KRAS remains one of the most challenging oncogenic targets in lung cancer because of its shallow binding surfaces, conformational flexibility, and limited availability of druggable pockets. In a preceding QSAR-guided screening and molecular docking study, compound C9, a quinazoline-based scaffold, was identified as a potential KRAS inhibitor. However, static docking alone is insufficient to fully characterize ligand stability, conformational persistence, and energetic behavior within dynamic solvent environments. Therefore, the present study employed molecular dynamics (MD) simulations and end-point free energy calculations to further investigate the dynamic interaction profile of C9 within the KRAS binding pocket. The four top-ranked docking poses of C9 (Modes 1-4) were subjected to 200 ns explicit-solvent molecular dynamics simulations. Structural stability and conformational behavior were evaluated using root-mean-square deviation root-mean-square fluctuation (RMSF), radius of gyration (Rg), dynamic cross-correlation matrix (DCCM), principal component analysis (PCA), center-of-mass distance analysis, and residue-wise ligand contact frequency profiling. Binding energetics were further assessed using MM-GBSA and MM-PBSA calculations with energy decomposition analyses. The four binding modes exhibited distinct dynamic and energetic behaviors during the simulations. Modes one and 3 demonstrated comparatively greater structural persistence and reduced conformational instability relative to Modes 2 and 4. Mode one maintained prolonged ligand contact persistence with key switch-region residues, compact conformational sampling, and relatively stable COM distance profiles throughout most of the trajectory. PCA further revealed a comparatively confined conformational basin for Mode 1, consistent with restricted collective motions and reduced conformational dispersion. However, MM-GBSA and MM-PBSA analyses identified unusually large van der Waals energy fluctuations in Modes 1, 2, and 4, suggesting transient steric instability or nonphysical energetic excursions during portions of the simulations. In contrast, Mode 3 exhibited comparatively more stable and physically interpretable interaction energy profiles with sustained negative interaction energies and reduced fluctuation amplitudes. Across all systems, electrostatic interactions represented the dominant favorable energetic contribution to KRAS-C9 binding. The combined structural, dynamic, and energetic analyses indicate that C9 is capable of adopting dynamically persistent binding conformations within the KRAS binding pocket. Among the evaluated docking modes, Modes one and 3 exhibited the most favorable balance between structural persistence and energetic stability. These findings provide computational support for the potential of the quinazoline-based scaffold C9 as a candidate KRAS-targeting compound and establish a mechanistic framework for future structure-guided optimization and experimental validation in KRAS-driven lung cancer systems.
YTH domain-containing protein 1 (YTHDC1) is a nuclear m6A reader with well-known roles in mRNA splicing, nuclear mRNA export, and DNA damage response. Although over 44 phosphosites, including those within the YTH domain, are detected in phosphoproteomic datasets, the phosphoregulatory mechanism that governs YTHDC1 function remains unknown. To delineate the phosphoregulatory landscape of YTHDC1, we conducted an integrative analysis of large-scale human phosphoproteomic datasets. The phosphosites were ranked according to their frequency of occurrence, and most recurrent sites were considered predominant. To explore their biological significance, we combined co-phosphorylation analysis with upstream kinase prediction and protein-interaction network mapping. Three phosphosites, S308 and S146 located in the intrinsically disordered regions and S424 in the YTH domain, were identified as predominantly perturbed across datasets. Functional enrichment analysis of phosphosites in other proteins (PsOPs) co-regulated with these YTHDC1 phosphosites revealed their potential association with mRNA processing and splicing. Considering that no kinases are validated for these sites, phosphomotif-based analysis identified upstream kinases such as MAPK14, CDK7, AKT1, and PAK1. Annotation of phosphosites in these kinases co-regulated with the predominant YTHDC1 phosphosites demonstrated their association with kinase activity, reiterating their potential role as upstream kinases. The PsOPs, including these kinases, as well as many validated binary interactors associated with splicing-related functions, were enriched in the YTHDC1 phosphoregulatory network. Notably, 30 of the co-regulated PsOPs were enriched in pathways that are linked to carcinogenesis, and 4 were in DNA repair inhibition, thereby corroborating their possible role in phosphorylation-dependent signaling associated with cancer. Considering that targeted molecular biology experiments to explore the role of multiple phosphosites are challenging, our approach provides a suitable framework to infer phospho-site centric regulatory networks. Current findings suggest a putative role of YTHDC1 and its predominant phosphosites in RNA splicing and highlight its regulatory potential in tumor-associated signaling networks.
