborious and lengthy longevity analysis. Here, we present a novel screening readout for the expedited discovery of compounds that restrain ageing of cell populations in vitro and enable extension of in vivo lifespan.MethodsUsing Illumina methylation arrays, we monitored DNA methylation changes accompanying long-term passaging of adult primary human cells in culture. This enabled us to develop, test, and validate the CellPopAge Clock, an epigenetic clock with underlying algorithm, unique among existing epigenetic clocks for its design to detect anti-ageing compounds in vitro. Additionally, we measured markers of senescence and performed longevity experiments in vivo in Drosophila, to further validate our approach to discover novel anti-ageing compounds. Finally, we bench mark our epigenetic clock with other available epigenetic clocks to consolidate its usefulness and specialisation for primary cells in culture.ResultsWe developed a novel epigenetic clock, the CellPopAge Clock, to accurately monitor the age of a population of adult human primary cells. We find that the CellPopAge Clock can detect decelerated passage-based ageing of human primary cells treated with rapamycin or trametinib, well-established longevity drugs. We then utilise the CellPopAge Clock as a screening tool for the identification of compounds which decelerate ageing of cell populations, uncovering novel anti-ageing drugs, torin2 and dactolisib (BEZ-235). We demonstrate that delayed epigenetic ageing in human primary cells treated with anti-ageing compounds is accompanied by a reduction in senescence and ageing biomarkers. Finally, we extend our screening platform in vivo by taking advantage of a specially formulated holidic medium for increased drug bioavailability in Drosophila. We show that the novel anti-ageing drugs, torin2 and dactolisib (BEZ-235), increase longevity in vivo.ConclusionsOur method expands the scope of CpG methylation profiling to accurately and rapidly detecting anti-ageing potential of drugs using human cells in vitro, and in vivo, providing a novel accelerated discovery platform to test sought after anti-ageing compounds and geroprotectors.

We previously reported trisubstituted pyrimidine lead compounds, namely, ARN22089 and ARN25062, which block the interaction between CDC42 with its specific downstream effector, a PAK protein. This interaction is crucial for the progression of multiple tumor types. Such inhibitors showed anticancer efficacy in vivo. Here, we describe a second class of CDC42 inhibitors with favorable drug-like properties. Out of the 25 compounds here reported, compound 15 (ARN25499) stands out as the best lead compound with an improved pharmacokinetic profile, increased bioavailability, and efficacy in an in vivo PDX tumor mouse model. Our results indicate that these CDC42 inhibitors represent a promising chemical class toward the discovery of anticancer drugs, with ARN25499 as an additional lead candidate for preclinical development.

DNA-encoded library technology grants access to nearly infinite opportunities to explore the chemical structure space for drug discovery. Successful navigation depends on the design and synthesis of libraries with appropriate physicochemical properties (PCPs) and structural diversity while aligning with practical considerations. To this end, we analyze combinatorial library design constraints including the number of chemistry cycles, bond construction strategies, and building block (BB) class selection in pursuit of ideal library designs. We compare two-cycle library designs (amino acid + carboxylic acid, primary amine + carboxylic acid) in the context of PCPs and chemical space coverage, given different BB selection strategies and constraints. We find that broad availability of amines and acids is essential for enabling the widest exploration of chemical space. Surprisingly, cost is not a driving factor, and virtually, the same chemical space can be explored with budget BBs.

The global surge in multi-drug resistant bacteria, including extended-spectrum β-lactamase (ESBL)-producing Escherichia coli has led to a growing need for new antibacterial compounds. Despite being promising, the potential of fish-derived antimicrobial peptides (AMPs) in combating ESBL-producing E. coli is largely unexplored. In this study, native African catfish antimicrobial peptides (NACAPs) were extracted from the skin mucus of farmed African catfish, Clarias gariepinus, using a combination of 10% acetic acid solvent hydrolysis, 5 kDa ultrafiltration, and C18 hydrophobic interaction chromatography. Peptides were then sequenced using Orbitrap Fusion Lumos Tribrid Mass Spectrometry. The identified peptides were screened for potential antibacterial activity using Random Forest and AdaBoost machine learning algorithms. The most promising peptide was chemically synthesized and evaluated in vitro for safety on rabbit red blood cells and activity against ESBL-producing E. coli (ATCC 35218) utilizing spot-on-lawn and broth dilution methods. Eight peptides ranging from 13 to 22 amino acids with molecular weights between 968.42 and 2434.11 Da were identified. Peptide NACAP-II was non-hemolytic to rabbit erythrocytes (p > 0.05) with a zone of inhibition (ZOI) of 22.7 ± 0.9 mm and a minimum inhibitory concentration (MIC) of 91.3 ± 1.2 μg/mL. The peptide is thus a candidate antibacterial compound with enormous potential applications in the pharmaceutical industry. However, further studies are still required to establish an upscale production strategy and optimize its activity and safety in vivo.

We present a novel small molecule antiviral chemotype that was identified by an unconventional cell-free protein synthesis and assembly-based phenotypic screen for modulation of viral capsid assembly. Activity of PAV-431, a representative compound from the series, has been validated against infectious viruses in multiple cell culture models for all six families of viruses causing most respiratory diseases in humans. In animals, this chemotype has been demonstrated efficacious for porcine epidemic diarrhoea virus (a coronavirus) and respiratory syncytial virus (a paramyxovirus). PAV-431 is shown to bind to the protein 14-3-3, a known allosteric modulator. However, it only appears to target the small subset of 14-3-3 which is present in a dynamic multi-protein complex whose components include proteins implicated in viral life cycles and in innate immunity. The composition of this target multi-protein complex appears to be modified upon viral infection and largely restored by PAV-431 treatment. An advanced analog, PAV-104, is shown to be selective for the virally modified target, thereby avoiding host toxicity. Our findings suggest a new paradigm for understanding, and drugging, the host-virus interface, which leads to a new clinical therapeutic strategy for treatment of respiratory viral disease.

