Browse Articles

How DeepMind’s genome AI could help solve rare-disease mysteries

Mechanically heterogeneous hydrogel with cell-programmed network restructuring promotes tissue regeneration by mechano-epigenetic modulation

DEENet: an edge-enhanced CNN–Transformer dual-encoder model for steel surface defect detection

Abstract Steel is an indispensable material in modern industry, and its surface quality directly affects the performance and service life of products. To address problems of insufficient feature extraction capability, weak detection of small defects, and blurred target contours that lead to degraded edge information in steel surface defect detection, this paper proposes a novel edge-enhanced dual-branch steel surface defect recognition model, DEENet. First, a dual-encoder module based on CNN and Transformer is designed to extract image features and enhance the feature extraction capacity of the backbone network. Second, a Dual Channel Fusion module is introduced to perform cross-enhancement between the local features captured by the CNN and the global semantic features modeled by the Transformer, achieving feature complementarity and improving the detection accuracy for small defects. Finally, an edge enhancement module, C2f_EEM, is designed to highlight gradient differences between defective and normal regions through differential operations, thereby strengthening contour information and improving the model’s sensitivity to defect edges. Experimental results on the NEU-DET dataset show that, compared with other algorithms, DEENet achieves a superior mean Average Precision (mAP) of 81.4%, enabling more accurate detection of steel surface defects and providing valuable reference for defect inspection in real-world production scenarios.

The autonomy paradox in AI-generated content adoption: Creative-specific alternative to TAM model in China’s micro-short drama industry

In China’s booming micro-short drama industry, Artificial Intelligence Generated Content (AIGC) presents creators with an ‘autonomy paradox’: improving efficiency while sparking fears of lost control, amplified by collectivist culture that heightens tensions between AI-driven productivity and loss of autonomy. Based on a mixed-methods study of 607 micro-short drama creators, this research proposes and tests the Creative Industries Technology Acceptance Model (CITAM), which builds upon TAM foundations while adapting constructs for creative contexts to reveal key dynamics in adoption intentions. Building upon TAM theoretical foundations while introducing innovation compatibility (IC) and creative autonomy retention (CAR), CITAM is grounded in Diffusion of Innovations and Self-Determination Theory to address both rational and psychological adoption factors in creative contexts. Using a mixed methods approach with SEM in 607 surveys and 10 in-depth interviews, the results reveal that CAR positively influences AIGC adoption through IC as a mediator, while CAR negatively moderates the positive influence of IC on adoption intentions, highlighting a modest but significant psychological tension. Qualitative insights on ‘Creativity Amplification’ complement this, showing that creators perceive AIGC as an idea enhancer, not a replacement for the essential ‘human spark.’ CITAM provides a customized extension of TAM for creative industries, offering practical guidance. The findings can help developers design tools that preserve the agency of the creator and inform policy makers about balancing AIGC innovation with creator rights. These discoveries offer an initial framework for the adoption of ethical AI in the global creative economy, calling for cross-cultural validation to improve generalizability in AI-driven creative ecosystems.

Mosaic partial epidermal reprogramming remodels neighbors and niches to refine skin homeostasis and repair

Abstract Adult stem cells and their niches communicate intricately for tissue maintenance and regeneration. However, effectively coordinating these complex interactions is challenging. Here, we demonstrate that transient dedifferentiation of a fraction of epithelial stem cell progenies orchestrates beneficial changes within the entire skin’s cellular networks to favor repair. We achieved this by inducing a mosaic and reversible expression of reprogramming factors ( Oct-4 , Sox2 , Klf4 , and c-Myc ) in the mouse epidermis. This in vivo partial epidermal reprogramming not only affected the partially reprogrammed cells, but also their microenvironment, including neighboring epithelial cells and T cells, conferring widespread healing characteristics even in the absence of injury. When a wound was introduced, these collective changes accelerated re-epithelialization in both wild-type and a hyperglycemic mouse disease model. Furthermore, the effects extended to dermal healing, leading to reduced scarring and altered angiogenesis. In conclusion, our work reveals that mosaic partial reprogramming of the epidermis influences various cell types within the skin during homeostasis and repair, leading to enhanced cutaneous wound healing.

