Long-term, flexible therapy strategies are needed for chronic diseases like cardiovascular ailments, diabetes, and chronic kidney disease and they have to deal with patient heterogeneity, changing physiological states and privacy constraints. This research suggests a framework for privacy-preserving and personalized artificial intelligence that combines Federated Learning (FL), a Res-HyperTransformerNet deep prediction model, Personalized Multi-Agent Reinforcement Learning (PMARL) and Explainable Artificial Intelligence (XAI) for adaptive chronic disease management. The framework is structured with two main aims: (i) to create an accurate and privacy-aware predictive model for chronic disease risk using diverse data sources, and (ii) to permit adaptive, individual therapy optimization through multi-agent reinforcement learning. Federated learning is utilized to conduct the training of Res-HyperTransformerNet over distributed Internet of Medical Things (IoMT) nodes without the requirement of transferring raw patient data. The embeddings that are created from patient data are then passed on to a PMARL module, where several agents optimize therapy dimensions like medications, diet, physical activity, and mental health interventions independently. To make the clinical support more understandable, the SHAP-based explainability method is used for both predictive and decision-making parts. The framework is tested using two public datasets—the CDC Chronic Disease dataset and the UCI Chronic Kidney Disease Risk Factor dataset. The performance is measured by employing classification metrics (accuracy, precision, recall, F1-score, MCC) as well as reinforcement learning metrics (reward score, convergence steps, episode return) and federated system metrics (communication overhead, convergence rounds, and training time). The experiments indicate that the proposed framework demonstrated improved performance in terms of predictive accuracy and policy convergence speed compared to the baseline deep learning and reinforcement learning models, and at the same time it is more cost-effective when it comes to communication in a federated setting. This means that the proposed method suggests the potential of being a reproducible and open AI framework for adaptive chronic disease therapy management. The combination of ResNet and Transformer blocks, i.e., Federated Res-HyperTransformerNet, achieved strong performance in both datasets, namely CDC Chronic Disease (Dataset 1) and UCI CKD Risk Factor (Dataset 2). The obtained accuracy for Dataset 1 was 98.61% and for Dataset 2, it was 97.75%, under the conducted experimental settings.
Neurodegenerative disorders such as Parkinson’s disease are closely associated with dysregulated activity of monoamine oxidase-B (MAO-B), which leads to dopamine depletion and oxidative stress. Despite the availability of numerous monoamine oxidase (MAO) inhibitors on the market, their irreversibility and associated side effects necessitate the development of more effective and reversible MAO-B inhibitors. In this study, a series of twenty-one indole-based derivatives (PSH1 - PSH21) , collectively designated as PSH , was synthesised and evaluated for inhibitory activity against MAO isoforms. Most synthesised compounds showed higher inhibitory activity toward MAO-B than MAO-A, indicating selectivity for MAO-B. Among the PSH derivatives, PSH18 exhibited the most potent MAO-B inhibitory activity (IC 50 = 0.95 ± 0.02 µM), followed by PSH6 ( IC 50 = 1.79 ± 0.40 µM) and PSH2 (IC 50 = 1.96 ± 0.08 µM). Compound PSH18 showed the highest selectivity index value of 42.11, followed by PSH6 (22.35) and PSH2 (20.41). Additionally, PSH18 was confirmed to be a competitive and reversible inhibitor of MAO-B with an inhibition constant value of 0.89 ± 0.035 µM. Notably, PSH18 exhibited good permeability across the blood–brain barrier in parallel artificial membrane permeability assay experiments, along with acceptable absorption, distribution, metabolism, excretion, and toxicity parameters predicted through in silico modelling, suggesting its potential as a central nervous system-targeted molecule. Molecular docking and a 200 ns molecular dynamics simulation demonstrated stable binding of the ligand to the MAO-B active-site pocket, driven by hydrophobic and π–π stacking interactions with key amino acid residues lining the aromatic cage. Moreover, the calculated binding energy indicated strong ligand–protein interactions. Overall, these results indicate that the PSH scaffold could serve as a promising lead structure for the development of potent and selective MAO-B inhibitors. These compounds may have therapeutic potential for the treatment of neurodegenerative diseases.