Alcohol-associated liver disease (ALD) encompasses a progressive spectrum of hepatic injury, with alcoholic hepatitis (AH) and alcoholic cirrhosis (AC) representing clinically severe and mechanistically interconnected stages. Despite significant disease burden, therapeutic strategies targeting core molecular drivers of disease progression remain limited. Identifying conserved regulatory determinants across AH and AC may provide a rational framework for mechanism-driven therapeutic intervention. S-adenosyl-L-methionine (SAMe), a key metabolic intermediate involved in methylation and redox homeostasis, has shown hepatoprotective potential; however, its direct molecular targets in ALD remain poorly characterized. An integrative in silico framework was employed to identify conserved molecular signatures and evaluate SAMe-target interactions. Publicly available transcriptomic datasets from the NCBI Gene Expression Omnibus (GEO) were analysed to identify differentially expressed genes (DEGs) in AH and AC, followed by Venn-based intersection to determine shared DEGs. Functional enrichment (GO and KEGG) and protein-protein interaction (PPI) network analyses were conducted to identify key regulatory hub genes. Selected hub proteins were subjected to molecular docking with SAMe and the stability of the resulting protein-ligand complexes was further evaluated using 100ns molecular dynamics (MD) simulations in conjunction with MM/GBSA binding free energy calculations. This integrative analysis identified 826 shared DEGs enriched in pathways associated with intracellular signalling, transcriptional regulation and extracellular matrix (ECM) organization. Network analysis revealed TGFB1, COL1A2, ESR1, PDGFRA, LUM and BCL2 as central hub genes. Molecular docking demonstrated favourable binding interactions of SAMe with these targets, with TGFB1 exhibiting the highest binding affinity (-7.0 Kcal/mol). MD simulations confirmed stable conformational dynamics of SAMe-bound complexes, particularly TGFB1, characterized by reduced structural fluctuations, increased compactness and sustained hydrogen bonding. Binding free energy analysis further supported the thermodynamic stability of these interactions, with the TGFB1-SAMe complex showing the most favourable energy profile. Collectively, these findings identify conserved molecular signatures linking AH and AC and suggest potential molecular interactions between SAMe and key regulatory proteins implicated in disease progression. By integrating transcriptomic, network and structural analyses, this study provides a systems-level framework for understanding the molecular landscape of ALD and offers a basis for future experimental studies aimed at evaluating therapeutic strategies targeting shared disease determinants.
Shigellosis is a significant threat to global public health, with an estimated 188 million cases resulting in high death rates, with the highest burden among children under 5 years old living in low and middle-income countries. The major treatment options historically have been fluoroquinolones, most commonly ciprofloxacin, and other antibiotics like azithromycin and third-generation cephalosporins. Fluoroquinolone-resistant S. flexneri poses an increased risk of causing disease globally and therefore represents a growing public health concern. Fluoroquinolone resistance in Shigella is primarily caused by chromosomal mutations in the DNA gyrase and topoisomerase IV subunits, which reduce the affinity of fluoroquinolones for their target sites, thereby reducing the effectiveness of therapy. In this study, we developed a machine-learning guided in silico approach to identify a single molecule capable of inhibiting both the DNA gyrase subunit A and topoisomerase IV, including resistant mutants, in Shigella flexneri. To accomplish this goal, machine learning models were trained on existing minimum inhibitory concentrations for various active compounds including different antibiotics against S. flexneri to screen a library of ciprofloxacin analogues derived from the PubChem database. The top-ranked compounds were then further evaluated utilising a multidisciplinary computational approach that included virtual screening, toxicity assessments, quantum simulations via density functional theory, molecular docking studies, molecular dynamic simulations, and binding free energy calculations. The multi-faceted computational workflow identified a lead compound, CA_1617 [PubChem CID: 1350066; IUPAC name: 6-Fluoro-1-(4-fluorophenyl)-7-(4-methyl-3-oxopiperazin-1-yl)-4-oxoquinoline-3-carboxylic acid] that demonstrated a high degree of dual-target affinity for both wild-type and mutant target proteins, had stable protein-ligand interactions, and had favourable pharmacokinetic characteristics relative to the reference antibiotic ciprofloxacin. These results support the potential of the lead molecule CA_1617 as a mutation-resistant inhibitor. However, additional experimental validation is required to determine its efficacy and translate the computational findings of this study into clinical use.
