Search Results (840)

The combination of computational and experimental methods has become indispensable for optimization and rational enzyme design. Recently, the development of artificial intelligence (AI)-based tools has further streamlined enzyme engineering pipelines, enabling more accurate designs, while reducing the number of variants required for experimental validation. However, due to the intricate complexity of enzymatic systems, significant challenges must be addressed before we take the next step to fully optimize the use of these AI-guided enzyme design methodologies. These challenges include un-curated datasets, the need to consider both the static and dynamic structure of enzymes, and the requirement for effective interdisciplinary collaborations to ensure the integration of computational and experimental approaches. Here, we present recent advances in AI-based computational enzyme design, discussing the main challenges in the field and how a combination with classical physics-based methods could help overcome them. We further explore novel trends that could completely modulate the future of protein design and provide our outlook on the key concepts and future opportunities that will shape the next steps of enzyme design. Full article

Monoacylglycerol lipase (MAGL) is a key enzyme at the interface between the endocannabinoid system and lipid metabolism, playing a pivotal role in the hydrolysis of the endocannabinoid 2-arachidonoylglycerol and in the regulation of lipid mediators involved in inflammation, pain, neurodegeneration and cancer. Owing to its therapeutic relevance, MAGL has emerged as an attractive pharmacological target, stimulating extensive research efforts aimed at the development of potent and selective modulators of its activity. Advances in medicinal chemistry, together with the increasing application of innovative computational approaches and biochemical methods to assess MAGL activity, have significantly expanded the chemical space of compounds capable of modulating this enzyme. This review provides a comprehensive overview of selective MAGL modulators reported in the scientific literature from 2020 to the present, excluding compounds described exclusively in patent literature and MAGL probes, as this area has been recently reviewed elsewhere, ranging from classical enzyme inhibitors to modulators acting through alternative strategies, such as targeted protein degradation. Overall, this review highlights the structural diversity and the main strategies that have emerged in recent years in modulating MAGL and it aims to guide the rational design of next-generation MAGL-targeting agents. Full article

Natural coumarins are a structurally privileged group of bioactive benzopyranone lactones widely spread across the Apiaceae, Rutaceae, and Leguminosae families, and hold significant potential as inhibitors of pathological protein aggregation in Alzheimer’s disease, Parkinson’s disease, and type 2 diabetes mellitus. The fully planar, rigid bicyclic structure of the coumarin nucleus (~3.4–3.5 Å thickness) is geometrically compatible with intercalative π–π stacking with aggregation-nucleating aromatic residues, including Phe19 of Aβ(1–42), providing a mechanistically coherent pharmacophoric basis for anti-aggregation activity according to computational and indirect biophysical evidence. This review critically evaluates the peer-reviewed literature on naturally occurring coumarins and their synthetic derivatives as candidate β-sheet intercalators, with analysis of SAR at C-3 to C-8 positions; multi-target-directed ligand designs with dual activities of inhibiting AChE, BACE-1, GSK-3β, and MAO-B, and as blood–brain barrier-penetrating neuroprotective agents validated in cellular and rodent models. The critical analysis identifies the translational gap between in vitro IC₅₀ values and attainable brain drug concentrations as the primary pharmacological obstacle. It identifies the absence of systematic investigation of coumarin against IAPP, a directly relevant amyloid target in metabolic neurodegeneration, as the most significant unmet research priority in the field. Full article

Small open reading frames (sORFs) encode micropeptides, which are a promising yet largely untapped resource for creating peptide design templates. Owing to their concise nature and functional efficiency, micropeptides often rely on essential structural elements and brief linear motifs, such as domains for membrane interaction, targeting sequences, and sites for protein–protein interactions, to fulfill their biological functions. This inherent simplicity makes them particularly suitable for a bottom-up design approach aimed at identifying, extracting, and systematically refining functional motifs to develop novel bioactive peptides. This review addresses the critical question of how micropeptides, particularly those involved in tumor regulation, can be explored as emerging therapeutic targets, functional templates for peptide design, and potential future therapeutic agents, by synthesizing current understanding of their mechanisms, functional significance in cancer, and the computational and design strategies for their clinical translation. We examined the current methods for analyzing the sequence and structural characteristics that underpin their functional activity and investigated how these attributes can be leveraged for drug discovery and design. Finally, we underscore the primary challenges and future prospects in converting sORF-encoded micropeptides into clinically relevant molecules with the aim of broadening the current scope of the druggable proteome. Full article

