Search Results (1,143)

This study presents an integrated geomechanical modeling framework for predicting multi-scale fracture networks and their activity in the Longmaxi Formation shale reservoir, northern Luzhou region, southeastern Sichuan Basin—an area shaped by complex, multi-phase tectonic deformation that poses significant challenges for resource prospecting. The workflow begins with quantitative characterization of key mechanical parameters, including uniaxial compressive strength, Young’s modulus, Poisson’s ratio, and tensile strength, obtained from core experiments and log-based inversion. These parameters form the foundation for multi-phase finite element simulations that reconstruct paleo- and present-day stress fields associated with the Indosinian (NW–SE compression), Yanshanian (NWW–SEE compression), and Himalayan (near W–E compression) deformation phases. Optimized Mohr–Coulomb and tensile failure criteria, coupled with a multi-phase stress superposition algorithm, enable quantitative prediction of fracture density, aperture, and orientation through successive tectonic cycles. The results reveal that the Longmaxi Formation’s high brittleness and lithological heterogeneity interact with evolving stress regimes to produce fracture systems that are strongly anisotropic and phase-dependent: initial NE–SW-oriented domains established during the Indosinian phase were intensified during Yanshanian reactivation, while Himalayan uplift induced regional stress attenuation with limited new fracture formation. The cumulative stress effects yield fracture networks concentrated along NE–SW fold axes, fault zones, and intersection zones. By integrating geomechanical predictions with seismic attributes and borehole observations, the study constructs a discrete fracture network that captures both large-scale tectonic fractures and small-scale features beyond seismic resolution. Fracture activity is further assessed using friction coefficient analysis, delineating zones of high activity along fold–fault intersections and stress concentration areas. This principle-driven approach demonstrates how mechanical characterization, stress field evolution, and fracture mechanics can be combined into a unified predictive tool, offering a transferable methodology for structurally complex, multi-deformation reservoirs. Beyond its relevance to shale gas development, the framework exemplifies how advanced geomechanical modeling can enhance resource prospecting efficiency and accuracy in diverse geological settings. Full article

Cathepsin L (CatL) is a critical protease involved in cleaving the spike protein of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), facilitating viral entry into host cells. Inhibition of CatL is essential for preventing SARS-CoV-2 cell entry, making it a potential therapeutic target for drug development. Six QSAR models were established to predict the inhibitory activity (expressed as IC₅₀ values) of candidate compounds against CatL. These models were developed using statistical method heuristic methods (HMs), the evolutionary algorithm gene expression programming (GEP), and the ensemble method random forest (RF), along with the kernel-based machine learning algorithm support vector regression (SVR) configured with various kernels: radial basis function (RBF), linear-RBF hybrid (LMIX2-SVR), and linear-RBF-polynomial hybrid (LMIX3-SVR). The particle swarm optimization algorithm was applied to optimize multi-parameter SVM models, ensuring low complexity and fast convergence. The properties of novel CatL inhibitors were explored through molecular docking analysis. The LMIX3-SVR model exhibited the best performance, with an $R^2$ of 0.9676 and 0.9632 for the training set and test set and RMSE values of 0.0834 and 0.0322. Five-fold cross-validation $R_{5-fold}^2$ = 0.9043 and leave-one-out cross-validation $R_{loo}^2$ = 0.9525 demonstrated the strong prediction ability and robustness of the model, which fully proved the correctness of the five selected descriptors. Based on these results, the IC₅₀ values of 578 newly designed compounds were predicted using the HM model, and the top five candidate compounds with the best physicochemical properties were further verified by Property Explorer Applet (PEA). The LMIX3-SVR model significantly advances QSAR modeling for drug discovery, providing a robust tool for designing and screening new drug molecules. This study contributes to the identification of novel CatL inhibitors, which aids in the development of effective therapeutics for SARS-CoV-2. Full article