Hydrogen sulfide is an endogenous gaseous signalling molecule with recognized roles in vascular regulation, redox homeostasis, and inflammation. In the placenta, H2S is essential for maintaining trophoblast function and promoting healthy vascular remodelling. Impaired H2S signalling has been implicated in placental disorders characterized by oxidative stress, particularly in preeclampsia. One of the principal drivers of oxidative stress in the placenta is H/R injury, which mimics the intermittent perfusion patterns seen in early placental maldevelopment. Although the protective roles of H2S have been described in several ischemia-reperfusion models, its genome-wide transcriptional effects on trophoblasts under hypoxia/reoxygenation-induced oxidative stress remain unknown. HTR-8/SVneo trophoblasts were subjected to H/R injury induced by varying oxygen concentrations to model the fluctuating oxygen environments of early placental development, followed by treatment with an exogenous H2S donor (NaHS). A CSE inhibitor (PAG) treatment was also given. RNA sequencing was performed to characterize global gene expression changes. Differentially expressed genes were analyzed using KEGG and Gene Ontology enrichment, protein-protein interaction network mapping, and transcription factor prediction. H/R induced extensive transcriptional remodelling, with robust activation of HIF-1, PI3K-Akt, MAPK, Rap1/Ras, NF-κB, and focal adhesion pathways. H/R [2/10% O2] triggered pronounced glycolytic, hypoxia-adaptive, anti-apoptotic, and pro-invasive signatures. NaHS modulated these responses in a context-dependent manner: it attenuated early chemokine-driven inflammation, enhanced angiogenic and ECM-remodelling programs, and strengthened metabolic adaptation under a higher hypoxic burden 2/10% H/R paradigm. PAG induced a chronic inflammatory angiogenic signature, indicating endogenous H2S restrains basal inflammatory activation. Integrated regulation of proliferation, migration, apoptosis, morphogenesis, and angiogenesis was observed through biological process analysis, with major changes noticed in NaHS-treated 2/10% H/R conditions. JUN, PTGS2, MAP3K5, DUSP1, SFN, NCF2, THBS2, and GADD45A emerged as the central interconnected hub-gene module through PPI analysis. Among these, JUN and PTGS2 appeared as potential integrators of trophoblast remodelling, redox stress, and inflammatory signalling. Our study provides the first evidence of transcriptomic analysis showing that H2S alters gene networks in trophoblast cells subjected to H/R-induced oxidative stress. The results highlight coordinated regulation of metabolic, angiogenic, and inflammatory pathways, providing fundamental understanding into how H2S may influence trophoblast adaptation to stress.
The HSP90-HOP interaction orchestrates transfer of client proteins from HSP70 to HSP90, promoting their conformational maturation and stabilisation and thereby sustaining oncogenic signalling. The study explores a promising alternative for cancer chemotherapy by targeting this interface rather than the traditional ATP-binding site, which may circumvent the toxicity associated with classical HSP90 inhibitors. Despite its therapeutic relevance, the HSP90-HOP interface remains underexplored, particularly in the context of structure-guided peptidomimetic inhibitors, highlighting a critical gap in strategies to modulate proteostasis in cancer. This study sought to identify a promising peptidomimetic molecule capable of disrupting the HSP90-HOP interface. A seven-residue template peptide was engineered from a crucial segment of HOP, with hotspot residues identified through in silico mutagenesis. These residues were subsequently employed to retrieve a library of 200 peptidomimetic molecules. An in-house developed classification-based machine learning model served as the primary screening tool to identify potential HSP90-HOP interaction modulators. The shortlisted compounds were subsequently evaluated by molecular docking, binding free energy estimation, machine learning-assisted scoring, and ADMET profiling to ensure structural stability, binding reliability, and pharmacokinetic suitability. This scrutiny resulted in the selection of five lead compounds, with MMs01053537 emerging as a top candidate. The ML model achieved an accuracy of 0.9055 and an ROC-AUC of 0.9537, indicating strong predictive performance of the model. The lead molecule MMs01053537 demonstrated a favourable binding score of -85.92 kcal/mol, along with a robust stability profile during molecular dynamics simulations. To further assess the consistency of the predicted binding mode across trajectory-derived conformations, ensemble docking and MD-enhanced binding free-energy analysis, along with statistical evaluation were performed. Collectively, these findings position MMs01053537 as a potential candidate for disrupting the HSP90-HOP interaction. However, experimental validation remains essential to confirm its therapeutic potential and support further biological evaluation of the compound.
Artificial intelligence (AI) is becoming central to genomics and multi-omics, but its concepts, architectures, applications, evaluation standards, and translational requirements remain fragmented. This scoping review mapped how AI is defined and operationalized in genomic science, including machine learning, deep learning, graph-based methods, foundation models, and large language models, and synthesized their data modalities, applications, evaluation practices, interpretability strategies, and governance challenges. We conducted a PRISMA-ScR scoping review with Joanna Briggs Institute guidance. Eligible studies applied AI to genomics or closely allied omics in research, clinical, or public health contexts. MEDLINE/PubMed, Embase, and supplementary registers were searched from January 2001 to 3 September 2025 without language restrictions. Records were screened in duplicate, and standardized items were extracted, including AI concept or method family, omics modality, task, metrics, interpretability, governance, and deployment considerations. Methodological reporting and quality were appraised using design-appropriate JBI tools and summarized descriptively as a normalized 0%-100% checklist-fulfillment index. From 3,785 records, 1,040 studies were included. Publication remained sparse until 2017 and then expanded steeply, with more than 90% appearing from 2018 onward. The normalized JBI checklist-fulfillment index was modest overall (mean 35.3%, SD 20.1; range 7.5%-87.5%) and was interpreted descriptively, not as a directly comparable quality score across designs. Conceptually, the field has moved from feature-engineered statistical learning toward representation learning systems modeling nucleotide sequences, regulatory context, single-cell states, multi-omics profiles, biomedical text, and clinical-genomic knowledge. Applications concentrated on variant interpretation, regulatory genomics, multi-omics integration, single-cell analysis, pathology/radiology-genomics fusion, and genomic decision support, with increasing use of deep learning, graph models, foundation models, and LLMs. Calibration, external validation, mechanistic interpretability, ancestry-aware fairness, privacy protection, and deployment models for sensitive genomic data were unevenly reported; prospective multisite evaluations were rare. AI in genomics has scaled rapidly since 2017-2018, but translation remains constrained by heterogeneous concepts, inconsistent benchmarks, incomplete reporting, and limited governance. Priorities include biologically meaningful benchmarks; calibrated uncertainty for genomic decision support; mechanism-linked interpretability; ancestry- and site-aware validation; privacy-preserving analysis of sensitive genomic data; and human oversight for variant interpretation, precision medicine, and public health genomics. https://osf.io/uexzh.
