Browse Articles
Discover research articles across all indexed journals
Synergistic anticancer effects of peanut hairy root culture crude extract in combination with gemcitabine against cholangiocarcinoma
Water-based UV-ozone activation enables aggregation-free processing of MFI nanosheets for membrane fabrication
Identification of suitable methods for static pile load capacity and recommendation for BNBC 2020
Abstract This research compares theoretical methods with static pile load test results to identify the most effective approaches for determining pile capacity across various soil and pile types, while also proposing modifications to the Bangladesh National Building Code (BNBC) 2020 guidelines. The study involved broad data collection from pile load test reports and subsoil investigations across six projects, focusing precast and cast-in-situ piles. Theoretical analyses were performed using i) bearing capacity equations as per BNBC 2020 (α-method and β-method), ii) Standard Penetration Test (SPT)-based BNBC 2020 method, and iii) SPT-based other methods from literature including Meyerhof, and Shioi Fukui equations. Following ASTM D1143 guidelines, eight evaluation methods were used to interpret load-settlement data from static load test. Statistical analyses involving Mean Absolute Percentage Error (MAPE), Bias Factor (λ), and Coefficient of Variation (COV) were conducted to evaluate predictive accuracy. Findings indicate that for clay soil, the SPT-based equations of Shioi & Fukui for skin friction and Meyerhof for end bearing can be considered as suitable alternatives for future BNBC revisions, showing lower variation with actual capacity. For sandy, silty, and layered soil devoid of clay, the SPT-based equations from BNBC 2020 are deemed appropriate.
Structural polymorphism of ex-vivo ALECT2 amyloid fibrils revealed by cryo-EM
Abstract ALECT2 amyloidosis is a rare systemic disease characterized by the pathological deposition of leukocyte cell-derived chemotaxin-2 (LECT2) as amyloid fibrils, primarily affecting the kidneys and liver. The molecular mechanisms underlying LECT2 aggregation remain poorly defined, hindering diagnostic and therapeutic development. Here, we present cryo-electron microscopy structures of ex-vivo ALECT2 fibrils extracted from a patient’s kidney. We identified three fibril polymorphs: a predominant single-protofilament morphology and two minor double-protofilament morphologies. The dominant single-protofilament morphology comprises the full-length 133-residue LECT2 protein and retains all three native disulfide bonds. Low-resolution reconstructions of double-protofilament morphologies suggest they adopt a similar fold to the single protofilament morphology, but form paired assemblies with different inter-filament interfaces. Mass spectrometry also reveals acetylation within the fibrils. These findings offer critical insights into the structural basis of ALECT2 amyloid formation and identify molecular features that could inform future diagnostic and therapeutic approaches.
Structural response of RC beam-column joints reinforced with smart alloys and polyvinyl alcohol fibers
Skeletal editing of tetrahydrofurans to pyrrolidines through O-to-N single atom swap
Potential role of leptin in colorectal cancer liver metastasis involving lipid metabolic reprogramming and immunosuppressive macrophage polarization
Metabolic adaptation drives self-organization during skin organoid morphogenesis
New predicted dual CDK-2/CDK-1 inhibitors from Aspergillus unguis isolate SP51-EGY with relative selectivity for colorectal cancer cells: a computational and experimental approach
Abstract Colorectal cancer is one of the deadliest cancers in the world. The main problem with cancer treatments is that they need to protect healthy cells. Cyclin-dependent kinases (CDKs), especially CDK-2 and CDK-1, are essential for regulating the cell cycle, cell growth, and tumor genesis. Their dysregulation is frequently detected in colorectal cancer, resulting in uncontrolled cell division and resistance to apoptosis, making them attractive targets for anticancer therapy. The fungal " Aspergillus unguis isolate SP51-EGY" “Sh cell” (shake mycelia) extract had the most significant cytotoxic action against HCT116 cancer cells (IC 50 = 3.49 µg/mL) and a selectivity index (SI = 23.33), indicating relative cytotoxicity against cancer cells. Computational modeling and cellular phenotypic data suggest a potential mechanism of dual CDK2/1 inhibition, forming a testable hypothesis that requires direct enzymatic validation. Compounds corresponding to peaks [7] and [14] have been tentatively assigned as promising CDK inhibitors based on molecular docking studies; due to their: (1) strong binding affinities (-13.23 to -46.05 kcal/mol) confirmed by stable molecular dynamics (RMSD < 2.0 Å); (2) cell cycle arrest mediated by dual CDK2/1 inhibition; and (3) favorable drug-like properties, such as molecular weight < 500 Da, good intestinal absorption, and minimal toxicity. Critical interactions with CDK2 (Tyr16, Phe81) and CDK1 (Met88) active site residues were identified by structural analysis. These interactions competed with ATP to disrupt the CDK2 cyclin A and CDK1/cyclin A/B complexes, which resulted in both G1/S and G2/M cell cycle arrests. These integrated computational and cellular findings suggest that the fungal-derived compounds are promising candidates for novel dual CDK2/CDK1 inhibitors, warranting further experimental validation through biochemical kinase assays.
Hybrid physics-machine learning models for quantitative electron diffraction refinements
Abstract High accuracy electron microscopy simulations required for quantitative crystal structure refinements face a fundamental challenge: while physical interactions are well-described theoretically, real-world experimental effects are challenging to model analytically. To address this gap, we present a hybrid physics-machine learning framework that integrates differentiable physical simulations with neural networks. By leveraging automatic differentiation throughout the simulation pipeline, our method enables gradient-based joint optimization of physical parameters and neural network components representing experimental variables, offering superior scalability compared to traditional second-order methods. We demonstrate this framework through application to three-dimensional electron diffraction (3D-ED) structure refinement, where our approach learns complex thickness distributions directly from diffraction data rather than relying on simplified geometric models. This method achieves state-of-the-art refinement performance across synthetic and experimental datasets, recovering atomic positions, thermal displacements, and thickness profiles with high fidelity. The modular architecture proposed can naturally be extended to accommodate additional physical phenomena and extended to other electron microscopy techniques. This establishes differentiable hybrid modeling as a powerful paradigm for quantitative electron microscopy, where experimental complexities have historically limited analysis.