High throughput screening (HTS) is routinely used to identify bioactive small molecules. This requires physical compounds, which limits coverage of accessible chemical space. Computational approaches combined with vast on-demand chemical libraries can access far greater chemical space, provided that the predictive accuracy is sufficient to identify useful molecules. Through the largest and most diverse virtual HTS campaign reported to date, comprising 318 individual projects, we demonstrate that our AtomNet® convolutional neural network successfully finds novel hits across every major therapeutic area and protein class. We address historical limitations of computational screening by demonstrating success for target proteins without known binders, high-quality X-ray crystal structures, or manual cherry-picking of compounds. We show that the molecules selected by the AtomNet® model are novel drug-like scaffolds rather than minor modifications to known bioactive compounds. Our empirical results suggest that computational methods can substantially replace HTS as the first step of small-molecule drug discovery.

Recent advancements in AI-driven technologies, particularly in protein structure prediction, are significantly reshaping the landscape of drug discovery and development. This review focuses on the question of how these technological breakthroughs, exemplified by AlphaFold2, are revolutionizing our understanding of protein structure and function changes underlying cancer and improve our approaches to counter them. By enhancing the precision and speed at which drug targets are identified and drug candidates can be designed and optimized, these technologies are streamlining the entire drug development process. We explore the use of AlphaFold2 in cancer drug development, scrutinizing its efficacy, limitations, and potential challenges. We also compare AlphaFold2 with other algorithms like ESMFold, explaining the diverse methodologies employed in this field and the practical effects of these differences for the application of specific algorithms. Additionally, we discuss the broader applications of these technologies, including the prediction of protein complex structures and the generative AI-driven design of novel proteins.

Auranofin, an FDA-approved drug for rheumatoid arthritis, has emerged as a promising antiparasitic medication in recent years. The gold(I) ion in auranofin is postulated to be responsible for its antiparasitic activity. Notably, aurothiomalate and aurothioglucose also contain gold(I), and, like auranofin, they were previously used to treat rheumatoid arthritis. Whether they have antiparasitic activity remains to be elucidated. Herein, we demonstrated that auranofin and similar derivatives, but not aurothiomalate and aurothioglucose, inhibited the growth of Toxoplasma gondii in vitro. We found that auranofin affected the T. gondii biological cycle (lytic cycle) by inhibiting T. gondiis invasion and triggering its egress from the host cell. However, auranofin could not prevent parasite replication once T. gondii resided within the host. Auranofin treatment induced apoptosis in T. gondii parasites, as demonstrated by its reduced size and elevated phosphatidylserine externalization (PS). Notably, the gold from auranofin enters the cytoplasm of T. gondii, as demonstrated by scanning transmission electron microscopy-energy dispersive X-ray spectroscopy (STEM-EDS) and Inductively Coupled Plasma-Mass Spectrometry (ICP-MS).IMPORTANCEToxoplasmosis, caused by Toxoplasma gondii, is a devastating disease affecting the brain and the eyes, frequently affecting immunocompromised individuals. Approximately 60 million people in the United States are already infected with T. gondii, representing a population at-risk of developing toxoplasmosis. Recent advances in treating cancer, autoimmune diseases, and organ transplants have contributed to this at-risk populations exponential growth. Paradoxically, treatments for toxoplasmosis have remained the same for more than 60 years, relying on medications well-known for their bone marrow toxicity and allergic reactions. Discovering new therapies is a priority, and repurposing FDA-approved drugs is an alternative approach to speed up drug discovery. Herein, we report the effect of auranofin, an FDA-approved drug, on the biological cycle of T. gondii and how both the phosphine ligand and the gold molecule determine the anti-parasitic activity of auranofin and other gold compounds. Our studies would contribute to the pipeline of candidate anti-T. gondii agents.

Preclinical models are ubiquitous and essential for drug discovery, yet our understanding of how well they translate to clinical outcomes is limited. In this study, we investigate the translational success of treatments for Clostridium difficile infection from animal models to human patients. Our analysis shows that only 36% of the preclinical and clinical experiment pairs result in translation success. Univariate analysis shows that the sustained response endpoint is correlated with translation failure (SRC = -0.20, p-value = 1.53 × 10-54), and explainability analysis of multi-variate random forest models shows that both sustained response endpoint and subject age are negative predictors of translation success. We have developed a recommendation system to help plan the right preclinical study given factors such as drug dosage, bacterial dosage, and preclinical/clinical endpoint. With an accuracy of 0.76 (F1 score of 0.71) and by using only 7 features (out of 68 total), the proposed system boosts translational efficiency by 25%. The method presented can extend to any disease and can serve as a preclinical to clinical translation decision support system to accelerate drug discovery and de-risk clinical outcomes.

Structure-based virtual screening is a key tool in early drug discovery, with growing interest in the screening of multi-billion chemical compound libraries. However, the success of virtual screening crucially depends on the accuracy of the binding pose and binding affinity predicted by computational docking. Here we develop a highly accurate structure-based virtual screen method, RosettaVS, for predicting docking poses and binding affinities. Our approach outperforms other state-of-the-art methods on a wide range of benchmarks, partially due to our ability to model receptor flexibility. We incorporate this into a new open-source artificial intelligence accelerated virtual screening platform for drug discovery. Using this platform, we screen multi-billion compound libraries against two unrelated targets, a ubiquitin ligase target KLHDC2 and the human voltage-gated sodium channel NaV1.7. For both targets, we discover hit compounds, including seven hits (14% hit rate) to KLHDC2 and four hits (44% hit rate) to NaV1.7, all with single digit micromolar binding affinities. Screening in both cases is completed in less than seven days. Finally, a high resolution X-ray crystallographic structure validates the predicted docking pose for the KLHDC2 ligand complex, demonstrating the effectiveness of our method in lead discovery.