Antibody-based therapeutics are a rapidly expanding class of treatments, with over 200 approved candidates and thousands in clinical trials. Computational pre-filtering using protein structure prediction models has the potential to reduce the cost of wet-lab screening, yet the relationship between model confidence measures and functional binding properties remains incompletely understood. Here, we evaluate whether confidence measures produced by contemporary open-source protein structure prediction models are suitable for in silico screening of nanobody-antigen interactions. We benchmark Boltz-2, Chai-1, IntFold, and AlphaFold3 using two complementary tasks: (i) ranking true nanobody-antigen binding complexes above non-binding bait pairs across 17 antigens, and (ii) detecting out-of-distribution sequences generated by alanine substitution of all complementarity-determining region residues. We further assess confidence measure sensitivity through progressive alanine mutagenesis on 13 nanobody-antigen complexes spanning the range of CDR3 lengths in our dataset and evaluate generalizability using data from a camelid immunization campaign against CD33. Boltz-2-derived confidence measures achieved the highest median performance for identifying true binders. Local confidence measures, including pLDDT and interface- or CDR-focused metrics, were most effective at detecting out-of-distribution sequences and exhibited the greatest sensitivity to mutations. No single confidence measure performed best across both tasks, and all evaluated protein structure prediction models showed limited generalization to previously unseen antigens. Our results suggest that robust in silico nanobody candidate selection should combine complementary global and local confidence measures rather than relying on a single metric. These findings provide practical guidance for integrating open-source protein structure prediction models into AI-driven nanobody discovery pipelines while highlighting the need for improved generalization across antigens.
Oral lichen planus (OLP) is a chronic inflammatory mucosal disease with a risk of malignant transformation and limited long-term therapeutic options. Paeoniflorin (PF), a natural monoterpene glycoside, exhibits multi-target anti-inflammatory and immunomodulatory properties, but its systematic mechanisms against OLP remain elusive. We employed an integrative framework combining network pharmacology, transcriptomic cross-validation, molecular docking, and molecular dynamics (MD) simulations. Public databases were mined to identify PF targets and OLP-related genes. Core targets were prioritized via protein-protein interaction (PPI) network topology and further validated using OLP tissue transcriptomic datasets (GSE52130 and GSE213349). Functional enrichment analyses were performed, followed by structural validation of PF-target binding via molecular docking and 100-ns MD simulations. Sixty-eight overlapping targets between PF and OLP were identified. PPI network analysis and transcriptomic cross-validation pinpointed eight core targets: AKT1, IL6, MMP9, STAT3, TNF, IL1B, PTGS2, and PDE4B. Mechanistically, these targets converged on the TNF, PI3K-Akt, and MAPK signaling pathways, regulating inflammatory response, cell migration, apoptosis, and protease activity at membrane microdomain and extracellular matrix interfaces. Molecular docking showed PF binding affinities comparable to or exceeding reference inhibitors (e.g., STAT3: -9.27 vs. Stattic -9.16 kcal/mol). MD simulations confirmed stable conformational binding, with the STAT3 and PDE4B complexes exhibiting the most balanced rigidity and lowest ligand RMSD (0.09-0.10 nm). This study provides a systems-level map of PF's multi-target intervention in OLP, highlighting a composite anti-inflammatory-immune reprogramming-pro-repair axis centered on core inflammatory kinases and proteases. The structural validation of key targets establishes a mechanistic rationale for PF as a promising therapeutic candidate, warranting further preclinical and clinical development for OLP management.