monoamine oxidase B inhibitorsParkinson's diseaseblood-brain barrier permeabilityADME prediction
The increasing complexity and sophistication of cyber threats have heightened the need for robust security mechanisms throughout the Software Development Lifecycle (SDLC). Traditional security methods have proven insufficient in addressing evolving risks, prompting the exploration of advanced technologies such as Artificial Intelligence (AI). However, integrating AI into SDLC raises challenges related to human-AI collaboration, trust, and alignment. This paper aims to explore the synergistic collaboration between Human Agentic and AI agents to enhance the security of SDLC. By integrating human expertise with AI capabilities, the study investigates how this collaboration can address existing security gaps and improve threat detection, vulnerability management, and secure code development. A qualitative approach is employed, reviewing existing literature on Human-Agentic AI frameworks, AI-driven security models, and their application in SDLC security. We also propose a conceptual framework for Human Agentic-AI collaboration in SDLC security, encompassing AI’s role in automated threat detection, vulnerability scanning, and model optimization, alongside human agents’ roles in decision-making, risk evaluation, and response validation. The proposed framework demonstrates that combining human intuition with AI’s computational power significantly enhances the security posture of SDLC processes. AI’s ability to quickly process and analyze large datasets complements human agents’ contextual understanding, leading to more effective identification of vulnerabilities and proactive security measures. The collaboration results in improved threat mitigation strategies, faster response times, and more reliable secure code deployment. The integration of Human Agentic-AI collaboration offers promising potential in enhancing SDLC security. The synergistic approach bridges the limitations of traditional methods and AI systems, offering a more dynamic and adaptable security solution. Future research should focus on optimizing human-AI interfaces and addressing trust and transparency challenges to further strengthen SDLC security.
International Journal of Polymeric Materials · مجلد 75 (8) · 835-858
Folate-conjugated chitosan nanoparticles represent a promising advancement in targeted drug delivery, offering enhanced specificity, biocompatibility, and controlled release profiles. Chitosan, a versatile biopolymer, provides a stable platform for nanoparticle synthesis, while folate conjugation enables selective targeting of folate receptors, which are overexpressed in various cancers and inflammatory diseases. This review explores the structural and functional properties of chitosan, mechanisms of folate receptor-mediated endocytosis, and innovative strategies for nanoparticle synthesis and functionalization. The review entails key formulation techniques, including nanoprecipitation, ionic gelation, and spray drying, alongside characterization methods essential for optimizing therapeutic efficacy. The biomedical applications of Folate-conjugated chitosan nanoparticles are highlighted in cancer therapy, gene delivery, and anti-inflammatory treatments, with an emphasis on emerging trends such as CRISPR-based targeting and dual-ligand strategies. This review critically examines the existing challenges and explores future directions, highlighting the promising role of folate-conjugated chitosan nanoparticles in the evolution of precision medicine.
targeted drug deliveryfolate receptorreceptor-mediated endocytosisnanoprecipitation
The SARS‐CoV‐2 main protease (M pro ) remains a critical therapeutic target for COVID‐19 intervention. In this study, we employed a comprehensive in silico repurposing strategy to identify FDA‐approved drugs with potential inhibitory activity against M pro . An initial molecular fingerprint screening of 3009 FDA‐approved compounds, based on similarity to the cocrystallized inhibitor X77, yielded 150 similar candidates. Subsequent structural similarity filtering reduced the list to 30 compounds, which were subjected to molecular docking. Apixaban demonstrated strong binding affinity and favorable interaction patterns within the M pro active site and was selected for further investigation. A 100‐ns molecular dynamics simulation comparing the apo and holo forms confirmed the structural stability of the M pro –apixaban complex. Apixaban remained stably positioned within the binding pocket, while reduced RMSF values in key active‐site residues indicated apixaban‐induced stabilization. Molecular mechanics–generalized Born surface area (MM‐GBSA) analysis estimated a binding free energy of −22.19 kcal/mol, with van der Waals interactions contributing predominantly to complex stability. Per‐residue decomposition highlighted significant contributions from Met165 and Gln189. Protein–ligand interaction fingerprinting (ProLIF) and protein–ligand interaction profiler (PLIP) analyses revealed persistent hydrogen bonds and hydrophobic contacts involving Asn142, Cys145, Gln189, Thr190, Gln192, and Arg188. Principal component analysis of trajectories (PCAT) demonstrated apixaban‐induced modulation of collective motions and partial closure of the binding pocket. Importantly, this study provides detailed atomistic insights into the potential interaction of apixaban with M pro . Together, these findings offer a plausible molecular rationale for its inhibitory activity and support further experimental and pharmacological evaluation of apixaban as a potential therapeutic candidate targeting SARS‐CoV‐2 M pro .