Culture-based microbial natural product discovery strategies fail to realize the extraordinary biosynthetic potential detected across earth's microbiomes. Here we introduce Small Molecule In situ Resin Capture (SMIRC), a culture-independent method to obtain natural products directly from the environments in which they are produced. We use SMIRC to capture numerous compounds including two new carbon skeletons that were characterized using NMR and contain structural features that are, to the best of our knowledge, unprecedented among natural products. Applications across diverse marine habitats reveal biome-specific metabolomic signatures and levels of chemical diversity in concordance with sequence-based predictions. Expanded deployments, in situ cultivation, and metagenomics facilitate compound discovery, enhance yields, and link compounds to candidate producing organisms, although microbial community complexity creates challenges for the later. This compound-first approach to natural product discovery provides access to poorly explored chemical space and has implications for drug discovery and the detection of chemically mediated biotic interactions.

Polypharmacology drugs-compounds that inhibit multiple proteins-have many applications but are difficult to design. To address this challenge we have developed POLYGON, an approach to polypharmacology based on generative reinforcement learning. POLYGON embeds chemical space and iteratively samples it to generate new molecular structures; these are rewarded by the predicted ability to inhibit each of two protein targets and by drug-likeness and ease-of-synthesis. In binding data for >100,000 compounds, POLYGON correctly recognizes polypharmacology interactions with 82.5% accuracy. We subsequently generate de-novo compounds targeting ten pairs of proteins with documented co-dependency. Docking analysis indicates that top structures bind their two targets with low free energies and similar 3D orientations to canonical single-protein inhibitors. We synthesize 32 compounds targeting MEK1 and mTOR, with most yielding >50% reduction in each protein activity and in cell viability when dosed at 1-10 μM. These results support the potential of generative modeling for polypharmacology.

In recent years, advances in biomedicine have revealed an important role for post-transcriptional mechanisms of gene expression regulation in pathologic conditions. In cancer in general and leukaemia specifically, RNA binding proteins have emerged as important regulator of RNA homoeostasis that are often dysregulated in the disease state. Having established the importance of these pathogenetic mechanisms, there have been a number of efforts to target RNA binding proteins using oligonucleotide-based strategies, as well as with small organic molecules. The field is at an exciting inflection point with the convergence of biomedical knowledge, small molecule screening strategies and improved chemical methods for synthesis and construction of sophisticated small molecules. Here, we review the mechanisms of post-transcriptional gene regulation, specifically in leukaemia, current small-molecule based efforts to target RNA binding proteins, and future prospects.

RNA methylation is a metabolic process validated for its association with various diseases, and thus, RNA methyltransferases (MTases) have become increasingly important in drug discovery. Yet, most frequently utilized RNA MTase assays are limited in their throughput and hamper this rapidly evolving field of medicinal chemistry. In this study, we describe a modular nanomole scale building block system that allowed the identification of tailored fluorescent MTase probes to unlock a broad selection of MTase drug targets for fluorescence‐based binding assays. Probe candidates were initially prepared on a 4 nanomole scale and could be tested directly from crude reaction mixtures to allow rapid probe identification and optimization. Using an alkyne‐azide click late‐stage functionalization strategy and in silico protein databank mining, we established a selection of fluorescent probes suitable for relevant drug targets from the METTL and NSUN families, as well as bacterial and viral MTases. Using this concept, a high‐throughput screening on the unexplored drug target METTL1 discovered three hit compounds with micromolar potency providing a (1H‐pyrazol‐4‐yl)pyridine‐based starting point for METTL1 drug discovery. [ABSTRACT FROM AUTHOR]

Comprehensive Summary Terpenoids are a large and diverse family of secondary metabolites that have gained more attention from chemists and pharmacologists for their intriguing skeletons and broad biological activities, including anti‐inflammatory, antitumoral, antimicrobial, and anti‐viral effects. Based on the previous reviews, this review comprehensively summarized recent advances on bioactive terpenoids, focusing on their intriguing structural features and promising biological activities, covering literature from 2020 to 2023. In this review, 532 terpenoids with remarkable chemical architectures and biological properties are classified into sesquiterpenoids (142), diterpenoids (129), sesterterpenoids (24), triterpenoids (52), and meroterpenoids (185), that were isolated from terrestrial and marine resources. This review would provide a reference for researchers to rapidly discover the interesting terpenoids for further study as promising drug leads. [ABSTRACT FROM AUTHOR]

Background: Identification of drug–target interactions is an indispensable part of drug discovery. While conventional shallow machine learning and recent deep learning methods based on chemogenomic properties of drugs and target proteins have pushed this prediction performance improvement to a new level, these methods are still difficult to adapt to novel structures. Alternatively, large-scale biological and pharmacological data provide new ways to accelerate drug–target interaction prediction. Methods: Here, we propose DrugMAN, a deep learning model for predicting drug–target interaction by integrating multiplex heterogeneous functional networks with a mutual attention network (MAN). DrugMAN uses a graph attention network-based integration algorithm to learn network-specific low-dimensional features for drugs and target proteins by integrating four drug networks and seven gene/protein networks collected by a certain screening conditions, respectively. DrugMAN then captures interaction information between drug and target representations by a mutual attention network to improve drug–target prediction. Results: DrugMAN achieved the best performance compared with cheminformation-based methods SVM, RF, DeepPurpose and network-based deep learing methods DTINet and NeoDT in four different scenarios, especially in real-world scenarios. Compared with SVM, RF, deepurpose, DTINet, and NeoDT, DrugMAN showed the smallest decrease in AUROC, AUPRC, and F1-Score from warm-start to Both-cold scenarios. This result is attributed to DrugMAN's learning from heterogeneous data and indicates that DrugMAN has a good generalization ability. Taking together, DrugMAN spotlights heterogeneous information to mine drug–target interactions and can be a powerful tool for drug discovery and drug repurposing. [ABSTRACT FROM AUTHOR]

