Early-stage decisions in the pharmaceutical development process carry outsized consequences for eventual clinical success, yet the computational tools underpinning these decisions remain fundamentally constrained. Computer-aided drug design (CADD) has transformed how researchers navigate chemical space and predict ligand–target interactions, but classical implementations rely on mechanical force-field approximations that fail to capture polarisation, charge transfer, and electron correlation effects central to molecular recognition and reactivity. Quantum-mechanical treatments that correctly describe these phenomena scale exponentially with system size on classical hardware, rendering them impractical for drug-relevant biomolecules. Quantum computing offers a physically motivated path beyond this scaling barrier: by exploiting superposition, entanglement, and interference, quantum algorithms can in principle simulate electronic structure with polynomial resource requirements for targeted problem classes. This article provides a theoretical review of how quantum computation integrates throughout the drug discovery pipeline, from target identification to lead optimisation. Its primary contribution is a pipeline-level mapping of quantum methods—including the Variational Quantum Eigensolver (VQE), quantum machine learning, and quantum-enhanced optimisation—to specific drug development stages. The review critically distinguishes near-term NISQ (Noisy Intermediate-Scale Quantum: current devices with 50–1,000 noisy qubits operating without full error correction) capabilities from fault-tolerant quantum computing (FTQC) requirements, quantifies current resource gaps through a worked CYP3A4 case study, and identifies algorithmic limitations (barren plateaus, ansatz expressibility, measurement overhead), hardware scalability, and error correction overhead as the principal barriers to practical deployment.
BackgroundImmune checkpoint inhibitors (ICIs), including pembrolizumab, are increasingly used in cancer treatment but can cause immune-related adverse events such as myasthenia gravis (MG). We report a case of pembrolizumab-induced myasthenia gravis characterized by isolated anti-titin antibody positivity in a patient with a history of thymoma and prior thymectomy.Case presentationA 65-year-old man presented with progressively worsening dysarthria, dysphagia, and dyspnea that developed 4 weeks after initiation of pembrolizumab. Neurological examination revealed external ophthalmoplegia and bilateral ptosis without significant limb weakness. Anti-acetylcholine receptor (AChR) antibodies and anti-muscle-specific kinase (MuSK) antibodies were negative, whereas anti-titin antibody was positive. Despite high-dose corticosteroid therapy, symptoms did not improve, but intravenous immunoglobulin treatment led to clinical improvement.ConclusionThis case is notable for isolated anti-titin antibody positivity in the absence of anti-AChR and anti-MuSK antibodies, without evidence of myositis or myocarditis, a phenotype rarely reported in ICI-related MG. The onset of MG 12 years after thymectomy may reflect persistent autoimmune susceptibility unmasked by immune checkpoint inhibition. This case expands the immunological spectrum of ICI-related MG and highlights the importance of neurological vigilance and consideration of anti-striational antibody testing, including anti-titin antibody, in patients with a history of thymoma receiving ICIs.
Structural activity cliffs are a well-established concept in structure–activity relationship studies, describing cases in which small structural modifications produce large changes in biological activity. Whether an analogous phenomenon exists for molecular physicochemical properties has received less attention and remains conceptually underdeveloped. Here, we examine the notion of property cliffs through a comparative analysis of structure–activity and structure–property landscapes. Using examples spanning acid–base behavior, lipophilicity, chemical reactivity, spectroscopy, molecular materials, and supramolecular systems, we argue that abrupt property changes associated with relatively small structural perturbations can occur across diverse chemical domains. We further discuss how property cliffs differ from activity cliffs in terms of their underlying mechanisms, potential frequency, and degree of mechanistic interpretability. Rather than proposing a universal definition, we outline a conceptual framework in which the identification of a property cliff depends on the property under consideration, the magnitude of the structural perturbation, and the property variation. Finally, we discuss how recognizing and systematically characterizing property cliffs may provide new opportunities for QSPR modeling, molecular machine learning, materials discovery, and the analysis of discontinuities in chemical space.
