The development of novel antibiotics is urgently needed due to the rise in multidrug-resistant (MDR) infections globally. Marine microorganisms have emerged as unexplored sources of bioactive compounds with unique mechanisms to combat antimicrobial resistance (AMR). This review summarizes peer-reviewed and recent research from PubMed, Scopus, and Web of Science to evaluate the scope of marine-derived metabolites against AMR pathogens. The focus was on studies describing antimicrobial metabolites produced by marine actinomycetes, bacteria, and fungi. Studies involving non-resistant pathogens or non-marine sources were excluded. The review identifies various secondary metabolites, including peptides, polyketides, macrolides, fatty acid amines, equisetin, and aminolipids. These compounds target AMR pathogens such as MRSA, VRE, and VRSE. Marine actinomycetes-including Streptomyces spp., Micromonospora spp., and Verrucosispora spp.-are key producers of antimicrobial compounds. Likewise, marine bacterial species such as Bacillus spp., Aequorivita spp., and Zobellia galactanivorans, as well as fungal species like Penicillium spp. and Fusarium spp., produce bioactive compounds. These findings highlight the potential of marine microbes as sources of novel antibiotics. Despite this potential, several limitations persist, including scalability issues, the noncultivability of many microbes, toxicology concerns, source-dependent variability, and insufficient in vivo validation. These challenges can be addressed by integrating modern omics technologies to advance drug discovery for AMR. This review underscores marine microbes as promising sources in the fight against AMR. To unlock their full potential, greater interdisciplinary collaboration is needed to develop sustainable solutions, accelerate antibiotic discovery, and address the global AMR crisis.

Thrombin (THR) is a key therapeutic target for anticoagulant therapy, yet the mechanism of β-sitosterol, a natural compound with antithrombotic potential, remains unclear. This study integrated AI-driven structural alignment, molecular docking, Molecular Dynamics (MD) simulations, binding free energy calculation, and Density Functional Theory (DFT) calculations to elucidate the recognition mechanism between THR and β-sitosterol. Simulations revealed that β-sitosterol binding is stabilized primarily by hydrophobic and van der Waals interactions, leading to the closure of the active site and conformational changes in the EF_loop (i.e., γ-loop). The large conformational changes within EF_Loop may be dominated by weak interactions between W168/ P184/ Q183/ S185 and the ligand β-sitosterol. Based on these insights, a series of novel sterol derivatives was designed with improved binding affinity and predicted antithrombotic activity, as indicated by the lowest binding free energy. This study not only reveals molecular recognition and inhibitory mechanism of βsitosterol at the atomic level, but also provides suggestions for structural optimization of novel inhibitors against human thrombin. Future work should include in vitro binding assays and in vivo functional studies to confirm the inhibitory activity. The conformational change of EF_loop with the recognition of β-sitosterol effectively occludes the catalytic site, thereby impairing thrombin's proteolytic activity. Among 13 designed sterol derivatives, the compound d3 was identified as a promising inhibitor with excellent ADMET properties. This work provides an anticoagulant mechanism for the dynamic identification of βsitosterol and supports the rational design of allosteric THR inhibitors.