To unravel the link between agate geochemistry, host volcanic rocks, and ore-forming processes, this study integrated elemental correlation analysis, interaction interpretation, and interpretable machine learning (LightGBM-SHAP framework with SMOTE and 5-fold cross-validation) using 203 in-situ element datasets from 16 global deposits. The framework achieved 99.01% test accuracy and 97.4% independent prediction accuracy in discriminating host volcanic rock types. Key findings reveal divergence between statistical elemental correlations and geological interactions. Synergies reflect co-migration/co-precipitation, while antagonisms stem from source competition or precipitation inhibition, unraveling processes like stepwise crystallization. Rhyolite-hosted agates form via a “crust-derived magmatic hydrothermal fluid—medium-low salinity complexation—multi-stage precipitation” model, driven by high-silica fluids enriching Sb/Zn. Andesite-hosted agates follow a “contaminated fluid—hydrothermal alteration—precipitation window differentiation” model, controlled by crustal contamination. Basalt-hosted agates form through a “low-temperature hydrothermal fluid—basic alteration—progressive mineral decomposition” model, with meteoric water regulating Na-Zn relationships. Zn acts as a cross-lithology indicator, tracing crust-derived fluid processes in rhyolites, feldspar alteration intensity in andesites, and alteration timing in basalts. This work advances volcanic-agate genetic studies via “correlation—interaction—mineralization model” coupling, with future directions focusing on large-scale micro-area elemental analysis. Full article

Although mulberry leaf (Morus alba L., ML) and Siraitia grosvenorii (SG) individually demonstrate anti-diabetic properties, their combined efficacy against type 2 diabetes mellitus (T2DM) remains unexplored. This study systematically explored the multi-target mechanisms and synergistic potential of the MLSG combination (MLSG) for T2DM intervention. We evaluated the in vitro inhibitory activities of MLSG, ML, and SG on α-amylase and α-glucosidase, alongside antioxidant capacity assessments through DPPH/ABTS radical scavenging, reducing power, and FRAP assays. Bioactive metabolites were identified using non-targeted metabolomics, while core targets and pathways were predicted using network pharmacology and validated through molecular docking. The results reveal MLSG’s significantly enhanced inhibition of α-amylase (IC₅₀ = 14.06 mg/mL) and α-glucosidase (IC₅₀ = 0.02 mg/mL) compared to individual extracts, exhibiting 1.3–15.5-fold higher potency with synergistic effects (combination index < 1). MLSG also showed improved antioxidant capacity, outperforming SG in DPPH/ABTS⁺ scavenging and reducing power (p < 0.05), and surpassing ML in ABTS⁺ scavenging, reducing power, and FRAP values (p < 0.05). Metabolomics identified 26 MLSG-derived metabolites with anti-T2DM potential, and network analysis pinpointed 26 active components primarily targeting STAT3, AKT1, PIK3CA, EGFR, and MAPK1 to regulate T2DM pathways. Molecular docking confirmed strong binding affinities between these components and core targets. Collectively, MLSG exerts potent synergistic anti-T2DM effects through dual-enzyme inhibition, elevated antioxidant activity, and multi-target pathway regulation, providing a solid foundation for developing MLSG as functional food ingredients. Full article

Cryo-electron microscopy single-particle analysis (cryo-EM SPA) has advanced three-dimensional protein structure determination, yet resolving intrinsically disordered proteins and regions (IDPs/IDRs) remains challenging due to conformational heterogeneity. This research evaluates cryo-EM’s capacity to map dynamic regions, assesses the adaptability of disorder prediction tools, and explores optimization strategies for dynamic structure prediction. Cryo-EM SPA datasets from 2000 to 2024 were categorized into different periods, forming a database integrating sequence data and disorder indices. Established prediction tools—AlphaFold2 (pLDDT), flDPnn, and IUPred—were evaluated for transferability, while a multi-level CLTC hybrid model (combining CNN, LSTM, Transformer, and CRF architectures) was developed to link local conformational fluctuations with global sequence contexts. Analyses revealed consistent advancements in average resolution and model counts over the past decade, although mapping disordered regions remained technically demanding. Both the adapted AlphaFold pLDDT and the CLTC model demonstrated efficacy in predicting structurally variable and poorly resolved regions. A subset of the cryo-EM missing residues exhibited intermediate conformational features, suggesting classification ambiguities potentially influenced by experimental conditions. These findings systematically outline the evolving capabilities of cryo-EM in resolving dynamic regions, benchmark the adaptability of computational tools, and introduce a hybrid model to enhance prediction accuracy. This study provides a framework for addressing conformational heterogeneity, contributing to methodological advancements in structural biology. Full article

