Browse Articles

Solar-driven co-production of C2H4 and H2O2 from CO2 and H2O

Selecting suitable moss indicators for routine bioindication of roadside air pollution

Engineering LmrR protein for L-proline-based asymmetric aldol biocatalysis

Accelerometer-based assessment of occupational standing time and its association with venous disorders – results of a cross-sectional field study

Abstract Pathophysiologically, prolonged standing is associated with varicose veins (VV) or pathological venous reflux. Prior work-related epidemiological studies are inconclusive due to crude and imprecise exposure assessments. This cross-sectional field study explored the association between work-related standing time and VV or pathological reflux, assessed using wearables. Daily standing time was tracked by accelerometer and categorized: ≤2 vs. >2 but ≤4 vs. >4 h. Long-term cumulative standing exposure were estimated by combining accelerometer and survey data. Employees underwent CEAP classification and duplex ultrasound to measure lower leg reflux. The associations between standing time and other predictors on VV or reflux were analyzed by multiple logistic regression. 198 employees working ≥30 h and ≥4 days per week for ≥2y in logistics, trade, retail, manufacturing, healthcare, research, and service sectors were included ( n  = 116♀, Ø-age = 40 ± 9y). 18% ( n  = 36) stood ≤2 h, 46% ( n  = 91) >2 but ≤4 h, 36% ( n  = 71) >4 h per day. Long-term cumulative standing exposure was 10,145.0 ± 8,396.0 h. 24% ( n  = 47) were described with VV, 38% ( n  = 74) with reflux. While neither daily nor long-term cumulative standing time were significantly associated with VV or reflux, age and family history were. Our findings partly contradict previous epidemiological studies, highlighting the value of objectively measured activity patterns for future research.

Intermediates of forming transition metal dichalcogenide heterostructures revealed by machine learning simulations

Abstract Two-dimensional (2D) transition metal dichalcogenide (TMD) van der Waals heterostructures (vdWHs) hold promise for high-performance electronics, but their large-scale synthesis remains limited by size constraints and alloying contaminations. Recently, a two-step vapor deposition method was reported for growing wafer-size TMD vdWHs with minimal impurities. In this study, we develop a machine learning potential (MLP) that captures the atomic-scale dynamic growth process of bilayer MoS 2 /WS 2 vdWHs under feasible growth conditions. Our simulations uncover a crucial metastable SMMS (M = Mo or W) intermediate structure that facilitates metal atom swap and alloying. Eliminating the alloying contamination requires preventing the embedding of bare metal atoms. The results also show that the SMMS structure exhibits favorable electronic properties and emerges as a low Schottky barrier contact electrode for MoS 2 field-effect transistors (FETs).

ITGA8 suppresses proliferation and metastasis of lung adenocarcinoma through the inhibition of glycolysis

Sono-mechanical nanostructures-enabled sustained precise ultrasound brain stimulation

Multimodal fusion for equipment health status assessment based on dynamic attention mechanism

The endomicrobiome and weed invasiveness in Mediterranean ecosystems worldwide

Load-bearing and failure behavior of welded horizontal joints in prefabricated shear wall structures

Digital medicine for infectious diseases

An in-silico study to design C60 fullerene-based nanosensors for the adsorption, detection, and removal of the narcotic drug γ-hydroxybutyric acid

Abstract γHydroxybutyric acid (GHB), a depressant of the central nervous system, is commonly used illegally and in drug-facilitated crimes; therefore, it is crucial to develop reliable and fast methods for detecting GHB. This study uses DFT theory to design and evaluate the performance of electrochemical and colorimetric nanosensors based on fullerene and its forms of doping with boron and zinc for GHB detection. The calculation results (bond length, HOMO-LUMO energy gap, infrared spectra and UV-visible absorption spectra) for C 60 showed very good overlap with experimental results in other literature, indicating the validity of the computational method used in this work. Several analyses (such as electronic structure calculations, adsorption energy evaluation, charge-transfer analysis, NBO, NCI/RDG, ELF, LOL, QTAIM, conductivity, recovery time, and optical response analyses) were performed to investigate the sensor performance. After comparing these results, Boron-Doped C 60 (BC 59 ) was found to be the best candidate for electrochemical sensing of GHB based on conductivity modulation & charge transfer behavior. In contrast, pure C 60 with the largest spectral shift (in the visible range) was introduced as a suitable candidate for colorimetric measurement. Zinc-doped C 60 adsorbs GHB best (based on adsorption properties), making it suitable for GHB removal and adsorption in purification applications. Overall, this computational study makes experimental efforts more targeted by qualitatively assessing sensor performance and reducing trial and error, and provides clear guidance for future experimental validation and development of efficient GHB detection platforms.

A new highly oxygen-deficient and cubic Pr3ZrO8-δ for intermediate-temperature thermochemical production of oxygen and hydrogen

Integrating geospatial intelligence and machine learning for flood susceptibility mapping

Membrane-embedded polar residues target membrane proteins for degradation by the quality control protease FtsH

Abstract The biogenesis of membrane proteins (MPs) is inherently error-prone, and is therefore monitored by quality control mechanisms that remove faulty MPs. A key challenge for this surveillance is to recognize misfolded MPs, but how this is achieved remains poorly understood. Here we reveal how FtsH, the main MP quality control protease in Escherichia coli , specifically targets faulty MPs. By analyzing the in vivo degradation of two substrates, we show that lipid-facing polar residues trigger FtsH-mediated degradation. In folded MPs, such polar residues are usually buried in the protein core. Their exposure to the membrane can therefore signal misfolding and promote degradation. Strikingly, lipid-facing polar residues can even trigger degradation of a folded protein, and do not require the extended cytosolic regions typically needed for other FtsH substrates. Recognition depends on the FtsH transmembrane domain and on specific polar residues within it. Thus, sensing misfolding within the membrane helps maintain the integrity of the membrane proteome.