Bicyclic boronates have recently emerged as promising candidates to invoke targeted biomolecular interactions, given their selectivity for specific functionalities. Despite this, the general stability of the C–B bond in vivo, for such heterocycles, remains an intractable challenge that can often preclude their utility in drug discovery. To address this challenge, de novo strategies that allow expedient access to strategically substituted boronates, that enable modulation of the C–B bond are urgently required. Herein we disclose an operationally simple, regioselective cross-coupling/cyclisation reaction of easily accessible vicinal boronic esters with 2-halophenols to rapidly forge 3-substituted bicyclic boronates. The utility of the platform was demonstrated via expedient access to Xeruborbactam derivatives, chemoselective manipulation of formed products and the convergent approach to bicyclic boronates with a pendent biomolecular probe. [ABSTRACT FROM AUTHOR]

α-Glucosidase (α-Glu) is implicated in the progression and pathogenesis of type II diabetes (T2D). In this study, we developed a rapid colorimetric technique using platinum nanoparticles stabilized by chitosan (Ch-PtNPs) to detect α-Glu activity and its inhibitor. The Ch-PtNPs facilitate the conversion of 3,3′,5,5′-tetramethylbenzidine (TMB) into oxidized TMB (oxTMB) in the presence of dissolved O2. The catalytic hydrolysis of 2-O-α-d-glucopyranosyl-l-ascorbic acid (AA-2G) by α-Glu produces ascorbic acid (AA), which reduces oxTMB to TMB, leading to the fading of the blue color. However, the presence of α-Glu inhibitors (AGIs) hinders the generation of AA, allowing Ch-PtNPs to re-oxidize colorless TMB back to blue oxTMB. This unique phenomenon enables the colorimetric detection of α-Glu activity and AGIs. The linear range for α-Glu was found to be 0.1–1.0 U mL−1 and the detection limit was 0.026 U mL−1. Additionally, the half-maximal inhibition value (IC50) for acarbose, an α-Glu inhibitor, was calculated to be 0.4769 mM. Excitingly, this sensing platform successfully detected α-Glu activity in human serum samples and effectively screened AGIs. These promising findings highlight the potential application of the proposed strategy in clinical diabetes diagnosis and drug discovery. [ABSTRACT FROM AUTHOR]

Background: Drug discovery strongly relies on the thorough evaluation of preclinical experimental studies. In the context of pulmonary fibrosis, micro-computed tomography (µCT) and histology are well-established and complementary tools for assessing, in animal models, disease progression and response to treatment. µCT offers dynamic, real-time insights into disease evolution and the effects of therapies, while histology provides a detailed microscopic examination of lung tissue. Here, we present a semi-automatic pipeline that integrates these readouts by matching individual µCT volume slices with the corresponding histological sections, effectively linking densitometric data with Ashcroft score measurements. Methods: The tool first geometrically aligns the vertical axis of the µCT volume with the cutting plane used to prepare the histological sample. Then, focusing on the left lung, it computes the affine registration that identifies the µCT coronal slice that best matches the histological section. Finally, quantitative µCT imaging parameters are extracted from the selected slice. In a proof-of-concept test, the tool was applied to a bleomycin-induced mouse model of lung fibrosis. Results: The proposed approach demonstrated high accuracy and time effectiveness in matching µCT and histological sections minimizing manual intervention, with an overall success rate of 95%, and reduced time required to align µCT and histological data from 40 to 5 min. Significant correlations were found between quantitative data derived from µCT and histology data. Conclusions: The precise combination of microscopic ex-vivo information with 3D in-vivo data enhances the accuracy and representativeness of tissue analysis and provides a structural context for omic studies, serving as the foundation for a multi-layer platform. By facilitating a detailed and objective view of disease progression and treatment response, this approach has the potential to accelerate the development of effective therapies for lung fibrosis. [ABSTRACT FROM AUTHOR]

In the race to combat ever-evolving diseases, the drug discovery process often faces the hurdles of high-cost and time-consuming procedures. To tackle these challenges and enhance the efficiency of identifying new therapeutic agents, we introduce VirtuDockDL, which is a streamlined Python-based web platform utilizing deep learning for drug discovery. This pipeline employs a Graph Neural Network to analyze and predict the effectiveness of various compounds as potential drug candidates. During the validation phase, VirtuDockDL was instrumental in identifying non-covalent inhibitors against the VP35 protein of the Marburg virus, a critical target given the virus's high fatality rate and limited treatment options. Further, in benchmarking, VirtuDockDL achieved 99% accuracy, an F1 score of 0.992, and an AUC of 0.99 on the HER2 dataset, surpassing DeepChem (89% accuracy) and AutoDock Vina (82% accuracy). Compared to RosettaVS, MzDOCK, and PyRMD, VirtuDockDL outperformed them by combining both ligand- and structure-based screening with deep learning. While RosettaVS excels in accurate docking but lacks high-throughput screening, and PyRMD focuses on ligand-based methods without AI integration, VirtuDockDL offers superior predictive accuracy and full automation for large-scale datasets, making it ideal for comprehensive drug discovery workflows. These results underscore the tool's capability to identify high-affinity inhibitors accurately across various targets, including the HER2 protein for cancer therapy, TEM-1 beta-lactamase for bacterial infections, and the CYP51 enzyme for fungal infections like Candidiasis. To sum up, VirtuDockDL combines user-friendly interface design with powerful computational capabilities to facilitate rapid, cost-effective drug discovery and development. The integration of AI in drug discovery could potentially transform the landscape of pharmaceutical research, providing faster responses to global health challenges. The VirtuDockDL is available at https://github.com/FatimaNoor74/VirtuDockDL. [ABSTRACT FROM AUTHOR]