Signal Transducer and Activator of Transcription 3 (STAT3) is a key mediator in Breast Cancer (BC) progression, contributing to tumor proliferation, metastasis, survival, and resistance to chemotherapy. Phosphorylation of STAT3 at tyrosine 705 promotes its dimerization and nuclear translocation, where it activates oncogenic transcriptional programs. Due to its central role in BC pathogenesis, STAT3 has emerged as a promising molecular target for therapeutic intervention. To synthesize, characterize, and evaluate the anticancer efficacy of synthetic and natural compounds with a focus on their ability to inhibit STAT3 phosphorylation, suppress breast cancer cell proliferation, and induce apoptosis and autophagy. A comprehensive literature review was conducted using databases such as PubMed, Scopus, Relemed, and ResearchGate. Relevant studies were identified that examined the synthesis, molecular mechanisms, and therapeutic potential of STAT3 inhibitors. Synthetic derivatives and phytochemicals were considered for their inhibitory effects on STAT3 activation and associated cellular outcomes in breast cancers. Several synthetic and natural compounds demonstrated significant inhibitory effects on STAT3 phosphorylation, leading to reduced breast cancer cell proliferation, migration, and survival. These agents effectively induced apoptosis and, in some cases, autophagy, highlighting their multifaceted anti-tumor mechanisms and elucidating the potential of these compounds as lead candidates for further preclinical and clinical development. Targeting STAT3 can be a significant therapeutic strategy, as both synthetic and natural compounds capable of inhibiting STAT3 signaling have been shown in preclinical studies. These findings provide valuable insights for cancer biologists, molecular researchers, and clinicians to explore STAT3 inhibitors as potential breast cancer therapeutics.

Medicinal plants and phytocompounds targeting skeletal muscle wasting in humans are under-represented in the majority of databases reporting plant/herb-diseases association. However, a large body of literature exists wherein plant extracts or active pharmaceutical ingredients thereof demonstrate potential benefit in skeletal muscle wasting diseases across model organisms. Underscoring the relevance of a repertoire documenting such medicinal plants, we introduce PDMD (Plants Database for Muscle Wasting Diseases), a manually curated plants database reported for muscle wasting diseases such as cachexia, sarcopenia, muscle atrophy, muscle frailty, impaired muscle regeneration, and muscle fatigue. PDMD was developed through systematic manual collection and curation of published studies from PubMed, Science Direct, etc, retrieving literature on plants conferring pharmacological efficacy against muscle wasting across experimental model organisms. Phytochemical and taxonomic information were extracted via tools like ClassyFire, PubChem. To handle the storage of an annotated listing of plants, MS-Excel and MySQL were used. Frontend was designed in Visual Studio Code and HTML/CSS. An Apache/PHP server was used to integrate MS-Excel data and charts. PDMD encompasses 206 medicinal plants, 230 API reported across 18 model organisms, offering taxonomical information, phytochemical classes, SMILES structure, geographical distribution, and other bioactivity indications. PDMD is cross-referenced with standard databases such as PubChem and PubMed for enhanced functionality. PDMD highlights overlooked plant-muscle links, bridging ethnopharmacology and botany gaps, and can aid hypothesis generation for novel therapies. PDMD highlights overlooked plant-muscle links, bridging ethnopharmacology and botany gaps, and can aid hypothesis generation. PDMD is freely available at https://www.jiit.ac.in/biotechhighlightes/Research-Databases/PDMD/index.html, and was last updated in September 2025.

Tuberculosis (TB), caused by Mycobacterium tuberculosis (Mtb), remains a major public health challenge. In this study, the Mtb proteasome was targeted as a promising site for novel drug development. A total of 190,295 natural compounds from the ZINC database were screened using a systematic approach involving Lipinski's rule of five, SwissADME, pkCSM analysis, PyRx, and molecular dynamics simulation to identify potential drug candidates. Finally, five compounds, namely, ZINC14688701, ZINC299835179, ZINC14638395, ZINC299839873, and ZINC14638400, were identified based on physicochemical, pharmacokinetics, drug-likeness properties, and free energy of binding. Among these, ZINC14688701 showed the highest free energy of binding (-9.3 kcal/mol) with the selected target Mtb proteasome through the amino acid residues Thr1, Arg19, Ser20, Thr21, Val31, Lys33, Gly47, Thr48, Ala49, Leu99, Ser141, and Ala180. Finally, the complex 'Mtb proteasome-ZINC14688701' was studied using Molecular Dynamics simulation (MD simulation) for 50 ns, in which RMSD, RMSF, Rg, H-bonds, and SASA showed complex stability. Physicochemical evaluations revealed that the compounds are non-toxic with favorable drug-likeness. Exploring the antibacterial natural inhibitors offers a promising strategy for novel drug development against infectious diseases. These findings suggest a potential inhibitor of Mtb proteasome that could be used for TB treatment.