IntroductionThe role of phytochemicals like xanthines, and (poly)phenols in neurodegenerative diseases has been highly established and explored. However, the mechanism of these molecules to cross the membranes reaching the blood and the brain has largely been understudied.MethodsIn this work, we used in silico methods to predict the permeability of more than 800 molecules across the membranes and specifically the blood-brain barrier. Our dataset included phytochemicals, like xanthine, phenolic terpenes, (poly)phenols and metabolites, including phase II conjugates, while some endogenous molecules and commercial drugs were used as positive/negative controls. We used QikProp to generate 42 computed physicochemical properties to predict the blood absorption, distribution to the brain, metabolism, and elimination of these molecules.ResultsAccording to Qikprop 555 of 800 molecules were inside the range of 95% of known drugs for all computed properties, while through passive di[usion, 78 molecules may reach the blood and 52 may reach the brain parenchyma. Furthermore, using data from natural molecules present in human CSF samples, we used machine learning to build a model that was capable to predict 171 novel molecules with the potential to reach the brain environment.ConclusionOverall, in this work we predicted the natural molecules with the potential to reach the brain, highlighting those that may do so by passive mechanisms. Our study creates an opportunity to track novel molecules in theblood and the brain and highlighting the most drug-promising natural molecules for brain drug development in the treatment and prevention of brain diseases.
The global interest in nanotechnology has significantly increased in recent years as a necessity to create new and natural antibacterial agents required in the management of superbug pathogens. The present study characterized the green-synthesized copper nanoparticles derived from Aloe vera leaf extract and determined their antibacterial activity against methicillin-resistant Staphylococcus aureus (MRSA) and multidrug-resistant (MDR) Klebsiella pneumoniae. The green synthesized copper oxide nanoparticle (CuO NPs) from the aqueous extract of Aloe vera leaves were characterized using ultra violet (UV)-visible spectra, Transmission Electron Microscopy, Fourier transform infrared, and X-ray diffraction (XRD) and then assayed for antibacterial activity. The study revealed that green synthesized CuO NPs contained several organic functional groups of phytochemicals with a wave by CuO NPs plasmon resonance. CuO NPs exhibited a plate-like hexagonal structure with a length of approximately 500 nm. The CuO NPs exhibited broad-spectrum antibacterial activity that was concentration-dependent, exhibited lower MIC and MBC values on MRSA strains (31.25–62.5 μg/mL and 125 μg/mL) than against MDR Klebsiella pneumoniae strains (62.5 μg/mL and 500 μg/mL) respectively. The CuO NPs synthesized from Aloe vera had strong bactericidal activity, offering a promising alternative strategy in the global effort to combat multidrug resistant bacterial infection. There is a need to conduct in vivo efficacy of the CuO NPs to validate their safety for human use.
Artificial intelligence (AI) has become an important tool in drug discovery by enabling the analysis of large-scale molecular dynamics (MD) simulation data and improving the understanding of protein-ligand interactions. However, identifying functionally relevant ligand-binding conformations from highly dynamic MD trajectories remains a major challenge. We propose a spectral analysis-based AI/machine learning (ML) framework to improve the identification of ligand-binding protein conformations. The framework compares the Fast Fourier Transform (FFT) and Discrete Wavelet Transform (DWT) by transforming protein feature time series into frequency-domain and time-frequency-domain representations, respectively. These spectral features capture global conformational dynamics and localized structural changes. A probabilistic majority-voting decision-fusion strategy integrates predictions from multiple AI/ML models, while the spectral feature space is used to mitigate class imbalance, improve the signalto-noise ratio, and enhance discriminative learning. The framework was evaluated on three G protein-coupled receptors (GPCRs): ADORA2A, OPRD1, and OPRK1. Compared with baseline models trained on raw time-series data, the proposed approach achieved improved predictive performance. FFT-based features effectively captured high-frequency conformational signatures, whereas DWT-based features identified localized high-energy patterns associated with ligand-binding events. The decision-fusion strategy further improved the sensitivity, the area under the receiver operating characteristic curve (AUC), and the overall consistency of the prediction across all targets. Spectral-domain analysis substantially enhances AI-driven identification of ligand-binding protein conformations by providing complementary representations of protein dynamics and improving classification performance. This framework offers a robust approach for analyzing MD simulations in the discovery of structure-based drugs and can facilitate the identification of biologically relevant conformations. Future work will validate the predicted conformations through molecular docking and extend the framework to additional therapeutic protein targets.
