Capital University of Science & Technology

The ever-increasing need for effective therapeutic management of thyroid cancer (TC) necessitates the exploration of novel approaches for advanced drug discovery. The current study employed a robust computational pipeline integrating Machine Learning (ML) algorithms, QSAR modeling, molecular docking, molecular dynamics (MD), density functional theory (DFT), and network pharmacology to identify novel Anaplastic Lymphoma Kinase (ALK) tyrosine kinase inhibitors. An initial library of 3546 compounds from the CHEMBL4247 database was systematically filtered to 578. This screening utilized Lipinski's rule of five, aided by QSAR and detailed PaDEL descriptor analysis. An ensemble ML model, specifically a Voting Classifier (VC) combining XGBoost, LightGBM, and ExtraTrees algorithms, attained high predictive accuracy (ROC-AUC = 0.99), facilitating a strong classification and prioritization of active leads. Molecular docking experiment identified five top hit ligands (60, 63, 124, 130, 204) having docking score ranging from -9.0 to -10.4 kcal/mol and also confirmed their strong binding affinities, which surpassed the native co-crystallized ligand used as a standard. Later on, ADMET studies were executed to explore their physicochemical properties. MD simulation trajectories and MM/PBSA analyses validated the notably conformational stability and favorable binding free energies of these hit complexes. Network pharmacology was incorporated to understand tentative mechanisms of action and potential off-targets, generating a protein-protein interaction (PPI) network. DFT-based frontier molecular orbital (FMO) analysis showed Ligand124 possessed the highest electrophilicity and optimal polarizability, consistent with its marked interaction stability in MD simulations. In addition, the molecular mechanisms of hit compounds against TC were elucidated using a network pharmacology approach, which revealed a compound-target network with crucial hub targets like AKT1 and TP53. Significant correlations with cancer-related pathways, such as PI3K-Akt and MAPK signaling, as well as key involvement in kinase activity, phosphorylation, and membrane signaling complexes, were observed by the enrichment analysis of the main targets. These comprehensive results imply that investigated hit compounds probably modulate the oncogenic signaling networks, especially those controlling cell survival, proliferation, and drug resistance, in order to achieve its anti-TC therapeutic actions. These findings highlight the fundamental ability of integrating ML and computational chemistry to accelerate therapeutic development for TC.

The ominous choreography of keystrokes, the clashing of code, and the unseen maneuvers of ghostly hackers, compose a symphony of chaos, reminding us the threat of cyber war. This research aims to exploit the AI methodologies by utilizing nonlinear autoregressive with exogenous inputs (NARX) networks for numerical treatment of fractional cyber warfare model, to tackle the challenges posed by inherent complexity and stiffness associated with the system involving fractional derivatives. The fractional cyber warfare model offers a versatile approach to simulate a wide range of attack surfaces with multifaceted attack patterns. This enables the depiction of multiple attack scenarios with greater accuracy and flexibility. The synthetic data for NARX was generated by utilizing Adams numerical solver to simulate various scenarios associated with the model, providing a reliable basis for training, testing and validating of the neural networks. The generated data was segmented into three parts for the purposes of training, testing, and validation. This segmentation allows us for enabling the successful execution of the designed NARX networks for calculating the approximate solution of the fractional order cyber warfare model. Comparing the outcomes of the Gr & uuml;nwald-Letnikov (GL) backward finite difference solver with obtained solutions verify the accuracy, consistency, and efficacy of the developed NARX methodology for simulating the cyber warfare system. This demonstrates the effectiveness of the methodology as an advanced computational tool for analyzing and addressing the challenges of cyber warfare. The comprehensive simulation-based performance of the predictive neural network was further verified and endorsed by analyzing the learning curves through mean squared error based fitness, autocorrelation, crosscorrelation, histograms and regression metrics for the system.