Prognostic value of the third thoracic vertebra skeletal muscle measurements in patients with digestive system malignancies: a comparative study with the third lumbar vertebra indices

Abstract Skeletal muscle mass assessment using computed tomography (CT) is crucial for evaluating nutritional status and prognosis in cancer patients. While the third lumbar vertebra (L3) level is widely accepted for this purpose, not all patients undergo abdominal CT scans. This study aimed to explore the potential of the third thoracic vertebra (T3) level as an alternative measurement site. This retrospective study included 257 patients with digestive system malignancies. Skeletal muscle area (SMA) and skeletal muscle index (SMI) were measured at both T3 and L3 levels using CT scans. Correlation analyses, linear regression models, and cox regression analyses were performed to evaluate the relationship between T3 and L3 measurements and their prognostic value. Strong correlations were observed between T3 and L3 measurements ( r  = 0.833 for SMA, r  = 0.747 for SMI). A multivariate linear regression model effectively predicted L3 SMA from T3 SMA (adjusted R ² = 0.829). Cox regression analyses revealed that lower T3 SMA and SMI were independently associated with increased mortality risk. Patients in the lowest quartile of T3 SMA had significantly higher mortality risk compared to those in the highest quartile (HR = 5.82, 95% CI: 1.86–18.16, P  = 0.002), after adjusting for confounders. Similar results were observed for T3 SMI and L3 measurements. T3 skeletal muscle measurements strongly correlate with L3 measurements and serve as independent prognostic factors in patients with digestive system malignancies. T3 measurements offer a viable alternative for assessing skeletal muscle mass and predicting prognosis when L3 measurements are unavailable.

Biochemical metabolic enhancement acting as a dominant driver in intra-leaf CO2 diffusional response to soil nitrogen supplying in Soybean

Biochemical metabolism and anatomical structure within leaf tissues have been proposed as the two principal mechanisms underpinning the rapid responsiveness of mesophyll conductance ( g m ) to environmental perturbations; nevertheless, empirical evidence distinguishing which of these factors acts as the dominant driver remains scarce. The response of intra-leaf CO 2 diffusion conductance including g m and stomatal conductance ( g sc ) to soil nitrogen (N) change in soybean was systematically quantified in leaf biochemical and structural characteristics. Our data revealed that (i) soil N made a positive effect on intra-leaf CO 2 diffusion and carbon assimilation that g m and A n (net photosynthetic rate) exhibited a significant positive response to increasing N supplying from 7.5 to 15.0 g urea m -2 while g sc showed no significant N-dependence. (ii) The enhanced intra-leaf CO 2 diffusion capacity induced by N application was principally attributable to the increase in g m .(iii) The enhancement of biochemical metabolism rather than the modifications in the leaf anatomical structure constituted the predominant mechanism by which N supplementation facilitated CO 2 diffusion and carbon assimilation in soybean. (iv)Furthermore, the improvement in water use efficiency (WUE) appeared to be more closely linked to aquaporin-mediated water relations, as supported by subsequent correlation analyses.These findings will advance our understanding of the key drivers that shape g m responsiveness to abrupt environmental variations.