The current study aims to determine the structure of the protein Kallikrein 11 and to screen for small natural product ligands to identify inhibitors of Kallikrein 11. Kallikreinrelated peptidase 11 (KLK 11) belongs to the Kallikrein family of Serine proteases. Kallikrein 11 is a multifunctional protease. In addition to causing cancer, this plays a critical role in a variety of physiological functions, including blood pressure regulation, sperm liquefaction, and skin desquamation. This study aims to identify the protein's 3D structure, perform virtual screening with a natural product database, and find ADME characteristics for the most desirable ligand retrieved. Additionally, it aims to evaluate the effectiveness of binding affinity-based scoring systems in differentiating active KLK11 inhibitors from decoy compounds through the use of Receiver Operating Characteristic (ROC) analysis. Using homology modelling protocols, the theoretical model of Kallikrein 11 will be predicted, and the resulting structure will be validated by several server tools. To identify new scaffold compounds that are effective against Kallikrein 11, the active site is examined, and the ligand database is used for virtual screening. The ROC-Area Under the Curve (AUC) is used to assess the effectiveness of inhibitors. The HIS94, ASP142, and SER235 residues in the KLK 11 protein are essential as the active site triad, and residues from GLY24 to ASN281 were chosen as a pocket for ligand molecule binding, according to the results of the virtual screening. With an AUC of 0.837, the results show a strong predictive ability, indicating that binding affinity is a trustworthy parameter for early virtual screening pipelines that target KLK11. Given its superior ADME qualities, the scaffolds containing the polyphenols and flavone pharmacophores were recognized as a potential lead drug against the KLK 11 protein. The findings confirm the reliability of the homology-modelled KLK11 structure and demonstrate that its catalytic triad and binding pocket can effectively distinguish active scaffolds through virtual screening. The strong ROC-AUC value indicates that binding-affinity-based selection is robust for early inhibitor discovery. Notably, the natural-product scaffolds displayed higher binding affinities than approved drugs, highlighting their potential as superior KLK11 inhibitor candidates. The research results demonstrated that the chosen ligand molecules with ADME parameter values are more acceptable medications, highlighting the ligand molecules' drug-like activity through the inhibition of KLK 11 protein. The identification of novel therapeutic scaffolds for cancer is aided by structural data, active site details, specific ligand molecules, and ROC-AUC of inhibitors.

Drug development is expensive and time-consuming. Advanced computational methods that mine drug-disease correlations are becoming increasingly popular and are gradually replacing traditional biological experiments. However, most existing techniques rely primarily on network information. They do not fully leverage integration details and rarely capitalize on drugdisease associations. This study proposes the RWRGDR framework, which uses Graph Neural Networks (GNN) for unsupervised feature learning to identify potential drug-disease interactions. The Random Walk with Restart (RWR) algorithm serves as a complementary mechanism to enhance prediction performance. The GraphSAGE algorithm first encodes low-dimensional representations, leveraging GAT for multi-head attention to weight the significance of neighbors. The RWR algorithm then captures the global network perspective from a given target node, complementing the initial embeddings with global topological descriptors. This convex integration fuses local features and long-range dependencies, ultimately leading to superior downstream predictions. Our model, based on the Multilayer Perceptron (MLP) classifier, achieved outstanding performance, with Area Under the Curve (AUC) and Area Under the Precision-Recall Curve (AUPRC) values of 0.84 and 0.91, respectively. This performance is highly competitive, surpassing previous techniques. Case studies validate its practical applicability. Comprehensive network exploration facilitates an in-depth understanding of complex interactions and extracts meaningful insights required for optimized predictions. Despite a relatively lower AUC, our model outperformed prior methods in AUPRC, highlighting its ability to prioritize highly ranked minority positives. RWRGDR represents a potentially reliable drug repositioning strategy, as demonstrated through case studies, indicating its practical significance, particularly for emerging conditions with no recognized treatments.