Skin cancer is one of the most common and life-threatening diseases in the world, and successful treatments rely on timely and precise detection. Deep learning has achieved remarkable success in medical image analysis, with hybrid models combining multiple techniques showing promising results in enhancing diagnostic accuracy. However, their performance is often compromised due to the challenges inherited in dermatological datasets, such as class imbalance, noisy annotations, and variations in image quality and acquisition conditions. Because of such issues generalization of results becomes limited and it also reduces the reliability of existing models in real-world clinical applications. Therefore, in this research, we propose a robust hybrid framework for skin lesion detection that integrates hybrid deep learning models via transfer learning along with dataset refinement techniques. To improve image quality and highlight lesion features, the model incorporates advanced preprocessing steps, including Z-score standardization, lesion region isolation via segmentation masking, artifact removal, noise reduction, and contrast enhancement using the Contrast Limited Adaptive Histogram Equalization (CLAHE) algorithm. These refinements enhance model learning and improve classification performance. High-level features are extracted using pretrained Convolutional Neural Network (CNN) architectures such as VGG19, MobileNetv2, ResNet50, AlexNet, DenseNet121, and EfficientNetB0, followed by hybrid feature fusion for effective classification. Unlike existing hybrid frameworks that focus primarily on architecture depth, this study introduces a novel integration of quantified dataset refinement (via CLAHE and Dull-Razor) with a dual-branch feature fusion network (DenseNet121 + EfficientNetB0). This approach specifically targets the research gap of improving feature extractability in low contrast dermatoscopic images before classification. Our hybrid framework demonstrated high accuracy (98.89

Fiber-reinforced composites (FRCs) offer enhanced strength, stiffness, and resistance to fatigue, corrosion, and impact compared to traditional materials, leading to their widespread use across diverse engineering sectors. However, their reliance on virgin synthetic fibers raises environmental and economic concerns. While previous review studies have focused on natural fiber composites or recycling approaches for specific waste streams, an integrated assessment of agricultural, industrial, and post-consumer waste fibers within a circular valorization framework remains limited. This review addresses this gap through systematic classification of waste fiber sources and comparative evaluation of processing routes, composite fabrication methods, mechanical performance, and sustainability implications. Agricultural residues such as coir, jute, sisal, rice husk, banana fibers and wheat straw, industrial byproducts including textile waste, paper mill sludge, glass and carbon fiber waste, and post-consumer materials such as waste tires and recycled papers and plastics are examined in terms of composition, reinforcement behavior, and application potential. Reported waste fiber-reinforced composites (WFRCs) demonstrate tensile strength of 100-900 MPa and modulus of 2-30 GPa, indicating competitive performance relative to conventional composites while reducing environmental footprints. Key challenges include variability in fiber properties, moisture absorption, poor fiber-matrix bonding, and lack of standardization. To address these issues, researchers use chemical surface treatments to improve adhesion, hybrid reinforcements to balance mechanical performance, and data-driven design methods to optimize composite functionality. Overall, the review integrates recent advances in waste fiber utilization and demonstrates the growing relevance of WFRCs in load-bearing and non-load-bearing applications.

This study aims to optimize the design of latent heat thermal energy storage system (LHTESS) embedded with twin heated walls to enhance its thermal performance using the Taguchi method. The system involves a rectangular enclosure containing six internal rectangular fins, positioned on both isothermal heated walls and fully immersed in stearic acid, a phase change material (PCM). The key design parameters of the LHTESS include fin displacement (fd), fin thickness (ft), and enclosure aspect ratio (Ar) of the LHTESS configuration. The lengths of the fins are determined using an exponential mathematical function. The Taguchi optimization process utilizes an L16 orthogonal array to identify the optimum design parameters. The numerical simulations are performed using Enthalpy-Porosity model after validating against the experimental results. The optimal configuration features an aspect ratio of 0.5, a fin displacement of 20 mm, and a fin thickness of 1.2 mm. The optimal LHTESS configuration facilitated strong natural convection due to the vertically oriented enclosure and adequately spaced fins. The optimum design achieves a melting time of 45.93 min representing a 54.4% improvement over the reference case with uniform fins. Additionally, the energy storage rate increases to 120 W as compared to the reference case having 54.6 W of energy storage rate.