Background and aims Design and setting Participants and Measurements Findings Conclusions Amphetamine‐type stimulants are the second‐most used illicit drugs globally, yet there are no US Food and Drug Administration (FDA)‐approved treatments for amphetamine‐type stimulant use disorders (ATSUD). The aim of this study was to utilize a drug discovery framework that integrates artificial intelligence (AI)‐based drug prediction, clinical corroboration and mechanism of action analysis to identify FDA‐approved drugs that can be repurposed for treating ATSUD.An AI‐based knowledge graph model was first utilized to prioritize FDA‐approved drugs in their potential efficacy for treating ATSUD. Among the top 10 ranked candidate drugs, ketamine represented a novel candidate with few studies examining its effects on ATSUD. We therefore conducted a retrospective cohort study to assess the association between ketamine and ATSUD remission using US electronic health record (EHR) data. Finally, we analyzed the potential mechanisms of action of ketamine in the context of ATSUD.ATSUD patients who received anesthesia (n = 3663) or were diagnosed with depression (n = 4328) between January 2019 and June 2022. The outcome measure was the diagnosis of ATSUD remission within one year of the drug prescription.Ketamine for anesthesia in ATSUD patients was associated with greater ATSUD remission compared with other anesthetics: hazard ratio (HR) = 1.58, 95% confidence interval (CI) = 1.15–2.17. Similar results were found for ATSUD patients with depression when comparing ketamine with antidepressants and bupropion/mirtazapine with HRs of 1.51 (95% CI = 1.14–2.01) and 1.68 (95% CI = 1.18–2.38), respectively. Functional analyses demonstrated that ketamine targets several ATSUD‐associated pathways including neuroactive ligand‐receptor interaction and amphetamine addiction.There appears to be an association between clinician‐prescribed ketamine and higher remission rates in patients with amphetamine‐type stimulant use disorders. [ABSTRACT FROM AUTHOR]

RNA is a central molecule in life, involved in a plethora of biological processes and playing a key role in many diseases. Targeting RNA emerges as a significant endeavor in drug discovery, diverging from conventional protein‐centric approaches to tackle various pathologies. Whilst identifying small molecules that bind to specific RNA regions is the first step, the abundance of non‐functional RNA segments renders many interactions biologically inert. Consequently, small molecule binding does not necessarily meet stringent criteria for clinical translation, calling for solutions to push the field forward. Converting RNA‐binders into RNA‐degraders presents a promising avenue to enhance RNA‐targeting. This mini‐review outlines strategies and exemplars wherein simple small molecule RNA binders are reprogrammed into active degraders through the linkage of functional groups. These approaches encompass mechanisms that induce degradation via endogenous enzymes, termed RIBOTACs, as well as those with functional moieties acting autonomously to degrade RNA. Through this exploration, we aim to offer insights into advancing RNA‐targeted therapeutic strategies. [ABSTRACT FROM AUTHOR]

For over seven decades, dopamine receptor 2 (D2 receptor) antagonists remained the mainstay treatment for neuropsychiatric disorders. Although it is effective for treating hyperdopaminergic symptoms, it is often ineffective for treating negative and cognitive deficits. Trace amine-associated receptor 1 (TAAR1) is a novel, pharmacological target in the treatment of schizophrenia and other neuropsychiatric conditions. Several TAAR1 agonists are currently being developed and are in various stages of clinical and preclinical development. Previous efforts to identify TAAR1 agonists have been hampered by challenges in pharmacological characterisation, the absence of experimentally determined structures, and species-specific preferences in ligand binding and recognition. Further, poor insights into the functional selectivity of the receptor led to the characterisation of ligands with analogous signalling mechanisms. Such approaches limited the understanding of divergent receptor signalling and their potential clinical utility. Recent cryogenic electron microscopic (cryo-EM) structures of human and mouse TAAR1 (hTAAR1 and mTAAR1, respectively) in complex with agonists and G proteins have revealed detailed atomic insights into the binding pockets, binding interactions and binding modes of several agonists including endogenous trace amines (β-phenylethylamine, 3-Iodothyronamine), psychostimulants (amphetamine, methamphetamine), clinical compounds (ulotaront, ralmitaront) and repurposed drugs (fenoldopam). The in vitro screening of drug libraries has also led to the discovery of novel TAAR1 agonists (asenapine, guanabenz, guanfacine) which can be used in clinical trials or further developed to treat different neuropsychiatric conditions. Furthermore, an understanding of unappreciated signalling mechanisms (Gq, Gs/Gq) by TAAR1 agonists has come to light with the discovery of selective compounds to treat schizophrenia-like phenotypes. In this review, we discuss the emergence of structure-based approaches in the discovery of novel TAAR1 agonists through drug repurposing strategies and structure-guided designs. Additionally, we discuss the functional selectivity of TAAR1 signalling, which provides important clues for developing disorder-specific compounds. [ABSTRACT FROM AUTHOR]