The ovarian cancer immunoreactive antigen domain-containing protein 1 (OCIAD1) is a mitochondrial protein implicated in mitochondrial morphology, energy metabolism, and differentiation. Although understudied, recent studies position it as a critical player in carcinogenesis and neurodegenerative disorders, making it a potentially druggable node in cellular signaling networks. However, the phosphoregulatory networks and the upstream kinases governing OCIAD1 remain unknown. A large-scale literature mining and analysis of 177 phosphoproteomic datasets with differential expression of OCIAD1 was carried out to map its phosphoregulatory network. The predominant phosphosites were determined based on localization probability, detection frequency, and differential regulation. Multipronged computational approaches were employed to gather novel candidate kinases that may target OCIAD1 phosphosites. Co-differential phosphorylation analysis was conducted with other proteins, including interactors and candidate upstream kinases, to infer functional and regulatory associations. The sites S108 and S123 emerged as predominant, together accounting for 70% of OCIAD1 phosphorylation. Co-differential phosphorylation analysis revealed associations with proteins involved in the cell cycle, DNA repair, autophagy, mitophagy, endocytosis, and apoptosis. Novel candidate kinases for OCIAD1 phosphosites were identified; notably, SRMS and YES1 emerged as potential upstream regulators of Y199. Furthermore, the phosphosites in the candidate kinases of sites, including PLK1 (T210), CDK13 (S383, S397), PRKD2 (S200), CIT (S1343), and RPS6KA3 (T577), showed strong positive co-differential regulation with OCIAD1 predominant sites, supporting their potential involvement as upstream kinases. This study presents the first systematic map of the OCIAD1 phosphoregulatory network and provides candidate upstream kinases that may contribute to its phosphorylation, which warrant further experimental validation. The strong co-differential regulation of proteins involved in autophagy, mitophagy, endocytosis, and neurodegenerative pathways, as well as of kinases that orchestrate these processes, suggests that OCIAD1 phosphoregulatory network maybe involved in mitochondrial quality control and mitochondria-associated neurodegeneration, establishing a foundation for therapeutic investigations targeting OCIAD1 signaling.
The rapid progression of Alzheimer's disease (AD) is primarily caused by compromised neurotrophin functions and decreased tropomyosin receptor kinase expression in the basal forebrain area. The two main pathogenic features of AD are cholinergic-dependent cognitive dysfunctions and amyloidogenic-induced neurodegeneration. Concurrent stimulation of major neurotrophin signalling pathways, such as tropomyosin receptor kinases receptor A and B (TrkA and TrkB), may reduce amyloid-β-mediated neurotoxicity and cholinergic denervation in the basal forebrain, improving cognitive performance and re-establishing neuronal communication. The development of new medications with dual agonist action towards TrkA and B receptors holds enormous therapeutic potential for managing the symptoms of neurodegenerative diseases. This study aims to develop novel dual TrkA/TrkB receptor agonists for the treatment of AD by enhancing neurotrophin signalling, reducing cholinergic denervation, and mitigating amyloid-β-induced neurotoxicity. An in silico drug discovery pipeline was employed, involving homology and pharmacophore modelling of amitriptyline, virtual screening of ChEMBL compounds, molecular docking, ADMET, MM/GBSA analysis, DFT calculations and molecular dynamics (MD) simulations for 100 and 300 ns to assess ligand stability and binding behaviour of the ligand-protein complexes. Six novel optimised quinoline analogues (OP-1 to OP-6) were identified as computationally predicted dual TrkA/TrkB agonists by molecular docking (-8.90 to -5.07 kcal/mol), MM/GBSA (-40.47 to -30.71 kcal/mol), ADMET and DFT analysis. Furthermore, OP-1, OP-2, and OP-3 exhibit stable binding interactions over 300 ns of MD simulations. The optimised compounds demonstrated favorable computational binding profiles, predicted pharmacokinetic properties, and stable receptor-ligand interactions, identifying them as promising candidates for further experimental validation as potential dual TrkA/TrkB modulators in Alzheimer's disease.
Antimicrobial resistance (AMR) is a major threat to global health. It reduces the effectiveness of current antibiotics and treatment for infectious diseases. The rise in AMR is mainly due to the overuse of antibiotics and the increased adaptability of harmful microorganisms. Among resistant bacteria, Methicillin-Resistant Staphylococcus aureus (MRSA) can resist an array of antibiotics. A key factor in resistance of MRSA is Penicillin-binding protein 2a (PBP2a). This protein decreases the effectiveness of β-lactam antibiotics and makes treatment more difficult. Therefore, finding new inhibitors that target PBP2a is crucial. In this study, Parmotrema perlatum, a himalayan lichen that has not been extensively studied for its antimicrobial properties, was chosen. Phytochemical research identified methyl orsellinate (MO) as a prominent secondary metabolite with antioxidant and antibacterial activities. However, initial docking analysis showed that MO had weak binding affinity for PBP2a. The molecular structure of MO was modified using a scaffold-morphing method to create a series of structural analogues. Molecular docking was conducted to assess their binding affinities and inhibitory potential. A detailed ADMET (Absorption, Distribution, Metabolism, Excretion, Toxicity) screening followed, to evaluate their pharmacokinetic and toxicity profiles. The stability of the top protein-ligand complexes using molecular dynamics (MD) simulations was assessed. MO-1 showed strong binding interactions with PBP2a and maintained stable trajectories throughout the simulation. Furthermore, MM/PBSA analysis indicated negative ΔG values, suggesting favourable binding. Overall, these results indicate that MO-derived analogue, MO-1 could be a computationally prioritised candidate for developing new therapies targeting MRSA. This study aims to open a new avenue to approach the problem of AMR with production of ethno-medicines using MO-1 to create effective therapies against MRSA and help reduce dependency on antibiotics.