Circadian fluctuation of soluble CD26 dictates the impact of the timing of cord blood transplantation on acute graft-versus-host disease

Regional disparities in breast cancer mortality in Brazil: a spatial analysis using uncorrected and adjusted data, 2000–2023

Abstract Breast cancer is the leading cause of cancer death among Brazilian women, yet mortality estimates are often underestimated due to ill-defined causes, incomplete diagnoses, underreporting, and data quality limitations. Using national mortality data from 2000 to 2023, we examined the spatial distribution of breast cancer mortality among women aged 20 years and older, comparing uncorrected and adjusted estimates. Adjustments were applied to correct ill-defined causes, incomplete diagnoses, underreporting, and other data quality limitations using methods developed by the World Health Organization and the Brazilian Institute of Geography and Statistics (IBGE). Age-standardized mortality rates were calculated for five time periods using the World Health Organization (WHO) standard population, and spatial patterns were analyzed using choropleth maps, Moran I, and Local Indicators of Spatial Association (LISA). A total of 328,319 breast cancer deaths were reported, increasing to 385,068 (+ 17.3%) after adjustment. Corrections had the greatest effect in the North and Northeast in 2000–2004 (up to + 69.9%), but declined substantially over time. Mortality remained consistently higher in wealthier regions, while adjustments revealed underestimation in historically underserved areas. These findings reveal enduring geographic inequalities in breast cancer mortality and underscore the urgent need for targeted interventions and improved surveillance systems.

DMS-YOLO: Small target detection algorithm based on YOLOv11

To address the challenges in vehicle detection from unmanned aerial vehicle (UAV) overhead images, such as small object size, low resolution, complex background, and scale variation, this paper proposes several targeted improvements to the YOLOv11n model. Firstly, inspired by the Cross Stage Partial Networks (CSPNet), a Dynamic Multi-Scale Edge Enhancement Network (DMS-EdgeNet) is designed to improve robustness to local target features. This module applies multi-scale pooling to extract edge features at various scales and dynamically fuses them through adaptive weighting. Secondly, the DynaScale Aggregation Network (DySAN) module is introduced into the neck network, and a multi-level jump connections structure is adopted to fuse low-level and high-level boundary semantics, thereby improving the detection capability of fuzzy boundary targets and improving target positioning accuracy under complex imaging conditions. Finally, a P2 small target layer is added to further improve the accuracy of small target detection. Based on these innovations, we propose a new architecture named Dynamic Multi-scale and Channel-scaled YOLO (DMS-YOLO), significantly improve the model’s ability to perceive small targets. Experimental results show that DMS-YOLO improves mAP50 and mAP50-95 by 7.0% and 2.9%, respectively, on the Aerial Traffic Images dataset, and by 5.1% and 3.1% on the VisDrone-DET2019 dataset, demonstrating superior performance over the YOLOv11n baseline.

TMEM63 proteins act as mechanically activated cholesterol modulated lipid scramblases contributing to membrane mechano-resilience

Brief mindfulness meditation increases risk-taking behavior

First report of extended-spectrum beta lactamase (ESBL) and carbapenemase-producing MDR Klebsiella pneumoniae from Fuchka

Extended-spectrum beta-lactamase (ESBL) and Carbapenemase-producing Gram-negative bacteria, particularly Klebsiella pneumoniae (ESBL-KP and CP-KP), in the food supply chain, poses a significant public health threat. Ready to eat (RTE) street foods, especially fuchka, a highly popular snack in Bangladesh, India and Southeast Asia, represents a critical breach in food safety. This study investigated the multidrug resistance (MDR) patterns of K. pneumoniae isolated from fuchka with an emphasis on phenotypic and genotypic detection of ESBL-KP and CP-KP. A total of 60 samples were collected from 15 fuchka selling points. K. pneumoniae was isolated and identified via staining, cultural methods, PCR and MALDI-ToF-MS biotyper. Antibiotic susceptibility test (AST) was accomplished using disk diffusion method, phenotypic ESBL producers was detected by combined disk test (CDT) and PCR was used to detect resistance determinant. Thirty K. pneumoniae isolates were confirmed by PCR and MALDI-ToF-MS. All showed resistance to amoxicillin and cefuroxime (100%), while most were sensitive to cefepime (96.7%), norfloxacin (96.7%), imipenem (93.3%) and meropenem (83.3%). All the isolates were MDR, with multiple-antibiotic resistance (MAR) index values ranged from 0.28 to 0.64. CDT confirmed 28 ESBL producers. Among ESBL-producing genes bla TEM, was most prevalent (64%), followed by bla SHV, bla OXA-1, bla CTX-M1, bla CTX-M3. Among carbapenem-resistant genes, bla BIC was most common (46%), while bla VIM was absent. Moreover, other resistance determinants ( aad A1, aac 3IV was most prevalent (40%) but aad A1, qnr A were not detected. The presence of multidrug-resistant, ESBL- and carbapenemase-producing K. pneumoniae in fuchka represents a critical threat to public health.