The convergence of artificial intelligence and biomedical research has catalyzed emerging impact in drug discovery and precision medicine. Large language models (LLMs), originally developed for natural language processing, have emerged as powerful tools capable of processing complex biomedical data, from molecular structures to clinical records. This comprehensive review examines the state-of-the-art applications of LLMs across the drug development pipeline, spanning target identification, molecular generation, property prediction, and drug repurposing in drug discovery, as well as clinical decision support, patient stratification, treatment personalization, and genomic interpretation in precision medicine. We analyze the technical methodologies underlying these applications, including multi-modal architectures, knowledge-guided approaches, and retrieval augmented generation systems. Through examination of recent advances, we highlight key achievements such as end-to-end drug discovery pipelines, multi-agent systems for clinical simulation, and knowledge-enhanced models achieving state-of-the-art performance. We critically assess current challenges, including data privacy, model interpretability, hallucination risks, and ethical considerations. Finally, we discuss future directions, emphasizing the potential for federated learning, explainable artificial intelligence, and integrated multiscale approaches to advance the field toward more reliable, transparent, and clinically applicable systems.
Milk kefir is a fermented drink traditionally consumed by humans for approximately 3,000 years, shaped by a bacterial and yeast consortium, with probiotic capacity. During the fermentation process, the microbial community synthesizes a broad-spectrum of compounds working synergically, as organic acids, bacteriocins, exopolysaccharides, and other bioactive metabolites that inhibit diverse bacterial or fungal pathogens like Salmonella typhimurium, Pseudomonas aeruginosa or Fusarium oxysporum. These compounds are mainly produced by the kefir’s lactic acid bacteria and modifies the environment and disrupt pathogen cell membranes or promote pores into the cell walls in pathogenic bacteria, among other mechanisms. Therefore, in this review, we explore the antimicrobial bioactive compounds produced by the kefir bacteria and associated yeast, such as organic acids, bacteriocins, bioactive peptides, and exopolysaccharides, among others, not only acting as individual molecules but also as results of a complex combination of metabolic products.
Therapeutic peptides have emerged as promising candidates for combating antimicrobial resistance, particularly against multidrug-resistant pathogens for which conventional antibiotics are becoming increasingly ineffective. Although artificial intelligence has accelerated antimicrobial peptide discovery through predictive modelling, generative design, and large-scale in silico screening, many current workflows remain fragmented, weakly interpretable, and only loosely connected to experimental feedback. In this Perspective, we propose a methodological framework for resistance-aware antimicrobial peptide discovery built upon three complementary principles: explainable and uncertainty-aware prediction, biologically constrained and rule-guided generative design, and iterative design–test–learn workflows capable of continuously incorporating new evidence. By organising existing methodologies into adaptive and transparent discovery systems, the proposed framework supports candidate prioritisation, optimisation, and iterative refinement under evolving resistance pressures. To demonstrate these concepts in practice, we present a workflow integrating interpretable prediction, uncertainty-aware prioritisation, rule extraction, and adaptive model updating. Collectively, these elements provide a roadmap for advancing antimicrobial peptide discovery from isolated predictive tasks toward integrated and evidence-driven discovery ecosystems.
Clinical drug development is characterized by high attrition rates: approximately 90% of candidates entering Phase I fail to obtain approval. Inadequate efficacy accounts for 40 to 50 percent of terminations in Phase II and Phase III, while safety concerns account for roughly 30 percent. Most efficacy failures result from incorrect biological hypotheses rather than physicochemical deficiencies: drugs frequently bind their intended targets but fail to alter disease progression because the mechanistic assumptions underlying their action are invalid. This perspective introduces a feasibility-gating framework that utilizes physical constraints as explicit, early-stage filters. The framework does not claim that physics alone can predict therapeutic success; rather, it asserts that a systematic evaluation of thermodynamic, kinetic, transport, and microenvironmental constraints can identify candidates likely to fail on physical grounds prior to substantial investment in synthesis and translational research, thereby conserving resources for the inherently experimental validation of biological hypotheses. The individual constraints are based on well-established pharmacological principles; their contribution lies in organizing these principles into a checklist-style decision logic with specific thresholds, explicit guidance on interpreting each threshold, and a clear connection to quantitative systems pharmacology (QSP) and physiologically based pharmacokinetic (PBPK) modeling. We position this framework relative to these established quantitative disciplines, implement each gating criterion, and demonstrate the logic retrospectively using documented development histories. Physical reasoning serves as an essential yet limited filter: it identifies certain predictable physical risks early in the development process without guaranteeing success for candidates that pass through.