Polo-like kinase 1 (PLK1) is a well-established oncogenic protein, as indicated by elevated levels of protein expression and function in a variety of tumor malignancies, and thus helps in metastasis of the disease. To target the PLK1, the study designed a cancer antigen that is capable of activating enhanced immune signaling pathways using an immunoinformatic approach. To achieve the study's objective initially, the potential B- and T-cell epitopes were screened through IEDB resources. By analyzing immunological profiles, the highly antigenic, non-toxic and non-allergenic epitopes were screened and further utilized for vaccine design. Using appropriate linkers and adapters, the multi-epitope vaccine was constructed, and subsequent structural modelling and validation were analyzed. Following that, the immunogenic potential was analyzed through molecular docking, molecular dynamic simulations, and immune simulations. In silico cloning and preliminary in vitro validation using selected CTL epitopes were performed. Four highly antigenic B-cell, five CTL, and five HTL epitopes along with linkers and 50s ribosomal adjuvants were utilized for vaccine construction. The subsequent physicochemical characterization and structural validation revealed that the PLK1 vaccine is considered highly stable. The molecular interactions with innate receptors revealed that the PLK1 vaccine had high binding affinity with the TLR4 receptor (-402.83 kcal/mol) compared to TLR2. The investigations under physiological conditions exhibited that the PLK1 vaccine with TLR complexes maintained structurally stable conformations. The elevated level of IFN-gamma and IL-2 productions was observed in the immune simulation analysis. The final recombinant length of the optimized gene expression was observed to be 6,488 bps. A decrease in PLK1 transcript levels was observed in the SKBR3 cells after treatment with specific CTL epitopes. The enhanced immune profiles of the constructed PLK1 demonstrated a favourable predicted interaction with TLR4, suggesting a potential receptor recognition that may associated with Th1-mediated immune responses. Overall, the PLK1 cancer antigen may act as a promising candidate in the context of cancer immunotherapy. Further, experimental validation in proper animal models is required to validate its immunogenic potential.
Myoglobin (Mb) is a heme-binding protein essential for oxygen storage, transport, and redox regulation, governed by a finely tuned network of electrostatic and hydrogen bonding interactions within the heme pocket. A point mutation His97Tyr leads to a hereditary disease, myoglobinopathy which disrupts heme-propionate interactions and leads to protein aggregation. Previous structural studies relied on a crystallographic model in which a Lys45Arg substitution was introduced to stabilize the heme pocket, masking the effects of the His97Tyr mutation and preventing clear interpretation of its role. To resolve this, we performed classical molecular dynamics simulations to systematically dissect the individual and combined structural effects of residues 45 and 97 by comparing the His97Tyr mutation in the native Lys45 (wild-type) and the crystallographic Arg45 residue. Our results show that residue 45 plays a dominant role in regulating long-range dynamical networks, with Arg45 producing more spatially coherent correlations than the diffuse dynamics observed in the wild-type Lys45 protein. Structural variability within the heme pocket is primarily driven by redistribution of heme-propionates, where residue 45 acts as the main determinant, while residue 97 contributes an additive effect. Although iron coordination remains preserved across all systems, the wild-type His97Tyr variant exhibits tilting of the heme group within the pocket, indicating increased conformational flexibility. Together, these findings suggest that the disease-associated mutation alters heme dynamics and highlights the critical role of sequence context in interpreting mutation-driven changes in myoglobin structure and function.