Glucagon-like peptide-1 receptor agonists (GLP-1 RAs), originally developed as antidiabetic agents, have emerged as candidate disease-modifying therapies for Parkinson’s disease (PD) and Alzheimer’s disease (AD), reflecting the widespread expression of GLP-1 receptors across neurons, glia, and the cerebral vasculature. This narrative review synthesizes mechanistic, clinical trial, and biomarker evidence for GLP-1 RAs in PD and AD, based on a literature search of PubMed, Embase, Scopus, and Web of Science, with ClinicalTrials.gov searched separately for trial protocols and results, through May 2026. Preclinical data implicate GLP-1 receptor agonism in anti-neuroinflammatory signaling, mitochondrial preservation, autophagy and proteostasis enhancement, anti-apoptotic effects, synaptic plasticity, cerebrovascular protection, and glymphatic clearance. Clinical evidence in PD is mixed and agent- and formulation-dependent: the phase II Exenatide-PD trial reported a motor benefit that persisted after washout (∼3.5-point between-group difference in MDS-UPDRS Part III), and the phase II LIXIPARK trial (lixisenatide) reported a significant 3.08-point between-group difference (95% CI 0.86–5.30; p = 0.007) favoring lixisenatide, whereas the larger, longer phase III Exenatide-PD3 trial, using an extended-release weekly formulation, found no significant effect on the same endpoint (adjusted coefficient 0.92; 95% CI −1.56 to 3.39; p = 0.47); oral semaglutide is under investigation in the ongoing phase II MOST-ABLE trial. In AD, the phase IIb liraglutide (ELAD) trial did not achieve statistical benefit on either its primary imaging endpoint (cerebral glucose metabolic rate; p = 0.14) or on most secondary cognitive measures, and the phase III semaglutide trials EVOKE and EVOKE+ did not meet their primary CDR-SB endpoint (estimated difference −0.08, 95% CI −0.35 to 0.20, p = 0.57 in EVOKE; 0.10, 95% CI −0.17 to 0.38, p = 0.46 in EVOKE+), despite biomarker signals including reductions in cerebrospinal fluid tau species and inflammatory markers. Comparative pharmacology reveals substantial heterogeneity in central nervous system penetration, receptor engagement, and systemic bioavailability across exenatide, liraglutide, semaglutide, and the imbalanced GLP-1/GIP co-agonist tirzepatide, complicating cross-agent extrapolation. Persistent challenges include separating direct neuroprotection from indirect metabolic and cardiovascular benefit, limited characterization of brain drug exposure, trial durations that may be short relative to the expected latency of a disease-modifying effect, and the absence of validated response biomarkers. We conclude that although the mechanistic and epidemiological rationale for GLP-1 RAs in neurodegeneration remains biologically plausible, current trial evidence does not yet support their use outside clinical trials, irrespective of comorbid diabetes status; future studies should prioritize earlier intervention, biomarker-guided patient selection, adequately powered subgroup analyses, and brain-penetrant formulations.
Diabetic neuropathy (DN) is a multifactorial complication of diabetes mellitus driven by chronic hyperglycemia, insulin resistance, and disturbed metabolic homeostasis, leading to progressive injury of both the peripheral and central nervous systems. This study investigated whether L-serine supplementation could attenuate DN through dose-dependent metabolic and neuroprotective mechanisms in a high-fat diet (HFD) plus streptozotocin (STZ)-induced diabetic rat model. Male Wistar rats (n = 8 per group) were allocated to five groups: normal control (NC), diabetic control (DC), pioglitazone (PIO; 1.5 mg/kg/day), low-dose L-serine (S1; 200 mg/kg/day), and high-dose L-serine (S2; 400 mg/kg/day). After 60 days of oral gavage, behavioural testing, glucose and insulin profiling, HOMA-IR calculation, brain histopathology, nerve growth factor (NGF) immunohistochemistry, and LC–MS/MS-based proteomic analysis of cerebral tissue were performed. Diabetic rats exhibited marked hyperglycaemia (355.33 ± 4.72 mg/dL), hyperinsulinaemia, severe insulin resistance (HOMA-IR 16.8 ± 3.2; a 14-fold increase), impaired thermal nociception, motor dysfunction, and pronounced neuronal degeneration. L-serine supplementation significantly improved metabolic status: S1 reduced HOMA-IR by 77.4% and S2 by 87.5% relative to diabetic controls (p < 0.001). High-dose L-serine produced greater improvements in thermal sensitivity, motor coordination (rotarod latency 26.67 ± 1.52 s vs. 16.1 ± 0.85 s in DC; p < 0.05), and NGF expression (8.6-fold increase vs. DC). Histopathology confirmed attenuation of neuronal injury and gliosis in both treatment groups. Exploratory, group-level proteomic profiling identified dose-specific molecular signatures: S1 was predominantly associated with carbohydrate, lipid, and biosynthetic pathways, whereas S2 was associated with synaptic, neurotransmission-related, and proteostasis pathways. Within the constraints of an exploratory design—group-level pooled proteomics, analysis of cerebral rather than peripheral-nerve tissue, and only two doses—these findings indicate that L-serine attenuates the metabolic and behavioural features of experimental diabetic neuropathy and generates the testable hypothesis of dose-dependent neuro-metabolic remodelling. The proteomic signatures are hypothesis-generating and require orthogonal validation before any mechanistic or translational inference can be drawn.