The majority of disease-associated proteins are considered “undruggable” due to the absence of well-defined binding pockets, the presence of extended interaction surfaces, and intrinsic structural disorder, which collectively limit the effectiveness of conventional small molecules and biologics. Representative examples include KRAS, p53, and c-MYC. Peptide therapeutics, particularly macrocyclic peptides, occupy a unique chemical space capable of targeting such recalcitrant protein–protein interactions (PPIs) where small molecules often fail. However, traditional peptide discovery, which relies heavily on high-throughput screening, is labor-intensive and frequently yields candidates with suboptimal pharmacological properties. The integration of artificial intelligence has begun to transform peptide discovery from a largely empirical process into a rational and design-driven paradigm. Modern deep learning approaches, including diffusion-based generative models, enable the de novo design of peptide binders with high affinity and structural precision, even for disordered or previously intractable targets. In this perspective, we highlight key structural and biological challenges associated with undruggable proteins and consider how peptide-based modalities are beginning to overcome these longstanding barriers. We further explore how advances in artificial intelligence and computational modeling may reshape the rational design of next-generation peptide therapeutics and propose an integrated experimental–computational framework to facilitate the development of clinically actionable candidates. Full article

The association of small molecules with lipid membranes plays a central role in drug delivery, pharmacokinetics, toxicity, and membrane biophysics, also being of fundamental importance in drug pharmacodynamics given that most drug targets are membrane-associated proteins. Accurate determination of solute–membrane association affinities, however, remains challenging due to the diversity of experimental systems, the complexity of membrane environments, and the intrinsic limitations of individual methodologies. This review provides a comprehensive overview of the experimental and computational approaches currently used to quantify small molecule association with lipid membranes. Standard experimental techniques, including spectroscopy-based methods, calorimetry, electrophoretic measurements, and surface-sensitive approaches, are discussed alongside established computational strategies ranging from continuum models to atomistic molecular dynamics simulations. Particular emphasis is placed on the formalisms required for data analysis, including partitioning models and thermodynamic frameworks, as well as on the assumptions underlying each method. The validity limits, sources of uncertainty, and common experimental and interpretative pitfalls are critically examined. By providing a unified and comparative perspective, this work establishes a structured framework for the quantitative study of solute–membrane interactions, guiding new researchers in the selection of appropriate methodologies and in the rigorous analysis of experimental and computational results. Moreover, it enables the consistent and quantitative rationalization of affinity parameters reported across the literature, supporting the development of curated datasets and predictive relationships that can inform the design of new and more effective drugs. Full article

Background/Objectives: Proteolysis Targeting Chimeras (PROTACs) are heterobifunctional small molecules that induce ubiquitin–proteasome–mediated degradation of target proteins and have matured from proof-of-concept chemistry to a clinically validated therapeutic modality, with the first Phase 3 readout reported in 2025. A systematic bibliometric analysis covering this pivotal-trial era, however, has been lacking. This study aimed to map the historical trajectory, current research front, and emerging frontiers of PROTAC research. Methods: We analyzed 2630 PROTAC-related publications indexed in the Web of Science Core Collection (WoSCC) from 2001 to 2025 using a combined toolkit of CiteSpace, HistCite, the Alluvial Generator, and R (ggplot2), covering co-occurrence networks, burst detection, keyword clustering, citation historiography, alluvial flow analysis, and reference co-citation timeline visualization. Results: China and the USA led global output, and the Chinese Academy of Sciences, China Pharmaceutical University, and Harvard University were the most productive institutions; the Journal of Medicinal Chemistry was the leading publishing venue, and Alessio Ciulli, Jian Jin, and Craig M. Crews anchored the author network. Keyword burst analysis showed that early research centred on E3 ubiquitin ligase recruitment and small-molecule PROTAC design, whereas the current hotspots, resolved through keyword clustering and co-citation timelines, included structural basis and ternary complex design, EGFR-directed degradation, oral bioavailability optimization, applications in multiple myeloma and Alzheimer’s disease, tumour-targeted delivery, and computational/AI-driven design. Conclusions: This study extends the bibliometric record of PROTACs across 2001–2025 and identifies oral bioavailability, E3 ligase repertoire expansion, and CNS-penetrant degrader design as the emerging frontiers likely to shape the next phase of the field. Full article