IntroductionReal-world data on dupilumab for moderate-to-severe atopic dermatitis (AD) in Latin America remain limited; therefore, we evaluated the real-world effectiveness and safety of dupilumab over 24 and 48 weeks in pediatric and adult patients in Colombia.MethodsThis observational, retrospective cohort included patients aged ≥6 years initiating dupilumab between 2020 and 2024 within a large integrated healthcare system. Clinician-reported (SCORAD, EASI) and patient-reported outcomes (DLQI, ADCT, POEM) were assessed at baseline and during routine follow-up, with primary evaluations at weeks 24 and 48, and analyzed using linear mixed-effects models. Therapeutic targets and minimal clinically important differences were assessed in paired datasets. Safety events were obtained from routine clinical documentation.ResultsA total of 547 patients met eligibility criteria (51.9% male; children 8.4%, adolescents 22.3%, adults 69.3%). Dupilumab was associated with reductionsacross all outcomes at weeks 24 and 48. At week 24, reductions were−59.3% in SCORAD, −71.9% in EASI, −66.8% in DLQI/CDLQI, −64.0% in ADCT and −60.0% in POEM (all p < 0.001), and were sustained or greater at week 48 (−65.2%, −79.3%, −67.1%, −66.4% and −66.8%, respectively). At week 48, 63.5% achieved EASI-75% and 36.5% achieved EASI-90. Improvements were consistent across age groups; adverse reactions were infrequent and mostly mild, most commonly headache and skin rash.ConclusionDupilumab demonstrated robust, durable clinical benefits and a favorable safety profile in routine practice, supporting its effectiveness in diverse real-world populations.
Pairwise learning is an emerging paradigm in cheminformatics that trains machine-learning models on pairs of molecules and their property differences rather than on individual compounds and absolute values. This formulation enables models to directly learn the effects of molecular modifications and more effectively supports comparative decision-making central to drug discovery. Here, we review recent advances in molecular pairwise learning and discuss how this framework can improve predictive performance, provide uncertainty quantification, and enable effective modeling in low-data and heterogeneous data source settings. We further highlight key applications in molecular optimization, outline current limitations, and identify future opportunities for pairwise learning to enhance drug development.
Background and PurposeBortezomib and Carfilzomib are first- and second-generation proteasome inhibitors that revolutionized the treatment of multiple myelomas. Despite their efficacy, they have been associated with off-target adverse events. Carfilzomib has a high rate of cardiovascular adverse events including heart failure, hypertension, ischemic heart diseases and arrhythmias. On the other hand, Bortezomib is known to cause peripheral neuropathy but has a safer cardiac profile. The mechanism behind the differential cardiac toxicity of Bortezomib and Carfilzomib is not completely understood. Thus, we aim to investigate the chronic effect of low nanomolar concentrations of Bortezomib and Carfilzomib on the electrophysiology of human-induced pluripotent-derived cardiomyocytes.Methods and Resultsin-house differentiated human-induced pluripotent-derived cardiomyocytes were incubated with Bortezomib or Carfilzomib for 16 h. Patch-clamp experiments were conducted to record spontaneous action potentials and ionic currents. Carfilzomib affected the electrophysiology of spontaneous action potentials by altering the calcium and potassium currents, without affecting the sodium current. Bortezomib showed a milder effect on action potentials, probably due to a lack of effect on the potassium current IKr and an opposite compensatory effect on the calcium and sodium currents.Conclusionthis study proposes a novel potential pro-arrhythmic mechanism that may contribute to elucidate the differential cardiotoxicity of Bortezomib and Carfilzomib.
Glioblastoma multiforme (GBM) is the most aggressive brain tumor and remains one of the most difficult malignancies to treat. The main obstacle in GBM treatment is the limited ability of most drugs to cross the blood-brain barrier (BBB), which together with the heterogeneity and complexity of GBM, contributes to therapeutic failure, drug resistance, and poor clinical translation. Moreover, despite only a few agents (temozolomide, lomustine, carmustine, and bevacizumab) have received approval from the U.S. Food and Drug Administration (FDA), their efficacy remains modest and there is the need to design more effective strategies. In this context, to address these limitations associated with conventional drug administration, biomaterials-based delivery systems have emerged as promising platform for their biodegradability, ability to enhance BBB penetration, promote drug accumulation at the tumor site, and ultimately improve therapeutic efficacy. In this review, we provide a critical description of the natural and synthetic biomaterials outlining their physicochemical properties, advantages for the fabrication of delivery systems and their limitations in preclinical and clinical settings. Further, we also provide a comprehensive overview of biomaterials functionalized with therapeutic agents for GBM treatment, highlighting their composition, delivery strategies, and preclinical and clinical outcomes. Overall, by bridging biomaterial design with tumor biology, this review aims to identify novel directions for the development of innovative biomaterials-based drug delivery systems to improve therapeutic options in GBM treatments.