Accurate clinical triage is critical for optimizing decision-making and resource allocation during infectious disease outbreaks such as COVID-19. In this study, we present an AI-driven decision-support tool for the triage of COVID-19 patients based on respiratory microbiome profiles derived from shotgun metagenomic sequencing. We analyzed 477 shotgun respiratory metagenomes from three independent public cohorts and generated genus-level taxonomic profiles, which were integrated with minimal clinical metadata (age, sex, and antibiotic exposure) to train supervised machine-learning models, including Random Forest, Support Vector Machine, and XGBoost. Model performance was evaluated using standard classification metrics, cross-validation, and particle swarm optimization for hyperparameter tuning. Across cohorts, we observed a consistent transition from microbiomes dominated by commensal taxa to dysbiotic states enriched in opportunistic and clinically relevant genera, particularly Acinetobacter and Staphylococcus, in severe and deceased patients. Among the evaluated models, XGBoost consistently achieved the best performance, reaching up to 96.1% accuracy, 97.6% F1-score, and 98.2% ROC-AUC in individual cohorts. When trained on the integrated dataset, XGBoost maintained robust performance (95.1% accuracy, 97.2% F1-score, 94.3% ROC-AUC) and demonstrated greater stability and lower variance compared to alternative models. Feature-importance analyses identified a compact and interpretable set of recurrent microbial predictors, and reduced-feature models retained substantial discriminative power when augmented with key clinical variables. These results support the respiratory microbiome as a valuable source of information for outcome-oriented clinical triage and position microbiome-informed machine learning as a scalable and interpretable decision-support approach for managing COVID-19 and future infectious disease scenarios.
Glioma is a highly aggressive central nervous system malignancy with poor clinical outcomes, and increasing attention has focused on whether environmental endocrine-disrupting chemicals contribute to its progression. This study aimed to systematically investigate the molecular mechanisms linking bisphenol A (BPA) to glioma using an integrated network toxicology and bioinformatics strategy. BPA-related targets were collected from public databases and intersected with glioma-associated genes to identify shared targets. A protein-protein interaction network was then constructed to screen hub genes, followed by transcriptomic validation using the GSE41031 dataset. Functional enrichment analyses were performed to characterize the biological processes and signaling pathways involved, and molecular docking was used to assess the binding potential of BPA with representative core targets. A total of 696 common targets were identified between BPA and glioma. Network analysis highlighted 20 hub genes, among which STAT3, AKT1, TNF, IL6, and TP53 showed the highest topological importance. Most hub genes were significantly dysregulated in glioma stem cells relative to normal neural stem cells. Enrichment analyses indicated that the shared targets were mainly associated with oxidative stress, hypoxia, xenobiotic response, steroid hormone signaling, apoptosis, focal adhesion, and the PI3K-Akt pathway. Molecular docking suggested moderate predicted binding compatibility between BPA and the five selected hub proteins. These in silico findings suggest that BPA-related targets are potentially associated with glioma-relevant inflammatory, stress-response, and survival-related signaling networks. This study provides a systems-level framework for understanding the potential contribution of BPA to glioma biology and identifies candidate molecular targets for future mechanistic and translational investigations.
Missense variants in breast cancer remain diagnostically challenging due to their functional diversity and complex genomic contexts. Conventional laboratory assays for evaluating pathogenicity are labor-intensive, costly, and often impractical for large-scale screening, creating a pressing need for accurate, scalable, and clinically interpretable computational approaches. In this study, we present a novel deep learning framework for predicting the pathogenicity of breast cancer missense variants, integrating comprehensive preprocessing, advanced imputation, rigorous model benchmarking, and explainability. Genetic variants were curated from multiple genomic databases, annotated using the Ensembl Variant Effect Predictor (VEP), and processed with Variational Autoencoders (VAE) for missing-value imputation. Seven deep learning models, MLP, CNN, DNN, RNN, LSTM, GRU, and Transformer, were trained and evaluated across 11 performance metrics. To quantify performance stability, each model was trained across five random seeds; mean AUC ± SD across seeds is reported as the primary performance estimate, with the best-seed run used only for LIME and PMI interpretability analyses. Recursive feature elimination, permutation importance (PMI), and Local Interpretable Model-Agnostic Explanations (LIME) were employed to enhance transparency. Statistical analyses, including Z-tests, ANOVA, and calibration assessments, validated performance consistency and inter-model differences. GRU achieved the highest internal AUC (0.9956 [95% CI 0.9936-0.9972]; mean across five seeds 0.9941 ± 0.0011), with precision 0.9967 and calibration ECE 0.0095. Externally, LSTM led with AUC 0.9457, exceeding all eleven standalone predictors benchmarked on the same set. Models showed strong alignment with conservation signals such as phyloP470way and Eigen-PC scores. Notably, the pipeline provides performance metrics with 95% confidence intervals and incorporates case-level LIME visualizations for true positive, true negative, false positive, and false negative predictions, bolstering interpretability and clinical relevance. This work delivers one of the most comprehensive evaluations of deep learning in breast cancer variant classification to date. By combining high-performance sequential models with interpretable AI tools, the proposed framework provides a reproducible, transparent benchmark for variant pathogenicity prediction and a foundation for future research use and translation in cancer genomics.