Antimicrobial resistance and persistent biofilm-associated infections have renewed interest in bacteriophages as alternatives or complements to conventional antibiotics. However, broader therapeutic adoption remains constrained by slow phage discovery, incomplete genome characterization, narrow host range, complex therapeutic matching, and manufacturing variability. Artificial intelligence (AI) offers computational approaches that may help address several of these limitations. This comprehensive narrative review discusses current AI applications across the bacteriophage pipeline, including metagenomic phage discovery, genome annotation, phage–host interaction prediction, personalized phage selection, cocktail optimization, and phage–antibiotic combination design. The review also examines AI-assisted synthetic biology approaches, including receptor-binding protein redesign, CRISPR-enabled engineering, generative genome design, and biosafety screening, as well as emerging applications in bioprocess optimization, yield prediction, purification analytics, quality assurance, and supply-chain management. Current evidence suggests that AI may accelerate phage identification, improve host-range prediction, support therapeutic optimization, and strengthen manufacturing consistency, potentially facilitating the transition of phage therapy from individualized rescue interventions toward more scalable antimicrobial platforms. Nevertheless, major limitations remain, including fragmented, taxonomically biased datasets; limited external validation; restricted interpretability; privacy concerns; biosafety oversight; and evolving regulatory frameworks. Future progress will depend on standardized datasets, multimodal validation, scalable manufacturing systems, experimental and clinical verification, and coordinated regulatory development. Full article

The efficacy of enzyme therapy is limited by their poor stability under physiological conditions. Thermostable enzymes, derived from extremophilic organisms or generated by advanced protein engineering, offer a revolutionary solution to this long-standing challenge. They are widely used in industrial biocatalysis. Their therapeutic applications are poorly investigated and spread across diverse disciplines. While most applications are in the preclinical stages, emerging evidence from animal models demonstrates proof-of-concept for thermostable antioxidant enzymes in cardiovascular, neurodegenerative, and inflammatory diseases. This review critically assesses the translational landscape, distinguishing between established therapeutic enzymes (e.g., asparaginase, PEGylated SOD) and emerging experimental candidates. This narrative review consolidates existing knowledge about thermostable enzyme engineering and their emerging functions as molecular therapies, particularly in oxidative stress-related diseases. This review synthesizes recent advances in structural biology, computational protein design, biomaterials engineering, and translational antioxidant strategies, highlighting how breaking down disciplinary barriers is accelerating the development of sustainable and self-regenerating antioxidant platforms. By integrating molecular precision with systems-level therapeutic design, engineered thermostable antioxidant enzymes exemplify the future of biological development, where multidisciplinary collaboration drives innovation against oxidative stress-driven pathologies. Engineered thermostable enzymes provide a versatile basis for next-generation therapeutics, with the potential to address medical needs through improved stability, targeted activity, and multifunctional design. Full article

Diabetes mellitus remains a major metabolic disorder requiring the development of new and effective α-glucosidase inhibitors. The present study aimed to identify, design, and optimize novel 3-amino-2,4-diarylbenzo[4,5]imidazo[1,2-α]pyrimidine derivatives with promising inhibitory activity against the α-glucosidase enzyme using a comprehensive in silico strategy. Approximately 300 molecular descriptors were calculated to characterize a dataset of 32 compounds (Peytam et al.) and to investigate the structural factors governing their biological activity. Based on these descriptors, a multiple linear regression model was developed to predict the inhibitory activities of the compounds against alpha-glucosidase. The developed model demonstrated satisfactory predictive performance and was internally and externally validated to ensure its accuracy, robustness, and reproducibility. In addition, the applicability domain analysis confirmed the reliability of the predictions. Using the validated QSAR model, seven new derivatives were designed with predicted pIC₅₀ values exceeding the maximum activity of the parent compounds. The leverage analysis demonstrated that all newly designed compounds were located within the applicability domain of the model, supporting the reliability of the predictions. To further evaluate their inhibitory potential, molecular docking studies were performed to investigate the interactions between the designed compounds and the α-glucosidase active site. The docking results revealed favorable binding interactions comparable to those reported for known α-glucosidase inhibitors. Furthermore, ADMET analysis indicated generally favorable pharmacokinetic properties, although potential CYP3A4 inhibition-related pharmacokinetic risks were identified and discussed. Molecular dynamics simulations, including replicated runs and MM/GBSA binding free energy calculations, confirmed the stability of the most promising protein–ligand complexes throughout the simulation period. In conclusion, this study proposes a robust and integrated computational workflow combining descriptor generation, QSAR modeling, applicability domain analysis, molecular docking, ADMET prediction, and molecular dynamics simulations for the rational design of potential α-glucosidase inhibitors. The findings highlight the therapeutic potential of the designed derivatives and provide a valuable in silico framework for the future development of antidiabetic agents. Full article