Antimicrobial resistance poses a significant global health threat in the 21st century. The COVID-19 pandemic, caused by SARS-CoV-2, has further emphasized the need for novel and effective alternative therapeutics, especially from natural sources. Despite the known bioactivity of citronella oil, its potential as a broad-spectrum antimicrobial and antiviral agent remains underexplored. This study aimed to evaluate the antibacterial, antifungal, and anti-viral potential of commercially produced citronella oil using in vitro and in silico approaches. The chemical constituents of citronella oil were identified using gas chromatography-mass spectrometry (GC-MS). Antibacterial activity was assessed using the disk diffusion method, while antifungal efficacy was evaluated via the poison plate method. In silico techniques, including molecular docking, ADMET profiling, density functional theory, molecular dynamics simulation, principal component analysis, and dynamic cross-correlation mapping, were used to predict antiviral activity against the SARS-CoV-2 main protease (Mpro). GC-MS identified 30 compounds in citronella oil. Antibacterial assays demonstrated no-table inhibition against Bacillus cereus (32 μL/mL), Staphylococcus aureus (128 μL/mL), Escherichia coli (8 μL/mL), and Klebsiella pneumoniae (32 μL/mL). The oil achieved 100% mycelial growth inhibition of Fusarium oxysporum at 70 μL/mL. Molecular docking identified three compounds- CID-91697, CID-5367548, and CID-609268- with higher binding affinities than the reference drug shikonin. Computational analyses highlighted CID-91697 as the most promising candidate for SARS-CoV-2 Mpro inhibition. The results show that commercially available citronella oil has strong antifungal properties and moderate antibacterial properties. This suggests that it could be a natural broad-spectrum antimicrobial agent. Furthermore, in silico analyses indicate that CID-91697 may function as a potential inhibitor of SARS-CoV-2 Mpro, thus underscoring its significance in antiviral drug discovery. Overall, the results from both in vitro and computer-based studies make a solid case for further preclinical research. This study presents the first report on the therapeutic potential of commercially produced citronella oil from Bangladesh as a dual-purpose antimicrobial and antiviral agent. The antifungal potential of citronella oil surpassed its antibacterial efficacy against the tested strains, and CID-91697 is recommended as a promising lead compound against SARS-CoV-2 according to the outcomes of in silico studies. These findings support further experimental and clinical validation of citronella oil as a sustainable, natural alternative for combating infectious diseases.

Biomarkers have revolutionized diagnostics and therapeutics by enabling early detection, prognosis, and treatment monitoring across a range of diseases, including cancer and neurodegenerative disorders. Their role in personalized medicine underscores their importance in modern healthcare. This review consolidates findings from diverse sources, exploring the classes, mechanisms, and emerging technologies for biomarker discovery. Techniques such as next-generation sequencing, immunohistochemistry, and mass spectrometry were critically evaluated for their efficiency in biomarker validation. The study identifies various cancer biomarkers, including genetic, proteomic, and metabolomic markers, and highlights their clinical applications. It underscores significant breakthroughs in non-invasive diagnostic tools, such as exosomal proteins, miRNAs, and saliva-based markers. Challenges such as limited sample sizes, regulatory hurdles, and clinical translation bottlenecks were also discussed. Despite significant advancements, integrating biomarkers into clinical practice remains challenging due to issues of specificity, sensitivity, and cost-effectiveness. Emerging approaches such as immune checkpoint inhibitors, tumor mutational burden assessments, and chemokine profiling have shown potential in enhancing cancer immunotherapy outcomes. Biomarkers are pivotal in advancing personalized medicine by refining diagnostic and therapeutic strategies. Addressing current limitations through innovative technologies and interdisciplinary collaboration can unlock their full potential, transforming disease management and patient care.