Background: Target identification and hit identification can be transformed through the application of biomedical knowledge analysis, AI-driven virtual screening and robotic cloud lab systems. However there are few prospective studies that evaluate the efficacy of such integrated approaches. Results: We synergistically integrate our in-house-developed target evaluation (SpectraView) and deep-learning-driven virtual screening (HydraScreen) tools with an automated robotic cloud lab designed explicitly for ultra-high-throughput screening, enabling us to validate these platforms experimentally. By employing our target evaluation tool to select IRAK1 as the focal point of our investigation, we prospectively validate our structure-based deep learning model. We can identify 23.8% of all IRAK1 hits within the top 1% of ranked compounds. The model outperforms traditional virtual screening techniques and offers advanced features such as ligand pose confidence scoring. Simultaneously, we identify three potent (nanomolar) scaffolds from our compound library, 2 of which represent novel candidates for IRAK1 and hold promise for future development. Conclusion: This study provides compelling evidence for SpectraView and HydraScreen to provide a significant acceleration in the processes of target identification and hit discovery. By leveraging Ro5's HydraScreen and Strateos' automated labs in hit identification for IRAK1, we show how AI-driven virtual screening with HydraScreen could offer high hit discovery rates and reduce experimental costs. Scientific contribution: We present an innovative platform that leverages Knowledge graph-based biomedical data analytics and AI-driven virtual screening integrated with robotic cloud labs. Through an unbiased, prospective evaluation we show the reliability and robustness of HydraScreen in virtual and high-throughput screening for hit identification in IRAK1. Our platforms and innovative tools can expedite the early stages of drug discovery. [ABSTRACT FROM AUTHOR]

Providencia stuartii, the most prevalent strain in its genus, is a significant cause of a range of potentially fatal diseases involving urinary tract infections, lung diseases, and endocarditis, often exhibiting resistance to multiple antibiotics, necessitating alternative therapeutic approaches. Unlike previous efforts focused on known virulence factors, this study delves into the physicochemical and functional characterization of 72 essential hypothetical proteins (EsHPs) from the P. stuartii BE246 genome. Subcellular localization identified 47 cytoplasmic EsHPs as potential drug targets, with the remainder 25 identified as vaccine targets. Subtractive analysis uncovered 68 virulent proteins, 4 of which were nonhomologous to both host and gut microbiota proteomes. Among these, WP_145928235.1 (147 amino acids), a cytoplasmic EsHP, emerged as a novel drug target due to its homologous broad‐spectrum hit. WP_145928235.1 displayed an instability index (II) of 33.14, an aliphatic index of 92.79, and the GRAVY score was ‐0.065 (hydrophilic). Molecular dynamics simulation‐aided conformational study and secondary structure analysis highlighted the protein's flexibility and propensity to interact with inhibitors. Key residues in the 60–90 region, including Asn62, Gly75, Glu81, and Asp85, were critical for interactions and stability, with low frustration indicating residual conservancy. WP_145928235.1 presents a promising target for therapeutic intervention against multidrug‐resistant P. stuartii infections. [ABSTRACT FROM AUTHOR]

Anaplastic lymphoma kinase (ALK; also known as ALK tyrosine kinase receptor) inhibitors (ALKi) are effective in treating lung cancer patients with chromosomal rearrangement of ALK. However, continuous treatment with ALKis invariably leads to acquired resistance in cancer cells. In this study, we propose an efficient strategy to suppress ALKi resistance through a meta‐analysis of transcriptome data from various cell models of acquired resistance to ALKis. We systematically identified gene signatures that consistently showed altered expression during the development of resistance and conducted computational drug screening using these signatures. We identified emetine as a promising candidate compound to inhibit the growth of ALKi‐resistant cells. We demonstrated that emetine exhibited effectiveness in inhibiting the growth of ALKi‐resistant cells, and further interpreted its impact on the resistant signatures through drug‐induced RNA‐sequencing data. Our transcriptome‐guided systematic approach paves the way for efficient drug discovery to overcome acquired resistance to cancer therapy. [ABSTRACT FROM AUTHOR]

Targeted protein degradation strategies leverage endogenous cellular degradation machinery to selectively eliminate a protein of interest. Emerging technologies are opening avenues in drug discovery and functional characterization of intracellular, membrane, and extracellular proteins. To view this SnapShot, open or download the PDF. [ABSTRACT FROM AUTHOR]

Coupling reactions play a crucial role in drug development enabling the rapid expansion of structure–activity relationships (SARs) during drug discovery programs to identify a clinical candidate and simplify subsequent drug development processes. In particular, their relevance in clinical medicine and drug discovery has increased significantly in the last two decades. To facilitate these metal-catalyzed coupling reactions, suitably designed ligands are necessary and from the industrial point of view, sustainable and cost-effective ligands are of current need. Benzotriazole, a non-toxic, thermally stable, and inexpensive bidentate ligand, exhibits strong electron donating and electron accepting properties, along with excellent solubility in various organic solvents. It has been extensively explored as a synthetic auxiliary in the past; in recent years, benzotriazole and its derivatives have been used as ligands in metal-catalyzed coupling reactions. The facile construction of carbon–carbon and carbon–heteroatom bonds in the presence of versatile benzotriazole ligands makes it an indispensable ligand for catalytic transformations. The present feature article mainly emphasizes the advances in the utilization of benzotriazole as a ligand in a diverse range of C–C, C–N, C–O, and C–S coupling reactions. [ABSTRACT FROM AUTHOR]

Target-first drug discovery relies heavily on protein structure information, which severely limits its application. In recent years, fragment-based drug Design (FBDD) has been identified as an alternative solution, where screening of smaller molecules for lower affinity allowed the use of focused libraries with a higher hit rate. It is shown that coupling an sp2-rich heteroaromatic group with a monofunctional sp3-rich core gives fragments (186 examples) with advantageous physical-chemical properties, covering a chemical space often neglected in traditional libraries. [ABSTRACT FROM AUTHOR]