YOLO-AMI: enhancing online quality monitoring in 3D printing with composite loss and parameter-free attention
Abstract Additive manufacturing (AM) is transforming industrial production; however, inevitable defects—such as spaghetti-like collapses, surface blemishes (“zits”), and stringing—substantially degrade product quality and mechanical performance. To overcome limitations of traditional inspection methods—often inefficient, subjective, or reliant on costly offline equipment—a high-precision, real-time defect-detection model, YOLO-AMI, is proposed, based on the YOLOv10 architecture. The neck network was reconstructed using the Asymptotic Feature Pyramid Network to enhance multi-scale feature fusion and suppress background noise. In addition, a parameter-free attention mechanism was integrated to adaptively emphasize critical features without increasing computational complexity. To improve detection of small defects, a composite loss function combining Normalized Wasserstein Distance and Intersection over Union was adopted. Experimental evaluation on a dataset of 6,000 AM images shows that YOLO-AMI attains a mean average precision (mAP@0.5) of 85.5%, precision of 87.1%, and recall of 83.2%, outperforming state-of-the-art models such as YOLOv8, YOLOv11, and RT-DETR-L. With an inference speed of 105.6 frames per second and a compact model size of 8.6 million parameters, the proposed approach achieves a favorable balance between accuracy and efficiency, providing a robust solution for intelligent online quality monitoring in Industry 4.0.
Co-delivering macrophage engager mRNA and PD-L1 antibody via tumor-responsive nanoparticles for glioblastoma immunotherapy
Abstract Bispecific immune cell engagers, particularly bispecific T-cell engagers, show limited efficacy in solid tumors such as glioblastoma (GBM) due to systemic toxicities, poor T cell infiltration, and restricted drug penetration. We develop PL@mBiME, a multifunctional lipid nanoparticle (LNP) platform that enables brain tumor–targeted delivery and sustained in vivo expression of mRNA encoding a bispecific macrophage engager (BiME). The BiME simultaneously targets ErbB2 on glioma cells and CD206 on M2 macrophages, reprogramming macrophages toward pro-inflammatory M1 phenotype while promoting macrophage–tumor cell bridging, enhancing tumor cell phagocytosis and antigen presentation. PL@mBiME incorporates pH-responsive charge reversal to improve tumor accumulation and lysosomal escape as well as glutathione-triggered release of surface-conjugated PD-L1 antibody to amplify anti-tumor immunity. Across multiple GBM models, this coordinated activation of innate and adaptive immunity induces tumor regression, prolongs survival, and generates durable immune memory without significant toxicity.
Preparation and evaluation of [64Cu]Cu-labeled bisphosphonate amide of DOTA using reactor-produced 64Cu as an affordable skeletal PET imaging agent
Learning lessons from over-crediting to ensure additionality in forest carbon credits
Abstract Independent evaluations have shown substantial over-issuance of REDD+ (Reducing Emissions from Deforestation and Degradation) credits traded on the voluntary carbon market. We synthesise these evaluations to estimate the additional forest conservation achieved by first-generation REDD+ projects and to identify mechanisms underlying over-crediting. We combine six independent ex post evaluations of avoided deforestation covering 44 REDD+ projects. These evaluations show that most projects reduced deforestation, but that they claimed an aggregate of 10.7 times more avoided deforestation than is justified by independent estimates. This discrepancy is not driven by the choice of forest cover data, but by selection bias in projects’ control areas and modelling approaches. Although recent initiatives that transfer assessment to unconflicted parties and restrict methodological flexibility are critical, they are insufficient. Ex post certification against credible counterfactuals is necessary if carbon markets are to represent causal reductions in deforestation.
Estimating the generation time for SARS-CoV-2 transmission using United States household data, December 2021–May 2023
Abstract Generation time, representing the interval between infection events in primary and secondary cases, is important for understanding disease transmission dynamics including predicting the effective reproduction number (Rt), which informs public health decisions. While previous estimates of SARS-CoV-2 generation times have been reported for early Omicron variants, there is a lack of data for subsequent sub-variants, such as XBB. We estimated SARS-CoV-2 generation times using data from the Respiratory Virus Transmission Network – Sentinel (RVTN-S) household transmission study conducted across seven U.S. sites from December 2021 to May 2023. The study spanned three Omicron sub-periods dominated by the sub-variants BA.1/2, BA.4/5, and XBB. We employed a Susceptible-Exposed-Infectious-Recovered (SEIR) model with a Bayesian data augmentation method that imputes unobserved infection times of cases to estimate the generation time. The estimated mean generation time for the overall Omicron period was 3.5 days (95% credible interval, CrI: 3.3–3.7). During the sub-periods, the estimated mean generation times were 3.8 days (95% CrI: 3.4–4.2) for BA.1/2, 3.5 days (95% CrI: 3.3–3.8) for BA.4/5, and 3.5 days (95% CrI: 3.1–3.9) for XBB. Our study provides estimates of generation times for the Omicron variant, including the sub-variants BA.1/2, BA.4/5, and XBB. These up-to-date estimates specifically address the gap in knowledge regarding these sub-variants and are consistent with earlier studies. They enhance our understanding of SARS-CoV-2 transmission dynamics by aiding in the prediction of Rt, offering insights for improving COVID-19 modeling and public health strategies.