Artificial intelligence (AI) is increasingly influencing nanopharmaceutical development by supporting the transition from empirical formulation screening toward predictive, data-driven, and translationally oriented design. Nanocarrier-based therapeutics are governed by nonlinear relationships among material composition, physicochemical attributes, manufacturing parameters, biological identity, pharmacokinetics, toxicity, and therapeutic performance. In this review, we examine how AI can contribute to nanopharmaceutical development from predictive formulation design to clinical translation. We synthesize current applications of machine learning, deep learning, physics-informed modeling, hybrid mechanistic–AI approaches, and automated optimization workflows, with emphasis on critical quality attribute modeling, multi-objective optimization, design of experiments, quality-by-design, process analytical technology, digital twins, and continuous manufacturing. We also discuss applications involving nano–bio interactions, pharmacokinetics, toxicity, immunogenicity, and precision nanomedicine. AI-based approaches can support rational nanocarrier design, identify nonlinear formulation–property relationships, guide optimization, improve process understanding, and integrate heterogeneous experimental, biological, and manufacturing datasets across diverse nanopharmaceutical platforms. These methods are particularly relevant for modeling protein corona formation, cellular uptake, intracellular trafficking, biodistribution, pharmacokinetics, toxicity, immunogenicity, and patient-specific responses. However, translational implementation remains limited by fragmented datasets, inconsistent reporting standards, limited interpretability, insufficient external validation, uncertain predictions, poorly defined applicability domains, and evolving regulatory expectations for adaptive computational models. Overall, AI should be viewed not only as an optimization tool, but also as a translational framework connecting formulation science, biological prediction, manufacturing control, and clinical implementation. Future progress will depend on standardized data infrastructures, explainable and externally validated models, uncertainty quantification, applicability-domain definition, hybrid mechanistic–AI frameworks, regulatory-ready documentation, and clinically relevant case studies. Full article

Recently, there has been great interest in AI-based approaches for de novo design of novel drug candidates. However, the generation of useful lead drug candidate compounds requires more than predicting engagement with the desired protein target. Candidate molecules must also be anchored in the real world of medicinal chemistry for their synthesis and modification as well as satisfying multiple drug development-related criteria. Here, we present Nevermore, an AI target-conditioned, database-grounded workflow for prioritizing candidate ligands from large compound libraries. Nevermore uses a geometry-aware protein–ligand affinity oracle to score target-specific binding and perform sparse integer edits in count-based Morgan fingerprint space. Nevermore then retrieves the most structurally similar molecules from public chemical databases. This design enables multi-objective search over predicted affinity and absorption, distribution, metabolism, excretion, and toxicity (ADMET) proxies while keeping all candidates anchored to valid database compounds. We evaluated Nevermore’s performance across three biologically distinct targets: Menin, a protein-interaction target relevant to leukemia; SARS-CoV-2 M^pro, a viral cysteine protease relevant to antiviral discovery; and epidermal growth factor receptor (EGFR), a kinase-superfamily oncology target with extensive experimentally tested compounds. Nevermore retrieved candidate sets with favorable predicted affinity–property trade-offs. These results support database-grounded fingerprint steering as a practical computational strategy for lead prioritization and for generating testable molecular hypotheses, although the prioritized candidates remain predictions, requiring follow-up experimental validation. Full article