Substance use disorder (SUD) is a chronic condition, with maintained abuse of a substance leading to physiological and psychological alterations and often changes in cognitive and social behaviours. Current therapies include psychotherapy coupled with medication; however, high relapse rates reveal the shortcomings of these therapies. The signalling, expression profile, and neurological function of the serotonin 2C receptor (5‐HT2C receptor) make it a candidate of interest for the treatment of SUD. Recently, psychedelics, which broadly act at 5‐HT2 receptors, have indicated potential for the treatment of SUD, implicating the 5‐HT2C receptor. The modern psychedelic movement has rekindled interest in the 5‐HT2C receptor, resulting in many new studies, especially structural analyses. This review explores the structural, molecular and cellular mechanisms governing 5‐HT2C receptor function in the context of SUD. This provides the basis of the preclinical and clinical evidence for their role in SUD and highlights the potential for future exploration. [ABSTRACT FROM AUTHOR]

Fragment-Based Drug Design (FBDD) plays a pivotal role in the field of drug discovery and development. The construction of high-quality fragment libraries is a critical step in FBDD. Conventional fragmentation approaches often rely on rigid rules and chemical intuition, limiting their adaptability to diverse molecular structures. The rapid development of Artificial Intelligence (AI) technology offers a transformative opportunity to rethink traditional methods. Here, we present DigFrag, a digital fragmentation method that highlights important substructures by focusing locally within the molecular graph. In addition, we feed the fragments segmented by machine intelligence and human expertise into the deep generative model to compare the preference for data from different sources. Experimental results show that the structural diversity of fragments segmented by DigFrag is higher, and more desirable compounds are generated based on these fragments. These results also demonstrate that data generated based on AI methods may be more suitable for AI models. Moreover, a user-friendly platform called MolFrag (https://dpai.ccnu.edu.cn/MolFrag/) is developed based on various fragmentation techniques to support molecular segmentation. Fragment-based drug design plays a pivotal role in the field of drug discovery and development, however, the construction of high-quality fragment libraries is a critical but challenging step. Here, the authors develop DigFrag, a digital fragmentation method based on the graph attention mechanism, showing higher structural diversity of the fragments and higher applicability to artificial intelligence-based drug design. [ABSTRACT FROM AUTHOR]

The accurate identification of protein–ligand binding sites is of critical importance in understanding and modulating protein function. Accordingly, ligand binding site prediction has remained a research focus for over three decades with over 50 methods developed and a change of paradigm from geometry-based to machine learning. In this work, we collate 13 ligand binding site predictors, spanning 30 years, focusing on the latest machine learning-based methods such as VN-EGNN, IF-SitePred, GrASP, PUResNet, and DeepPocket and compare them to the established P2Rank, PRANK and fpocket and earlier methods like PocketFinder, Ligsite and Surfnet. We benchmark the methods against the human subset of our new curated reference dataset, LIGYSIS. LIGYSIS is a comprehensive protein–ligand complex dataset comprising 30,000 proteins with bound ligands which aggregates biologically relevant unique protein–ligand interfaces across biological units of multiple structures from the same protein. LIGYSIS is an improvement for testing methods over earlier datasets like sc-PDB, PDBbind, binding MOAD, COACH420 and HOLO4K which either include 1:1 protein–ligand complexes or consider asymmetric units. Re-scoring of fpocket predictions by PRANK and DeepPocket display the highest recall (60%) whilst IF-SitePred presents the lowest recall (39%). We demonstrate the detrimental effect that redundant prediction of binding sites has on performance as well as the beneficial impact of stronger pocket scoring schemes, with improvements up to 14% in recall (IF-SitePred) and 30% in precision (Surfnet). Finally, we propose top-N+2 recall as the universal benchmark metric for ligand binding site prediction and urge authors to share not only the source code of their methods, but also of their benchmark. Scientific contributions This study conducts the largest benchmark of ligand binding site prediction methods to date, comparing 13 original methods and 15 variants using 10 informative metrics. The LIGYSIS dataset is introduced, which aggregates biologically relevant protein–ligand interfaces across multiple structures of the same protein. The study highlights the detrimental effect of redundant binding site prediction and demonstrates significant improvement in recall and precision through stronger scoring schemes. Finally, top-N+2 recall is proposed as a universal benchmark metric for ligand binding site prediction, with a recommendation for open-source sharing of both methods and benchmarks. [ABSTRACT FROM AUTHOR]

Potassium ion channels play a fundamental role in regulating cell membrane repolarization, modulating the frequency and shape of action potentials, and maintaining the resting membrane potential. A growing number of studies have indicated that dysfunction in potassium channels associates with the pathogenesis and treatment of depression. However, the involvement of potassium channels in the onset and treatment of depression has not been thoroughly summarized. In this review, we performed a comprehensive analysis of the association between multiple potassium channels and their roles in depression, and compiles the SNP loci of potassium channels associated with depression, as well as antidepressant drugs that target these channels. We discussed the pivotal role of potassium channels in the treatment of depression, provide valuable insights into new therapeutic targets for antidepressant treatment and critical clues to future drug discovery. [ABSTRACT FROM AUTHOR]

Background: Bioactive peptides are important bioactive molecules composed of short-chain amino acids that play various crucial roles in the body, such as regulating physiological processes and promoting immune responses and antibacterial effects. Due to their significance, bioactive peptides have broad application potential in drug development, food science, and biotechnology. Among them, understanding their biological mechanisms will contribute to new ideas for drug discovery and disease treatment. Results: This study employs generative adversarial capsule networks (CapsuleGAN), gated recurrent units (GRU), and convolutional neural networks (CNN) as base classifiers to achieve ensemble learning through voting methods, which not only obtains high-precision prediction results on the angiotensin-converting enzyme (ACE) inhibitory peptides dataset and the anticancer peptides (ACP) dataset but also demonstrates effective model performance. For this method, we first utilized the protein language model—evolutionary scale modeling (ESM-2)—to extract relevant features for the ACE inhibitory peptides and ACP datasets. Following feature extraction, we trained three deep learning models—CapsuleGAN, GRU, and CNN—while continuously adjusting the model parameters throughout the training process. Finally, during the voting stage, different weights were assigned to the models based on their prediction accuracy, allowing full utilization of the model's performance. Experimental results show that on the ACE inhibitory peptide dataset, the balanced accuracy is 0.926, the Matthews correlation coefficient (MCC) is 0.831, and the area under the curve is 0.966; on the ACP dataset, the accuracy (ACC) is 0.779, and the MCC is 0.558. The experimental results on both datasets are superior to existing methods, demonstrating the effectiveness of the experimental approach. Conclusion: In this study, CapsuleGAN, GRU, and CNN were successfully employed as base classifiers to implement ensemble learning, which not only achieved good results in the prediction of two datasets but also surpassed existing methods. The ability to predict peptides with strong ACE inhibitory activity and ACPs more accurately and quickly is significant, and this work provides valuable insights for predicting other functional peptides. The source code and dataset for this experiment are publicly available at https://github.com/Zhou-Jianren/bioactive-peptides. [ABSTRACT FROM AUTHOR]