A heterogeneous population code at the first synapse of vision

Abstract Vision begins when photoreceptors convert fluctuations in light intensity into temporal patterns of glutamate release that drive the retinal network. The input-output relation at this first stage has not been studied in vivo so it is not known how it operates across a photoreceptor population. Using glutamate imaging in zebrafish, we find that individual type 1 cones (PR1; ancestral red cones), which dominate daylight vision in non-avian vertebrates, encode visual stimuli with high reliability and time-precision but routinely vary in sensitivity to luminance, contrast and frequency across the population. Variations in input-output relations are generated by feedback from the horizontal cell network that effectively decorrelate feature representation. A model capturing how zebrafish sample their visual environment indicates that heterogenous cone outputs expand the dynamic range of the retina to improve the coding of natural scenes. Moreover, we find that different kinetic release components are used to encode distinct stimulus features in parallel: sustained release linearly encodes low amplitude light and dark contrasts, but transient release encodes large amplitude dark contrasts. This study reveals an unexpected degree of functional heterogeneity within a population of cones and illustrates how separation of different visual features begins in the first synapse in vision.

Preparation and characterization of holmium doped ZIF-8 nanocrystals for white light emitting phosphors

Abstract We report on the generation of white light from Ho-doped ZIF-8 nanocrystals synthesized by a simple wet chemical technique at room temperature. The influence of Holmium (Ho) concentration and heat treatment on both structural and optical responses has been investigated. The crystal structure and morphology of as-synthesized (As) and post-thermal annealed (TA) nanocrystals have been examined using X-ray diffraction (XRD) and Transmission Electron Microscope (TEM).  In addition, the thermal stability and bond structure have been studied via Thermogravimetric analysis (TGA) and Fourier Transform Infrared Spectroscopy (FTIR), respectively. Moreover, the photoluminescence (PL) measurements were performed using two excitation wavelengths: 445 nm and 355 nm. The XRD and TEM evidenced the sodalite nanocrystalline nature of the samples. The TGA and FTIR analyses demonstrated the degradation behavior of ZIF-8 due to the incorporation of Ho ions and thermal calcination. The PL results showed intense luminescence response, in particular, from the heat-treated samples. Furthermore, the PL exhibited a naked-eye observed white light emission. This was attributed to the structural modifications induced by the dopant and thermal treatment. Our findings could be used as a guideline towards the potential application of lanthanide-doped MOFs (ZIF-8) in light-emitting purposes.

Intergenerational transmission of violence in Bangladesh: Mediated through maternal attitudes towards intimate partner violence, disciplinary beliefs, and life satisfaction