Background: Diabetes remains a major global health challenge, contributing significantly to premature mortality due to its potential progression to organ failure if not diagnosed early. Traditional diagnostic approaches are subject to human error, highlighting the need for modern computational techniques in clinical decision support systems. Although these systems have successfully integrated deep learning (DL) models, they still encounter several challenges, such as a lack of intricate pattern learning, imbalanced datasets, and poor interpretability of predictions. Methods: To address these issues, the temporal inception perceptron network (TIPNet), a novel DL model, is designed to accurately predict diabetes by capturing complex feature relationships and temporal dynamics. An adaptive synthetic oversampling strategy is utilized to reduce severe class imbalance in an extensive diabetes health indicators dataset consisting of 253,680 instances and 22 features, providing a diverse and representative sample for model evaluation. The model’s performance and generalizability are assessed using a 10-fold cross-validation technique. To enhance interpretability, explainable artificial intelligence techniques are integrated, including local interpretable model-agnostic explanations and Shapley additive explanations, providing insights into the model’s decision-making process. Results: Experimental results demonstrate that TIPNet achieves improvement scores of 3.53% in accuracy, 3.49% in F1-score, 1.14% in recall, and 5.95% in the area under the receiver operating characteristic curve. Conclusions: These findings indicate that TIPNet is a promising tool for early diabetes prediction, offering accurate and interpretable results. The integration of advanced DL modeling with oversampling strategies and explainable AI techniques positions TIPNet as a valuable resource for clinical decision support, paving the way for its future application in healthcare settings. Full article

In software development, software requirement engineering (SRE) is an essential stage that guarantees requirements are clear and unambiguous. However, incomplete inconsistency, and ambiguity in requirement documents often occur, which can cause project delay, cost escalation, or total failure. In response to these challenges, this paper introduces a machine learning method to automatically identify the risk levels of software requirements according to ensemble classification methods. The labeled textual requirement dataset was preprocessed utilizing conventional preprocessing techniques, label encoding, and oversampling with the synthetic minority oversampling technique (SMOTE) to handle class imbalance. Various ensemble and baseline models such as extra trees, random forest, bagging with decision trees, XGBoost, LightGBM, gradient boosting, decision trees, support vector machine, and multi-layer perceptron were trained and compared. Five-fold cross-validation was used to provide stable performance evaluation on accuracy, area under the ROC curve (AUC), F1-score, precision, recall, root mean square error (RMSE), and error rate. The bagging (DT) classifier achieved the best overall performance, with an accuracy of 99.55%, AUC of 0.9971 and an F1-score of 97.23%, while maintaining a low RMSE of 0.03 and error rate of 0.45%. These results demonstrate the effectiveness of ensemble-based classifiers, especially bagging (DT) classifiers, in accurately predicting high-risk software requirements. The proposed method enables early detection and mitigation of requirement risks, aiding project managers and software engineers in improving resource planning, reducing rework, and enhancing overall software quality. Full article

High-resolution 3D pavement images have become a valuable data source for automated surface distress detection and assessment. However, accurately identifying and segmenting cracks from pavement images remains challenging, due to factors such as low contrast and hair-like thinness. This study investigates key factors affecting segmentation performance and proposes a novel deep learning architecture designed to enhance segmentation robustness under these challenging conditions. The proposed model integrates a multi-resolution feature extraction stream with gated attention mechanisms to improve spatial awareness and selectively fuse information across feature levels. Our extensive experiments on a 3D pavement dataset demonstrated that the proposed method outperformed several state-of-the-art architectures, including FCN, U-Net, DeepLab, DeepCrack, and CrackFormer. Compared with U-Net, it improved F1 from 0.733 to 0.780. The gains were most pronounced on thin cracks, with F1 from 0.531 to 0.626. Our paired t-tests across folds showed the method is statistically better than U-Net and DeepCrack on Recall, IoU, Dice, and F1. These findings highlight the effectiveness of the attention-guided, multi-scale feature fusion method for robust crack segmentation using 3D pavement data. Full article

