Browse Articles

Dynamic monitoring of railway bridges via coupling digital twins with deep reinforcement learning

Genome-wide DNA methylation profiling during metabolic dysfunction-associated steatohepatitis-related hepatocarcinogenesis in patients in Japan and the United States

A novel hybrid clustering approach for robust ramp event characterization

Abstract Large power fluctuations in a brief amount of time, or ramp events, are an increasing concern for grid operators due to the rise in renewable energy generation and the unreliable hour-ahead predictions. To balance these ramp events, grid operators need to be aware of their anticipated occurrence intervals and range. Prior studies used binary ramp event categorization, whereas other studies employed non-causative classification techniques. Existing clustering methods, Z-score and k-means, have strengths but distinct limitations. To address these, this paper introduces the ZK-means hybrid approach, integrating Z-score normalization with k-means partitioning, forming a centroid-based clustering algorithm to enhance adaptability, noise resistance, and interpretability in ramp classification. Its need arises from the growing demand for accurate and efficient ramp analysis to support reliable grid operation and forecasting. Two comparison phases for the ZK-means approach were conducted: First, it was evaluated against its constituent methods to assess the benefits of their combination; second, it was compared to the density-based spatial clustering of applications with noise (DBSCAN) algorithm to verify its robustness and general applicability. Although DBSCAN can capture local variations in data, it produced inconsistent cluster numbers and required frequent parameter tuning across the ten years. In contrast, ZK-means achieved more stable clustering patterns and lower within-cluster variance, demonstrating superior reliability for long-term ramp event characterization. The new categorization method is applied to a real case study, and the results reveal that the new hybrid method offers significant improvements in the quality, robustness, and interpretability of the clustering process and its resulting cluster characteristics, as it combines the stability of normalization with the scalability of k-means, offering a robust and practical solution for large-scale, high-dimensional clustering. While this new method does entail a slight increase in time-speed, computational complexity, and energy consumption compared to its constituent methods, it remains faster than DBSCAN and the enhanced insights it provides offer critical advantages for effective grid management.

An interpretable machine learning model for biomarker identification and diagnostic nomogram development in hypo-productive thrombocytopenia

Abstract Thrombocytopenia is a common hematological disorder with diverse etiologies. This study aimed to identify potential biomarkers and to develop an interpretable diagnostic nomogram for the early recognition of hypo-productive thrombocytopenia. This retrospective study included 185 patients with thrombocytopenia who were admitted to the Department of Hematology at the Affiliated Suzhou Hospital of Nanjing Medical University between 2020 and 2025. Patients were categorized into hypo-productive thrombocytopenia ( n  = 114) and hyper-destructive thrombocytopenia ( n  = 71) according to etiology. Feature selection was performed using least absolute shrinkage and selection operator (LASSO) regression, support vector machine recursive feature elimination (SVM-RFE), and Boruta algorithms, followed by univariate and multivariate logistic regression analyses. Six machine learning (ML) models were developed and compared based on residual analysis and receiver operating characteristic (ROC) curve. The optimal model was interpreted using SHapley Additive exPlanations (SHAP) analysis. The diagnostic nomogram was evaluated by the ROC, calibration curves, and decision curve analysis (DCA). Finally, age, platelet-to-lymphocyte ratio (PLR), neutrophil-to-lymphocyte ratio (NLR), and mean platelet volume (MPV) were identified as key predictors. The generalized linear model(GLM)–based nomogram demonstrated strong discriminative ability (area under the curve [AUC] = 0.945, 95% confidence interval [CI] 0.900–0.976) with excellent calibration. DCA demonstrated a higher net benefit for the nomogram than individual predictors. The interpretable machine learning derived nomogram may serve as a practical and non-invasive tool to assist in the early identification of hypo-productive thrombocytopenia.

The genetic architecture of cortical similarity networks

Abstract The genetic architecture of human brain networks is central to understanding cortical organisation and evolution, the causal links between brain structure and function, and the pathogenesis of neuropsychiatric disorders. Using N > 48,000 subjects, we investigated common genetic effects on Morphometric INverse Divergence (MIND), a heritable, multi-modal structural MRI metric of inter-areal similarity and connectivity. Genetic correlations between MIND network edges were largely reducible to two gradients, each aligned with distance from one of the two phylogenetically primitive areas (paleocortex and archicortex) predicted by the dual origin theory of cortical evolution. MIND was more heritable than comparable measures of functional (f)MRI connectivity, and the paleocortically-aligned MIND gradient was genetically correlated with, and causally predictive of, fMRI connectivity. Finally, we identified genetic overlaps between MIND gradients and neuropsychiatric and biomedical traits. These results provide fresh insight into the dual origins of the cortex and their implications for brain function and health.

An autonomous lab for data-driven homogeneous catalysis

Cooperative chelation for high-performance Perovskite light-emitting diodes

Trigger-day hCG effects on DNA methylation and neurodevelopment in ART offspring

Structure of the NAT10 acetyltransferase and mechanism of tRNA acetylation

Abstract NAT10 is the sole eukaryotic acetyltransferase that catalyzes N4-acetylcytidine (ac 4 C) modification of RNA. While dysregulation of NAT10 is associated with cancer and premature aging syndromes, the requirement for its acetyltransferase activity and how NAT10 coordinates catalysis and RNA binding remain poorly understood. Here, we report single particle cryo-electron microscopy structures of eukaryotic ( Chaetomium thermophilum ) NAT10 in complex with a designer cytidine-CoA cofactor-based ligand in the presence and absence of ADP. NAT10 forms a symmetrical heart-shaped dimer where a Gcn5-related N-acetyltransferase (GNAT) domain with an atypically opened active site is flanked by conserved helicase and RNA-binding domains. Biochemical reconstitution of NAT10 in the presence of the adapter protein THUMPD1 reveals that tRNA acetylation is enabled by two conserved active site residues (His548 and Tyr549 in Ct NAT10) and two basic patches: one proximal and one distal from the active site, and suggests that binding orientation rather than affinity drives catalysis. Finally, we harness structure-guided mutations in cellular studies to demonstrate the necessity for NAT10 catalytic acetyltransferase activity in fungal thermoadaptation and mammalian etoposide-induced cellular senescence, respectively. Our findings provide a structural foundation for understanding NAT10-catalyzed cytidine acetylation, with implications for regulation and therapeutic targeting of its distinct RNA acetyltransferase activity.