Triple negative breast cancer (TNBC) is an aggressive disease characterized by a poor prognosis, high decline rates, lack of hormonal receptors for targeted therapy, limited effectiveness of existing treatments, and the emergence of chemoresistance. Sirtuin1 (SIRT1) is an epigenetic modifier and nicotinamide adenine dinucleotide (NAD+) dependent class III histone deacetylase (HDACs) protein. It promotes the modulation of several tumour suppressors and oncogenes. The evidence also suggests that inhibiting SIRT1 activity using selective SIRT1 inhibitors could restore E-cadherin expression and suppress EMT-mediated metastasis in TNBC cells. Though we have many SIRT1 protein inhibitors, they exhibited off-target effects, low isoform selectivity, and inefficiency at the clinical trial stage. In this study, we aim to identify SIRT1 isoform inhibitors by utilizing an integrated computational approach. Three stages of ML modelling were performed to find the best model from the SIRT1-based dataset. The QDA + ROS and XGBClassifier + ROS models were identified as the most robust, and they were subjected to the SHAP framework (XAI approach) to address the "black box" nature of the developed ML models. The NPASS natural compound dataset was first screened with the applicability domain of the developed models, followed by a two-step virtual screening with UniDock and AutoDock GPU in Scientiflow. Finally, the selected compounds were taken for molecular dynamics simulation, with rigorous trajectory analysis, and preliminary experimental validation was done. The compounds with NPASS IDs: NPC216682, NPC480509, NPC210910, and NPC247082 were identified as the most promising hits. Among these hits, Praziquantel (NPC480509), the only available test compound with reported anticancer properties, revealed the cytotoxic nature on MDA-MD-231 and MCF7 breast cancer cell lines, whereas non-cytotoxic on normal breast cell line (MCF10A). This study is exploratory. Exact SIRT1 selective inhibition by Praziquantel, along with other hits, may be further studies thorugh in-vitro and in vivo evaluation to understand the exact mechanism of action of these hits. The integrated in silico, and preliminary in-vitro approach proposed in this study could lead to innovative outcomes when applied in a pharmaceutical framework distinct from traditional methods.
The interaction between RNAs and RNA-binding proteins (RBPs) is fundamental for gene expression and regulation of cellular homeostasis. The growing interest in understanding protein-RNA complexes and their use in developing biotechnological solutions has highlighted the need for computational resources to enable detailed structural analysis of these interactions. Despite the availability of structural databases, there is still a significant gap in specialized databases that integrate, in a curated, systematic, and up-to-date manner, structural information on these complexes. Here, we propose RNApedia, a specialized, curated database of protein-RNA complexes accessible via an interactive and user-friendly web interface. The database brings together systematic analyses of 56,133 protein-RNA pairs. It integrates structural descriptors, including accessible and hidden surface areas, atomic contacts and interaction types, RNA classification, protein domains, RNA modifications, and, when available, affinity data. RNApedia is a scalable and integrative platform for exploring protein-RNA interactions, serving as a promising resource for structural bioinformatics and data-driven approaches, including applications in artificial intelligence. All data are freely available for download at: https://bioinfo.dcc.ufmg.br/rnapedia.
Astrocytes are essential for maintaining neuronal homeostasis, yet their stage-specific contribution to mild cognitive impairment (MCI) and Alzheimer's disease (AD) remains insufficiently understood. This study aimed to investigate astrocyte-associated transcriptional and metabolic alterations across the control-MCI-AD continuum using integrated transcriptomic and genome-scale metabolic modeling approaches. Transcriptomic profiles from hippocampal CA1 tissue (GSE28146) were analyzed across four clinical conditions (control, early MCI, advanced MCI, and AD). Astrocyte-associated expression programs were inferred using the unsupervised deconvolution algorithm CDSeq and validated through canonical marker enrichment, correlation with external reference signatures, and comparison with an independent single-nucleus RNA-seq astrocyte pseudobulk dataset. The inferred profiles were integrated into a curated human astrocyte genome-scale metabolic model to generate condition-specific models, which were analyzed using flux balance analysis (FBA) and flux variability analysis (FVA). The analyses supported stage-dependent remodeling of astrocyte-associated transcriptional and metabolic programs during disease progression. Early MCI was associated with signaling and stress-adaptation changes, whereas advanced MCI and AD showed broader disruption of synaptic support, redox homeostasis, and inflammatory-related programs. Model predictions indicated a progressive reduction in a biomass-derived maintenance proxy from control to advanced MCI, followed by a partial rebound in AD, suggesting a compensatory shift toward reactive-like astrocyte states rather than full functional recovery. Flux variability analysis revealed reduced metabolic flexibility across disease stages, particularly in glutamate-glutamine cycling, glutathione/redox metabolism, glycolysis-pyruvate metabolism, cholesterol handling, and one-carbon/folate metabolism. These findings support the view that astrocytes undergo progressive, stage-specific metabolic reprogramming during the transition from healthy aging to AD. Early alterations in redox regulation, neurotransmitter cycling, and mitochondrial function may contribute to early neuronal vulnerability. This work highlights astrocyte-centered pathways as potential targets for future experimental validation and therapeutic exploration.