The increasing demand for energy efficiency in manufacturing has driven the need for advanced modeling techniques to optimize power consumption in machining processes. This study presents a novel approach to modeling power consumption in milling processes using machine learning, leveraging a custom-designed braking device integrated into the milling machine’s main spindle to measure friction forces with high precision. A comprehensive dataset of observations, including parameters such as speed, force, intensity, apparent power, active power, and power factor, was collected under loaded conditions. Nine machine learning models—Linear Regression, Random Forest, Support Vector Regression, Polynomial Regression, Multi-Layer Perceptron with 2 and 3 layers, K-Nearest Neighbors, Bagging, and a hybrid Probabilistic Random Forest—Multi-Layer Perceptron (PIRF–MLP)—were evaluated using 5-fold cross-validation to ensure robust performance assessment. The PIRF–MLP model achieved the highest performance, demonstrating superior accuracy in predicting utile power. The feature importance analysis revealed that force and speed significantly influence power consumption. The proposed methodology, validated on a milling machine, offers a scalable solution for real-time energy monitoring and optimization in machining, contributing to sustainable manufacturing practices. Future work will focus on expanding the dataset and testing the models across diverse machining conditions to enhance generalizability. Full article

Background/Objectives: All-inside ACL reconstruction has emerged as a minimally invasive alternative to conventional techniques, with potential advantages in graft configuration and reduced surgical trauma. This study aimed to compare the clinical outcomes of all-inside and full tibial tunnel ACL reconstruction, focusing on graft diameter, postoperative stability, and functional recovery. We hypothesized that the all-inside technique would allow for thicker grafts and yield superior postoperative knee stability and functional outcomes, with postoperative anterior laxity as a major outcome of interest. Methods: This retrospective comparative study reviewed patients who underwent ACL reconstruction between January 2020 and February 2024. From January 2020 to September 2022, a four-strand hamstring autograft with full tibial tunnel technique (FT-4) was used, while from September 2022, a six-strand hamstring autograft with the all-inside technique (AI-6) was adopted to enable thicker grafts and optimize fixation. Among a total of 103 patients, 1:1 propensity score matching (PSM) was performed based on age, sex, BMI, laterality, ALL reconstruction, meniscal lesion, and preoperative anterior laxity (SSD). Graft diameter and clinical outcomes, including knee stability and functional scores, were compared between the matched groups. Results: After PSM, two comparable groups of 29 patients each were established. Graft diameter was significantly larger in the AI-6 group (9.5 ± 0.7 mm) compared to the FT-4 group (7.8 ± 0.8 mm, p < 0.001), while other baseline characteristics remained well balanced between the groups. At the final follow-up, both groups exhibited significant improvements in anterior laxity, functional scores, and pivot shift grades (all p < 0.001). The AI-6 group demonstrated superior outcomes with a significantly higher Lysholm score (82.2 ± 6.7 vs. 75.6 ± 8.9, p = 0.002), lower WOMAC score (8.0 ± 4.3 vs. 12.9 ± 10.5, p = 0.023), and reduced anterior laxity (1.6 ± 1.1 mm vs. 2.5 ± 1.4 mm, p = 0.005) compared to the FT-4 group, whereas no significant differences were observed in the IKDC, Tegner, Korean knee score, or pivot shift test results. A simple linear regression revealed a significant negative correlation between graft diameter and postoperative anterior laxity (B = −0.398, p = 0.048). Conclusions: The present study demonstrated that the use of a six-strand hamstring graft configuration in the AI-6 technique resulted in significantly thicker grafts and was associated with reduced postoperative anterior knee laxity compared to the FT-4 technique. While interpretation of these findings requires caution in light of MCID thresholds, the AI-6 group showed favorable outcomes in anterior laxity and selected functional scores, such as the Lysholm and WOMAC. This technique may offer practical clinical value, particularly in populations prone to smaller graft diameters, as it facilitates adequate graft thickness through multifold preparation, with the all-inside approach accommodating the inherent graft shortening. Full article