Introduction Child discipline, while intended to instill appropriate behavior, often manifests as violent practices in low- and middle-income countries, including Bangladesh. Maternal exposure to violence, attitude towards intimate partner violence (IPV), and disciplinary beliefs serve as key determinants of physical disciplinary practices. These dynamics illustrate how exposure to violence in adulthood can shape parenting behaviors, highlighting the urgency of addressing cultural attitudes that sustain harsh physical discipline. Materials and methods This study analyzed nationally representative cross-sectional data from the 2019 Bangladesh Multiple Indicator Cluster Survey (MICS) which included 30044 mother-child (children aged between 2 and 14 years) pairs. Physical disciplinary practice is analyzed as an ordered outcome, considering maternal experience of physical violence as the primary exposure along with their attitudes toward IPV and disciplinary beliefs as mediators. This study used ordinal logistic regression within a structural equation modeling framework and bootstrapping technique to analyze indirect associations, providing robust inference that accounts for sampling variability and accommodates binary mediators. Results Mothers exposed to violence had significantly higher odds of physically disciplining their children (odds ratio, OR=1.77 and 95% confidence interval, CI=[1.60, 1.95]). Three mediators significantly increased the odds of adopting harsh physical disciplinary practice by 2% through maternal positive attitudes toward IPV, by 51% through their disciplinary beliefs, and by 6% through their overall life satisfaction. The total association indicated that maternal exposure to violence nearly tripled the odds (OR = 2.89 and 95% CI= [2.52, 3.31]) of physical disciplinary practices. Conclusion This study suggested that supportive environment for children can be fostered by reducing violence against women, promoting mothers’ life satisfaction, and reshaping women’s perceptions of spousal abuse and disciplinary beliefs.

The biological role of local and global fMRI BOLD signal variability in multiscale human brain organization

Abstract Variability drives the organization and behavior of complex systems, including the human brain. Understanding the variability of brain signals is thus necessary to broaden our window into brain function and behavior. Few empirical investigations of macroscale brain signal variability have been undertaken, given the difficulty in separating biological sources of variance from artefactual noise. Here, we characterize the temporal variability of the most predominant macroscale brain signal, the fMRI BOLD signal, and systematically investigate its statistical, topographical, and neurobiological properties. We contrast fMRI acquisition protocols, and integrate across histology, microstructure, transcriptomics, neurotransmitter receptor and metabolic data, fMRI static connectivity, and empirical and simulated magnetoencephalography data. We show that BOLD signal variability represents a spatially heterogeneous, central property of multi-scale multi-modal brain organization, distinct from noise. Our work establishes the biological relevance of BOLD signal variability and provides a lens on brain stochasticity across spatial and temporal scales.

Serial femtosecond crystallography data processing at the global science data hub center at KISTI

Rare event detection by progressive clustering undersampling

Capturing rare events in severely imbalanced datasets is challenging, as the learning and optimization processes are often biased toward the majority class. To address this issue, this study explores various resampling techniques and introduces a novel method called Progressive Clustering Undersampling (PCU). This technique removes negative instances that are distant from positive ones. PCU was compared with eight common undersampling and two oversampling techniques, consistently outperforming them on highly imbalanced and noisy datasets. The workflow demonstrates that rare anomalies can be effectively predicted using unsupervised methods based on frequency-driven decision boundaries. Progressive clustering ultimately identifies clusters with the highest concentration of positive instances. These delineated clusters are then saved by supervised models and used in the preparatory phase before prediction. The proposed method produces two outputs: one optimized for a high F1-score and the other for high precision. Overall, this approach presents a promising solution for identifying rare anomalies in complex, imbalanced data environments.

Phased-assembly-driven pangenome graphs for structural variant genotyping and complex trait mapping in dairy cattle

Abstract Structural variants are an underexplored source of genetic diversity. As part of the FarmGTEx Project, here we report a Holstein breed-specific pangenome graph (H20D) using Minigraph-Cactus and 40 phased haploid assemblies from 20 cows. H20D outperforms both assembly- and read-based long-read callers, and far exceeds short-read approaches, identifying over 10,000 additional structural variants per sample. It also significantly improves structural variant detection and genotyping relative to graphs built across breeds or from fewer/unphased assemblies, with particular advantages in complex regions. Using H20D, we genotype variants in 173 cattle and performed a GWAS, where a larger fraction of structural variants than SNPs reach genome-wide significance, implicating them as potential causal variants. Together, these results demonstrate the power of phased, within-breed pangenome graphs for accurate SV genotyping and trait mapping in dairy cattle.