1,2,3-triazoles have emerged as important structural motifs in chemical biology, attracting increasing focus in recent years because of their wide range of uses. These compounds can be efficiently synthesized using click reactions. Its versatility makes it valuable in drug discovery and materials science. Significant advancements have been made in recent years in the process of making 1,2,3-triazoles, reflecting the growing curiosity and investment in the area of research. The catalytic performance of transition metals Cu, Ir, Rh, Ru, Ni, Pd, Au, Ag, and Zn, which were used as ligands and salts in the azide–alkyne cycloaddition method of 1,2,3-triazole synthesis, has been covered in this review. Cu-complexes and salts were found to be more effective for selective synthesis among all transition metals. Furthermore, it is determined that some azide–alkyne reactions are entirely catalytic in nature and cannot be carried out as such by switching transition metals. [ABSTRACT FROM AUTHOR]

Image-based cell profiling is a powerful tool that compares perturbed cell populations by measuring thousands of single-cell features and summarizing them into profiles. Typically a sample is represented by averaging across cells, but this fails to capture the heterogeneity within cell populations. We introduce CytoSummaryNet: a Deep Sets-based approach that improves mechanism of action prediction by 30–68% in mean average precision compared to average profiling on a public dataset. CytoSummaryNet uses self-supervised contrastive learning in a multiple-instance learning framework, providing an easier-to-apply method for aggregating single-cell feature data than previously published strategies. Interpretability analysis suggests that the model achieves this improvement by downweighting small mitotic cells or those with debris and prioritizing large uncrowded cells. The approach requires only perturbation labels for training, which are readily available in all cell profiling datasets. CytoSummaryNet offers a straightforward post-processing step for single-cell profiles that can significantly boost retrieval performance on image-based profiling datasets. Author summary: Image-based cell profiling experiments measure thousands of features in millions of cells in microscopy images. Measuring cell response to various treatments in this way has proven powerful for many applications in biological research and drug discovery. However, each cell population responding to a treatment is usually population-averaged, losing information about the diversity of cells within each sample. We developed CytoSummaryNet, a machine learning method that substantially improves the ability to predict a treatment's mechanism of action from cell imaging profiles. Rather than averaging all cells equally, it learns which cells are most informative; it appears to focus on large, uncluttered cells while downweighting dividing cells and debris. Training CytoSummaryNet only requires information about which samples were exposed to which treatments, making it simpler to implement than previous techniques. By learning biologically relevant ways to summarize single-cell data into treatment profiles, CytoSummaryNet could help scientists leverage cell heterogeneity to extract richer insights from cell imaging screens and ultimately accelerate the development of new therapies. [ABSTRACT FROM AUTHOR]

Objective Methods Results Significance The cannabinoid cannabidiol has established antiseizure effects in drug‐resistant epilepsies such as Dravet syndrome and Lennox–Gastaut syndrome. Amorfrutin 2, honokiol, and magnolol are structurally similar to cannabinoids (cannabis‐like drugs) but derive from non‐cannabis plants. We aimed to study the antiseizure potential of these compounds in various mouse seizure models. In addition, we aimed to characterize their molecular pharmacology at cannabinoid CB1 and CB2 receptors and at T‐type calcium channels, which are known targets of the cannabinoids.Brain and plasma pharmacokinetic profiles were determined. Antiseizure activity was assessed against hyperthermia‐induced seizures in a Scn1a+/− mouse model of Dravet syndrome. We then elaborated on the most promising compounds in the maximal electroshock (MES) test in mice and the Gabrb3+/D120N mouse model of Lennox–Gastaut syndrome. Fluorescence‐based assays were used to examine modulatory activity at cannabinoid CB1 and CB2 receptors and T‐type calcium channel subtypes CaV3.1, CaV3.2, and CaV3.3 overexpressed in mammalian cells. Automated patch‐clamp electrophysiology was then used to confirm inhibitory activity on CaV3.1, CaV3.2, and CaV3.3 channels.Magnolol and honokiol had high brain‐to‐plasma ratios (3.55 and 7.56, respectively), unlike amorfrutin 2 (0.06). Amorfrutin 2 and magnolol but not honokiol significantly increased the body temperature threshold at which Scn1a+/− mice had a generalized tonic–clonic seizure. Both amorfrutin 2 and magnolol significantly decreased the proportion of mice exhibiting hindlimb extension in the MES test. Furthermore, magnolol reduced the number and duration of atypical absence seizures in Gabrb3+/D120N mice. The three compounds inhibited all T‐type calcium channel subtypes but were without specific activity at cannabinoid receptors.We show for the first time that amorfrutin 2 and magnolol display novel antiseizure activity in mouse drug‐resistant epilepsy models. Our results justify future drug discovery campaigns around these structural scaffolds that aim to develop novel antiseizure drugs for intractable epilepsies. [ABSTRACT FROM AUTHOR]