Granulocyte-macrophage colony-stimulating factor (GM-CSF) is a multifunctional cytokine with therapeutic applications in oncology and neurodegenerative diseases. However, its clinical use is limited by the high cost of eukaryotic production systems. Here, we developed a cost-effective Escherichia coli-based platform for high-yield production of biologically active recombinant human GM-CSF (rhGM-CSF) using SUMO fusion technology. The engineered pET-SUMO-GM plasmid enabled expression of a 33 kDa fusion protein, accounting for 23–25% of total cellular protein, though it primarily accumulated in inclusion bodies. A multi-step purification strategy—including nickel affinity chromatography, Ulp protease cleavage, and hydrophobic chromatography—yielded >99.5% pure rhGM-CSF. In vitro functional assays demonstrated equivalent activity to the WHO international standard (ED50: 0.045 vs. 0.043 ng/mL in TF-1 cell proliferation). In vivo, the preparation significantly restored neutrophil counts (3.4-fold increase, p ≤ 0.05) in a murine cyclophosphamide-induced myelosuppression model. Our results establish a scalable, prokaryotic-based method to produce functional rhGM-CSF, overcoming solubility and folding challenges while maintaining therapeutic efficacy. This approach could facilitate broader clinical and research applications of GM-CSF, particularly in resource-limited settings. Full article

Background: Pan-drug-resistant (PDR) Acinetobacterinfections are an escalating ICU threat, demanding both patient-level triage and facility-wide forecasting. Objective: The aim of this study was to build a dual-scale AI framework that (i) predicts PDR status at infection onset and (ii) forecasts hospital-level PDR burden through 2027. Methods: We retrospectively analyzed 270 Acinetobacter infection episodes (2020–2024) with 65 predictors spanning demographics, timelines, infection type, resistance-class flags, and a 25-drug antibiogram. TabTransformer and XGBoost were trained on 2020–2023 episodes (n = 210), evaluated by stratified 5-fold CV, and externally tested on 2024 episodes (n = 60). Metrics included AUROC, AUPRC, accuracy, and recall at 90% specificity; AUROC was optimism-corrected via 0.632 + bootstrap and DeLong-tested for drift. SHAP values quantified feature impact. Weekly PDR incidence was forecast with an attention–LSTM model retrained monthly (200 weekly origins, 4-week horizon) and benchmarked against seasonal-naïve, Prophet, and SARIMA models (MAPE and RMSE). Quarterly projections (TFT-lite) extended forecasts to 2027. Results: The CV AUROC was 0.924 (optimism-corrected 0.874); an ensemble of TabTransformer + XGBoost reached 0.958. The 2024 AUROC fell to 0.586 (p < 0.001), coinciding with a PDR prevalence drop (75→38%) and three covariates with PSIs > 1.0. Isotonic recalibration improved the Brier score from 0.326 to 0.207 and yielded a net benefit equivalent to 26 unnecessary isolation-days averted per 100 ICU admissions at a 0.20 threshold. SHAP highlighted Ampicillin/Sulbactam resistance, unknown acquisition mode, and device-related infection as dominant drivers. The attention–LSTM achieved a median weekly MAE of 0.10 (IQR: 0.028–0.985) vs. 1.00 for the seasonal-naïve rule, outperforming it on 48.5% of weeks and surpassing Prophet and SARIMA (MAPE = 6.2%, RMSE = 0.032). TFT-lite projected a ≥ 25% PDR tipping point in 2025 Q1 with a sustained rise in 2027. Conclusions: The proposed framework delivers explainable patient-level PDR risk scores and competitive 4-week and multi-year incidence forecasts despite temporal drift, supporting antimicrobial stewardship and ICU capacity planning. Shrinkage and bootstrap correction were applied to address the small sample size (EPV = 2.1), which poses an overfitting risk. Continuous recalibration and multi-center validation remain priorities. Full article