Angiotensin-Converting Enzyme (ACE, EC 3.4.15.1), also termed as kininase II or dipeptidyl carboxypeptidase I, is a crucial component of the Renin angiotensin system (RAS). ACE is extensively recognized for its role in the regulation of the cardiovascular system. Emerging evidence supports a neurobiological role of ACE, promoting investigations into its central functions responsible for the modulation of memory and cognitive function, particularly through the use of clinically approved ACE inhibitors (ACE-Is). This review comprehensively analyzes preclinical and clinical findings to date to evaluate the contribution of ACE to cognitive processes and discusses its potential as a therapeutic target for cognitive improvement. Mechanistically, ACE influences cognition through regulation of angiotensin II (Ang II) signaling, as well as modulating other neuropeptides such as substance P. These pathways intersect with processes implicated in cognitive decline, including oxidative stress, neuroinflammation, and cholinergic neurotransmission. Pre-clinical studies consistently demonstrate that ACE-Is confer neuroprotective effects by attenuating Ang II-mediated signaling, restoring redox and inflammatory balance, and improving synaptic plasticity. Clinical evidence provides partial support to these findings, although results remain heterogeneous. Variability in clinical outcomes appears to depend on factors such as patient age, baseline comorbidities, and central availability of the ACE-I. Collectively, this study highlights a non-canonical role of central ACE inhibition in modulating cognitive function, suggesting its promise as a therapeutic strategy for addressing age and disease-associated cognitive dysfunction. Future research should thus prioritize delineation of tissue-specific actions of ACE along with comprehensive examination of ACE genotype-drug interactions, and head-to-head comparison of centrally active ACE-Is with other RAS modulators for their nootropic potential.
Nearly 90% of drug candidates that enter clinical trials fail to receive approval from the U.S Food and Drug Administration. Approximately one-quarter of these failures are due to adverse effects in the CNS, most of which are caused by alterations in neural function. To address this issue, we employed a neuro-behavioral assay, originally designed for identification and characterization of the effects of general anesthetics on the neural oscillations of an endogenous brainstem nucleus in a fish model system. To establish proof of concept, we validated the applicability of this high-content in vivo assay by assessing possible neural (side-)effects of diphenhydramine, best known under its brand name Benadryl. In the neuro-behavioral assay, this antihistamine revealed a robust, reversible, dose-dependent neural effect, reflected by a pronounced drop in the oscillation frequency readout of the assay. This effect, presumably caused by use-dependent blocking of sodium channels, is consistent with secondary actions of diphenhydramine as a sedative and/or local anesthetic. The results of our study demonstrate the potential of the neuro-behavioral assay to predict functional (side-)effects of drug candidates in the central nervous system at earlier stages of drug development than possible with the currently available technologies.
IntroductionChikungunya fever remains a growing global health challenge, with a subset of infected individuals developing long-lasting and severe disease manifestations. Although significant progress has been made in chikungunya virus (CHIKV) antiviral research, no licensed therapeutic options are currently available. In this context, drug repurposing offers a strategic approach to accelerate the identification and development of effective CHIKV treatments.MethodsIn this study, we conducted a high-throughput phenotypic screening assay based on a reporter CHIKV to evaluate 1,600 clinically approved or well-characterized compounds from the Repurposing, Focused Rescue, and Accelerated Medchem (ReFRAME) library and initiated the characterization of the in vitro antiviral profile of the most promising candidate.ResultsThe developed high-throughput screening platform effectively identified 13 candidate antivirals. Farudodstat, a human dihydroorotate dehydrogenase (DHODH) inhibitor, emerged as the most promising candidate. It demonstrated robust inhibitory activity against epidemic CHIKV strains and efficacy that varied across cell lines of distinct origins, consistent with a host-dependent mechanism of action. As expected from the role of DHODH in the de novo pyrimidine biosynthesis pathway, supplementation with exogenous uridine restored viral replication in the presence of Farudodstat, indicating that blockade of DHODH underlies its anti-CHIKV activity.DiscussionThe potency and outstanding selectivity of Farudodstat as a CHIKV inhibitor in vitro, together with its favorable safety and pharmacokinetic properties, support the potential of Farudodstat as a promising oral antiviral candidate for the treatment of CHIKV fever.