Repurposing polyamines to prevent life-threatening arrhythmias in Short QT Syndrome type 3

Rapid warming in South America during the last deglaciation

Abstract Understanding tropical land temperature response to rising atmospheric CO 2 in the past is crucial for better constraining future climate projections. However, the evolution of regional land temperatures on paleoclimate timescales remains uncertain due to the paucity of precise records. Here we reconstructed temperatures across the last deglaciation using nucleation-assisted microthermometry in a stalagmite from central-eastern South America. We show that cave temperatures increased by 5.8 ± 0.3 °C (2 standard errors of the mean, SEM) from the Last Glacial Maximum to the early Holocene, broadly tracking global atmospheric CO 2 and Antarctic temperatures. Our results reveal an abrupt regional warming across the Antarctic Cold Reversal-Younger Dryas (ACR-YD) transition, linked to the weakening of the Atlantic Meridional overturning circulation (AMOC). Notably, the most rapid warming at our cave was still slower than projections of future long-term warming, highlighting the unprecedented nature of the current greenhouse gas forcing.

Structural diversity of heat-sensing channel TRPV3 with Olmsted syndrome mutations

Precision culturomics enabled by unlabeled single-cell morphology and Raman spectra

Choline metabolism drives metastasis in BRCA1-deficient ovarian cancers by activating FAM3C

Divergent 3D genome architecture of male germ cells across vertebrates

Axion electrodynamics in a topologically trivial antiferromagnet

Solving the vibrational Schrödinger equation with artificial neural networks

Abstract Artificial neural networks are universal function approximators and have shown great ability in computing the ground-state energy of the electronic Schrödinger equation, yet have not established themselves as a practical and accurate approach for solving the vibrational Schrödinger equation for realistic polyatomic molecules. Here, we propose an efficient neural-network approach for solving the vibrational Schrödinger equation and provide a detailed illustration using the methane molecule. To demonstrate the power of the proposed method, we then apply it to propane, an 11-atom molecule with 27 vibrational degrees of freedom. Using a neural network with fewer than 15,000 parameters, we obtain the ground-state energy within 1 cm −1 of the reference value obtained from a diffusion Monte Carlo calculation, as well as vibrational energies for three excited states involving C-C-C stretching/bending modes that agree with the corresponding experimental values within the experimental uncertainties. The proposed method is expected to provide highly accurate vibrational energies and wavefunctions for molecules with more than 20 atoms.

Solvent-triggered reconfiguration of optical physical unclonable functions

Abstract Optical physical unclonable functions provide artificial fingerprints through randomized light–matter interactions, but are limited by static architectures that lack adaptive defense capabilities. Although reconfigurable optical physical unclonable functions based on phase-change materials have been proposed to overcome this constraint, their reliance on light or heat makes them susceptible to unintended environmental activation. Here, we propose a solvent-triggered reconfiguration strategy for optical physical unclonable functions based on polymeric microcube arrays confined within square microwells while retaining translational and rotational degrees of freedom. A volatile solvent induces swelling that establishes wall–cube contact; evaporation-driven detachment drives non-deterministic rearrangement into new spatial configurations, regenerating the optical fingerprint. A machine-learning-based authentication framework provides robust identification of encoded physical configurations. The resulting system exhibits remarkable stability under various environmental and mechanical stresses, while exposure to volatile solvents serves as an effective trigger for reconfiguration, offering a robust pathway to decouple the intrinsic trade-off between environmental stability and reconfigurability.

A syntenic pangenome of Gardnerella reveals novel plasmids and phage, taxonomic boundaries, and species-level stratification of metabolic and virulence potential

Abstract Gardnerella species are key drivers of bacterial vaginosis (BV), a prevalent condition affecting nearly one in three women of reproductive age and associated with adverse reproductive outcomes. Despite decades of study, progress in defining Gardnerella diversity has been hindered by inconsistent taxonomy and poor-quality genomic resources. Here we sequenced 392 Gardnerella isolates spanning asymptomatic and BV-associated microbiota and integrated this collection with all publicly available genomes to create a curated, high-quality reference set of 312 genomes. We resolved 21 genomic lineages encompassing 11 species and 15 subspecies using phylogenomics, average nucleotide identity (ANI), digital DNA–DNA hybridization (dDDH) and assigned each a provisional taxonomic name. Long-read assemblies enabled construction of a syntenic Gardnerella pangenome, revealing lineage-specific repertoires of virulence, metabolic, and defense, including variable sialidases (NanH), vaginolysin, and amino-acid biosynthetic pathways alongside conserved genomic organization. Comparative methylome profiling uncovered restriction-modification system diversity suggesting barriers to genetic exchange. Finally, we identified native cryptic plasmids in Gardnerella , overturning the assumption that the genus lacks plasmids. Together, these results establish a complete genomic and functional framework for Gardnerella , providing a reproducible foundation for mechanistic and translational studies of BV and a model for resolving taxonomy and functional stratification in other urogenital-associated bacteria.

Dynamic palladium catalysis enables chiral amplification toward acyclic Schiff base atropisomers