Background: In healthy adults, probiotic supplementation alone does not increase Urolithin A (UroA) and rarely increases butyrate, both microbiome-derived metabolites that influence key biological functions involved in regulating gastrointestinal symptoms. Accordingly, this clinical trial evaluated key biological functions of a multi-species synbiotic with 24 probiotic strains and a polyphenol-based prebiotic using capsule-in-capsule delivery technology. Methods: We conducted a randomized, placebo-controlled trial among healthy participants (n = 32). Participants were administered a daily synbiotic (53.6 billion AFU multi-species probiotic and 400 mg Indian pomegranate extract; DS-01) or matching placebo for 91 days. Samples were obtained at baseline Day 0, and Days 7, 14, 49, and 91. Endpoints included changes in fecal microbiome composition, urinary UroA, fecal butyrate, serum CRP, and safety. Results: The synbiotic significantly increased alpha-diversity of Bifidobacterium and Lactobacillus spp. at all timepoints, including at end-of-study (Day 91, p < 0.0001) and increased native beneficial microbes. UroA production was significantly increased in the synbiotic arm at short-term (Day 7, 12-fold, p < 0.02) and long-term (Day 91, 49-fold, p < 0.001) timepoints. A higher proportion of synbiotic participants were capable of converting polyphenols into UroA (Day 91, 100% vs. 44.4%; p < 0.01). Mechanistically, synbiotic participants showed an increased abundance of Lactobacillus species involved in UroA precursor metabolism and UroA-producing Gordonibacter species. The synbiotic also significantly increased fecal butyrate levels (p < 0.03), and butyrate-producing species, in low-baseline butyrate producers, through Day 91, and was associated with reduced systemic inflammation. Conclusions: This multi-species synbiotic significantly increases diversity and abundance of key beneficial bacteria, enhances UroA production and butyrate levels, and is associated with lowered systemic inflammation. This is the first synbiotic to increase both UroA and butyrate. Full article

The emergence of multi-drug-resistant bacteria (known as superbugs) represents one of the greatest challenges for human health and modern medicine. Due to their remarkable ability to rapidly develop resistance to currently used antibiotics, new molecular targets for bacteria and substances capable of effectively combating related infections are still being sought. Lasso (known also as lariat) peptides are an unusual subclass of ribosomally synthesized and post-translationally modified peptides (RiPPs) with a structurally constrained knotted fold resembling a lasso. They are synthesized by certain groups of microorganisms as a result of complex processes involving intricate structural changes leading to the formation of the lasso structure. Reproducing these processes using known peptide synthesis methods poses a major challenge for synthetic chemistry. Lasso peptides exhibit a range of bioactivities including antibacterial activity. Due to the lasso structure, the peptides are capable of binding to new molecular targets, including atypical sides of ribosomes, in relation to currently used antibiotics. Thus, creating new mechanisms that inhibit metabolic processes leading to the death of pathogenic bacteria. This feature makes lasso peptides a potential “last chance” weapon in the fight against emerging superbugs. Full article

Early and accurate identification of brain tumors is essential for determining effective treatment strategies and improving patient outcomes. Artificial intelligence (AI) and deep learning (DL) techniques have shown promise in automating diagnostic tasks based on magnetic resonance imaging (MRI). This study evaluates the performance of four pre-trained deep convolutional neural network (CNN) architectures for the automatic multi-class classification of brain tumors into four categories: Glioma, Meningioma, Pituitary, and No Tumor. The proposed approach utilizes the publicly accessible Brain Tumor MRI Msoud dataset, consisting of 7023 images, with 5712 provided for training and 1311 for testing. To assess the impact of data availability, subsets containing 25%, 50%, 75%, and 100% of the training data were used. A stratified five-fold cross-validation technique was applied. The CNN architectures evaluated include DeiT3_base_patch16_224, Xception41, Inception_v4, and Swin_Tiny_Patch4_Window7_224, all fine-tuned using transfer learning. The training pipeline incorporated advanced preprocessing and image data augmentation techniques to enhance robustness and mitigate overfitting. Among the models tested, Swin_Tiny_Patch4_Window7_224 achieved the highest classification Accuracy of 99.24% on the test set using 75% of the training data. This model demonstrated superior generalization across all tumor classes and effectively addressed class imbalance issues. Furthermore, we deployed and benchmarked the best-performing DL model on embedded AI platforms (Jetson AGX Xavier and Orin Nano), demonstrating their capability for real-time inference and highlighting their feasibility for edge-based clinical deployment. The results highlight the strong potential of pre-trained deep CNN and transformer-based architectures in medical image analysis. The proposed approach provides a scalable and energy-efficient solution for automated brain tumor diagnosis, facilitating the integration of AI into clinical workflows. Full article