IntroductionDermatitis of the external ear canal is an inflammatory condition characterized by itching, desquamation, erythema, and discomfort. Although topical corticosteroids are commonly used as first-line therapy, prolonged administration may lead to adverse effects, including skin atrophy and irritation. Therefore, alternative treatments with a safer profile are being explored. Ozonized oil has gained attention due to its antimicrobial, anti-inflammatory, and tissue-repairing properties. This study aimed to evaluate the effectiveness and safety of ozonized oil-based ear drops in reducing symptoms and improving clinical signs in patients with non-infectious dermatitis of the external ear canal.Materials and MethodsIn this prospective observational study, adult patients diagnosed with non-infectious dermatitis of the external ear canal were enrolled. Participants were treated with ozonized oil-based ear drops twice daily for 2 weeks. Clinical effectiveness was assessed before and after treatment using the Modified Itch Severity Scale (MISS), the Visual Analog Scale (VAS) for pain, and otoscopic evaluation of hyperemia, edema, and desquamation. Statistical analysis was performed using the Wilcoxon signed-rank test.ResultsThirty-five patients were enrolled, and 33 completed the study. Treatment resulted in a significant reduction in itching frequency and intensity (p < 0.001), as well as improvement in otoscopic signs (p < 0.001). Sleep onset improved significantly, while nighttime awakenings did not. Mild otalgia occurred in three patients and resolved. No adverse effects were reported.ConclusionOzonized oil-based ear drops were associated with an improvement in symptoms and otoscopic findings in this prospective observational study. However, controlled studies are required to confirm these findings.
Osteoarthritis is a chronic, degenerative joint disease characterized by cartilage breakdown, inflammation, and pain, significantly affecting the quality of life of affected individuals. Current treatments focus on symptom relief but do not halt disease progression, highlighting the need for novel therapeutic strategies. Cannabinoid receptor type 2 (CB2R) has emerged as a promising target for this purpose due to its role in inflammation and pain modulation, with minimal psychoactive effects compared to cannabinoid receptor type 1 (CB1R). This study performed a structure-based virtual screening to identify potential CB2R ligands with high affinity and favorable pharmacokinetic properties. Molecular docking studies led to the selection of compound P415 (PubChem CID: 154468691, 4-[[(1R,2S,6S,14R,15R,16R)-11,15-dimethoxy-5-methyl-13-oxahexacyclo [13.2.2.12,8.01,6.02,14.012,20]icosa-8(20),9,11-trien-16-yl]methoxymethyl]-3,5-dimethyl-1,2-oxazole) for further evaluation, using WIN55,212-2 (PubChem CID: 5311501, [(11R)-2-methyl-11-(morpholin-4-ylmethyl)-9-oxa-1-azatricyclo[6.3.1.04,12]dodeca-2,4(12),5,7-tetraen-3-yl]-naphthalen-1-ylmethanone) as a reference agonist. Molecular dynamics simulations (500 ns) were also conducted in triplicate to assess the stability and dynamic behavior of CB2R-ligand complexes. The RMSD and RMSF analyses revealed distinct conformational stability patterns, with the CB2R-P415 complex showing a more stabilized profile over time with an average RMSD of 0.4 nm. Analysis of the most stable replicates through MM/PBSA binding free energy calculations yielded −51.00 kcal/mol for CB2R-P415 and -55.65 kcal/mol for CB2R-WIN55,212-2. ADMET predictions indicated that P415 possesses drug-like properties, including a log P of 3.36, TPSA of 62.95 Å2, and an LD50 of 3.086 mg/kg, with no predicted AMES toxicity or hepatotoxicity. These results suggest that P415 is a promising CB2R ligand with high binding affinity and structural stability, warranting further experimental validation for potential therapeutic applications in osteoarthritis treatment.