This review examines the emerging role of machine learning (ML) in pharmaceutical chemistry, with emphasis on molecular design, synthetic feasibility, and structure–property–performance (SPP) relationships. By enabling pre-synthesis prediction of physicochemical properties, reaction pathways, and pharmaceutical performance, ML can reduce empirical trial-and-error experimentation and support more efficient exploration of chemical space. A structured narrative review design with PRISMA-aligned systematic search elements was used to evaluate 101 studies, enabling transparent literature identification, eligibility screening, and thematic synthesis across heterogeneous ML applications in pharmaceutical chemistry. This review examines structure–property relationships (SPRs) and property–performance relationships (PPRs), with emphasis on key pharmaceutical endpoints such as solubility, permeability, stability, dissolution, and bioavailability. An integrated SPP framework is proposed to connect molecular structure, intermediate properties, and final performance outcomes while incorporating retrosynthetic analysis and experimental feedback and closed-loop optimization. Recent frontier developments are also discussed, including molecular foundation models, multimodal language–graph models, diffusion-based molecular generation, E(3)-equivariant models, and MolMIM-like latent-space optimization. This review also covers co-folding and joint ligand–protein modeling, Boltz-2-like affinity prediction, AlphaFold 3-related biomolecular interaction modeling, and absorption, distribution, metabolism, excretion, and toxicity (ADMET) prediction. Key limitations include dataset leakage, benchmark inconsistency, assay variability, conformational and protonation-state effects, reproducibility challenges, regulatory constraints, and the gap between computational prediction and prospective experimental validation. Future progress is expected to depend on hybrid physics–ML models, uncertainty-aware prospective validation, autonomous experimentation, explainable artificial intelligence, and sustainability-aware molecular design. Overall, ML is evolving from a predictive tool into a chemically informed decision-support framework for rational, synthesis-aware, and experimentally validated pharmaceutical development. Full article

The Yezo virus (YEZV) is a recently discovered tick-borne orthonairovirus with pathogenic potential, causing acute febrile illness in humans. Viral nucleoproteins (N) play a key role in genome packaging, replication, and modulation of host immune responses, making their structural characterization essential for understanding viral pathogenesis and developing targeted countermeasures. However, the absence of structural data for YEZV proteins significantly hinders these efforts. This study presents the first solution structure of the YEZV N domain 1 (D1). A highly purified, soluble, tag-free recombinant YEZV N D1 was produced from the native sequence of the clinical YEZV isolate. The native-state conformation was resolved through an integrated approach combining size-exclusion chromatography coupled with small-angle X-ray scattering (SEC-SAXS), AlphaFold 3 structure prediction, and all-atom molecular dynamics simulations. The YEZV N D1 structure adopts a stable, predominantly α-helical globular fold that remains monomeric under near-physiological conditions. SEC-SAXS data show excellent agreement with computational models, revealing moderate conformational flexibility. The characterized recombinant YEZV N D1 and its first solution structure reported here providing essential insights into understanding of YEZV molecular architecture. These findings lay a foundation for rational serological assay development and structure-guided therapeutic design against this and other emerging orthonairoviruses. Full article

Genetically encoded biosensors are now central tools, deployed either as intracellular reporters to advance basic research, or as whole-cell reagents that detect analytes in diverse sample-types. Across the diversity of molecular scaffolds and modes of operation, biosensors serve a common functional purpose: translating ligand presence into a readable signal. Despite this shared logic, biosensor development as a field of practice remains fragmented: different scaffolds and modalities are advanced in separate, often lab-specific pipelines with diverse assays, metrics, and design practices. Moreover, libraries, selection histories and performance data generated during routine campaigns rarely outlive the projects that produced them. In this perspective, we focus on this fragmentation as a field-level bottleneck and argue that it deserves explicit attention in its own right. We discuss how modest, incremental steps—such as structured development records, adherence to high-information screening formats, library annotation, and community-level deposition infrastructure—could make biosensor development more reproducible, more comparable, and easier to build on across projects and laboratories. We further argue that such infrastructure will become increasingly valuable as computational protein design matures—not as a competing approach, but as the source of diverse, comparable, and context-annotated experimental data that sequence-function models and design benchmarks ultimately depend on. Full article