Impact of polystyrene microplastic exposure at low doses on male fertility: an experimental study in rats

Abstract Polystyrene microplastics (PS-MPs), widely used in commercial and pharmaceutical products, are emerging endocrine-disrupting pollutants with potential reproductive toxicity. This study evaluated the dose-dependent effects of PS-MPs on adult male rats by assessing semen quality, reproductive hormones, oxidative stress, mitochondrial and inflammatory markers, and testicular histology. Rats were assigned to six groups: a control group and five groups receiving PS-MPs orally (0.1, 1, 10, 20, or 40 µg/kg BW) for 45 days. PS-MP exposure reduced sperm count and motility, increased abnormal sperm, decreased testosterone, and elevated FSH and LH. Mitochondrial biogenesis/function markers (PGC-1α, UCP1, TFAM) were downregulated, while NF-κB, caspase-3, and TBARS were increased, accompanied by significant depletion of antioxidant defenses (GSH, GR, GPx, SOD, GST, CAT, TAC) and pronounced testicular histopathology. These effects were dose-dependent, and PS-MPs were detected in testicular tissue by pyrolysis-GC/MS at the doses of 10 µg/kg and higher. Collectively, the data identify mitochondrial dysfunction–driven oxidative stress and associated inflammation as a key mechanism by which PS-MPs induce spermatogenic failure, hormonal disruption, and testicular damage, highlighting their potential as potent male reproductive toxicants.

Feedback neurons based on perovskite memristor with nickel single-atom engineered reduced graphene oxide cathode

A hybrid actor–critic and BERT framework for intelligent course recommendation in IoT-aware e-learning systems

Sub-pangenome analysis reveals structural variants associated with fruit color and bacterial wilt resistance in eggplant

Abstract Eggplant ( Solanum melongena L.) is a globally important Solanaceae crop, yet trait-relevant genomic variants remain poorly characterized. Here, we perform population genomic analyses of 226 eggplant accessions sampled mainly from a major domestication center spanning Southeast Asia and South China, and find that genetic relationships closely track geographic origin. We generate chromosome-scale assemblies for 11 representative accessions using long-read sequencing and integrate six published genomes to build a pangenome resource. Using this resource, association scans identify a 12.4 Mb inversion on chromosome 10 segregating at 50.44% frequency that is strongly associated with fruit color, likely through hitchhiking with SmMYB1 . We also detect variants associated with bacterial wilt resistance, including a premature stop codon in SmCYP82D47 and copy number variations in SmEPS1 and SmRoq1 homologs. Together, our results illuminate the evolution and phenotypic impact of large structural variants and provide genomic resources for eggplant genetics and breeding.

Designing an explainable algorithm based on XGBoost and genetic algorithm for predicting hospitalization needs of COVID-19 patients

Abstract Timely identification of COVID-19 outpatients who are at risk of hospitalization is critical for preventing clinical deterioration and optimizing healthcare resources. Although machine-learning models have demonstrated high predictive accuracy, their limited interpretability often hinders clinical adoption. This study aims to develop a hybrid explainable framework that combines the predictive strength of XGBoost with clinically interpretable rule-based explanations to support decision-making in real clinical settings. A retrospective dataset of 1278 COVID-19 patients was analyzed after applying strict inclusion and exclusion criteria. Twenty-seven clinical, laboratory, and demographic variables were preprocessed using outlier detection, multiple imputation by chained equations, and stratified train–test splitting validated through a Kolmogorov–Smirnov test. XGBoost was trained and benchmarked against logistic regression, random forest, LightGBM, and a neural network. For interpretability, candidate rules were extracted from a constrained Random Forest and optimized via a genetic algorithm (GA) using accuracy–support multi-objective fitness. Clinical validation of rules was performed by ten physicians using the Content Validity Index (CVI; threshold ≥ 0.85). XGBoost achieved superior predictive performance with an AUC of 0.85, sensitivity of 73.5%, specificity of 88.7%, AUPRC of 0.72, and a Brier Score of 0.085. Baseline models demonstrated lower discrimination and calibration. Fairness evaluation indicated stable model behavior across demographic and comorbidity subgroups. Sensitivity analysis identified SpO 2 , CRP, age, D-dimer, ferritin, and lymphocyte percentage as the most influential predictors. From 400 initial rules, 80 were selected and refined, and 40 clinically valid rules were finalized through expert review. The explainability framework expands upon the classical Decision Tree Surrogate method, producing global IF–THEN rules that outperform local attribution tools such as SHAP and LIME in practical interpretability. The proposed hybrid system successfully integrates high-accuracy machine-learning predictions with clinically validated, interpretable rules, offering a transparent decision-support tool for hospitalization risk assessment in COVID-19 outpatients. Its modular design enables rapid adaptation to future infectious disease outbreaks, supporting broader clinical deployment and improved triage decision-making.