Browse Articles

Daily briefing: The brain builds a sentence neuron by neuron

Human serum albumin profiling by top-down analysis enables multi-class liver fibrosis staging: a cross-platform validation study

Direct development of stem tetrapods across the fin-to-limb transition

Modern amphibians are characterized by an aquatic larval stage that ends abruptly with a period of widespread tissue remodeling (metamorphosis) upon transition to terrestrial adulthood. A transient larval stage ending in gradual metamorphosis is often assumed for the earliest digited tetrapods, but direct evidence of this larval stage is lacking. Exceptionally preserved stem tetrapod hatchlings show that a transient larval stage was absent in tetrapods both before and after the fin-to-limb transition; instead we identified soft- and hard-tissue evidence of direct development, falsifying hypotheses of an ancestral origin of metamorphosis or of a gradual larval-postlarval transition serving as a template for lissamphibian metamorphosis. We argue that a transient larval period culminating in metamorphosis originated near or within the tetrapod crown group as part of a broader suite of traits associated with terrestrialization.

Decoupling pitch and angle of attack with a variable-incidence wing extends quadrotor flight endurance

Abstract The aerodynamic efficiency of conventional lifting-wing quadrotors is severely constrained by the rigid coupling between fuselage pitch attitude and wing angle of attack (AoA), which prevents the vehicle from maintaining optimal lift-to-drag performance across varying flight speeds. This study presents a novel variable-incidence lifting-wing quadrotor that effectively decouples fuselage attitude from wing aerodynamics. The proposed mechanism incorporates a 143 g lightweight lifting wing and a dual-actuator drive system, with a total structural mass penalty of only 209 g. High-fidelity Computational Fluid Dynamics (CFD) simulations, reveal that the variable-incidence strategy alleviates detrimental rotor-wing interference. Specifically, a fuselage pitch of − 20° combined with a wing AoA of 10°–12° reestablishes a strong suction peak on the wing’s upper surface, maximizing passive lift. Systematic flight experiments validate the decoupling control strategy, demonstrating that the prototype achieves a maximum normalized power-saving ratio of 42.98% at 10 m/s relative to hover. Furthermore, even when accounting for the 209 g structural weight penalty, the design provides a net 13.22% efficiency improvement over the fixed-incidence concept of the original RflyLW. Leveraging these aerodynamic enhancements, predictive analyses indicate that the UAV can achieve an extended maximum flight endurance of 46.8 min and a practical range of 28.1 km. These results, showing high consistency between numerical predictions and experimental data, offer a robust and highly efficient solution for extending the endurance of UAVs in energy-constrained environments.

Hour of the wolf

Wolves have made a spectacular comeback in Europe. But attacks on livestock and humans have caused a heated debate about the limits of coexistence

An in situ non-contact 3D microscopic measurement system for temporal bone anatomy based on stereo imaging

ScS-triggered slip on megathrust interfaces after the 2011 <i>M</i> _W 9.0 Tohoku-Oki earthquake

We report an extraordinary observation of ground motion in Japan after the moment magnitude ( M W ) 9.0 2011 Tohoku-Oki earthquake attributed to a multiplate-interface slip event triggered by shear wave that traveled to the Earth’s core and back. The megathrust earthquake generated a strong ScS phase with a peak-to-peak amplitude exceeding 1 centimeter in Japan. Superposed on this waveform, an eastward steplike displacement of up to 5 to 6 millimeters was recorded in Global Navigation Satellite System (GNSS) data throughout Japan. This likely originated from slip on the megathrust interfaces triggered by the nearly simultaneous arrival of the ScS wave across Japan. Such an ScS triggering is a previously unrecognized source of seismic hazard, which can potentially (re)activate the mainshock area and the broader surrounding megathrust interfaces.

The context of new media art for photography generation based on diffusion models

Decoupling of global metabolic flux and proteome partitioning in bacteria

Bacteria regulate homeostatic growth by adjusting proteome composition. In Escherichia coli , this coordination is mediated by guanosine tetraphosphate and pentaphosphate, collectively termed (p)ppGpp, which couple amino acid supply with ribsosome production. We identified a distinct architecture in Bacillus subtilis , in which guanosine triphosphate (GTP), not (p)ppGpp, controls proteome allocation. Translational inhibition resulted in GTP depletion and suppressed amino acid biosynthesis through feedback inhibition without altering ribosome abundance, establishing a regulated decoupling between total amino acid flux and proteome composition, with flux deviating from proteome-based predictions. By artificially adjusting GTP concentrations, we recoupled flux and proteome, restoring growth to maximal amounts. The regulated suboptimality enables a trade-off to balance growth and stress resilience. Similar GTP-based strategies were present in other Firmicute species, indicating possible evolutionary conservation. Proteome composition and metabolic flux have distinct regulatory layers in some bacteria.

Development of a portable device for the detection of anti-PEG antibodies in human plasma

AI finds legal loopholes, even when not asked

Models repeatedly discovered ways to exploit regulations, and current safeguards largely failed to stop them

Correction: Exploitation of phage Carin-5’s own DNA polymerase to sequence its T-hypermodified genome

Lamprey 3D single-cell transcriptomics reveals ancestral and specialized features of the vertebrate brain

The lamprey occupies a pivotal position for elucidating vertebrate brain evolution. Using spatial transcriptomics and single-nucleus RNA sequencing, we generated a three-dimensional molecular atlas of the lamprey brain, identifying 209 distinct cell clusters across 14 regions. Cross-species comparisons revealed broad conservation of regional spatial architecture, defining an ancestral organizational blueprint. Within this conserved framework, however, marked lineage-specific divergence emerged. We observed extensive neuronal specialization across vertebrate lineages, accompanied by regulatory shifts associated with spatial reorganization and functional diversification of neuronal populations. Additionally, our results suggest that a cerebellum-like architecture predates the jawed vertebrate cerebellum. Together, these findings identified constraints on neural organization and detected cellular innovations driving evolutionary diversification.

These ‘master’ proteins protect us from deadly mutations — and could inspire new drugs

Effect of blended NPSB fertilizer and seeding rates on yield and yield components of bread Wheat (Triticum aestivum L.) in Banja District, Northwestern Ethiopia

Maternal trans-vaccenic acid shapes neonatal T cell development and early-life immune imprinting

How maternal nutrition influences neonatal immune development and imprinting through breastfeeding remains largely unclear. We report that maternal supplementation with trans-vaccenic acid (TVA), the predominant naturally occurring trans-fatty acid in human breast milk, promoted neonatal T cell development in mice. Neonates fed by mothers on a TVA-enriched diet showed an expanded naïve cluster of differentiation 4 (CD4 + ) T cell population and enhanced adaptive immunity against infection. TVA reprogrammed neonatal naïve CD4 + T cells through a G protein–coupled receptor–CCCTC-binding factor axis and promoted T helper cells (Th1)–skewing by cooperating with the transcription factor TBX21. Early-life exposure to maternal TVA via breastfeeding supported long-lasting antiviral immunity in adulthood. Our findings establish the multifaceted benefits of maternal nutrition and breastfeeding via TVA in promoting infant immune homeostasis and protective immunity.

High-resolution photovoltaic power forecasting using machine learning models under seasonal and stress conditions

Tracking the roots of zoonoses

Ancient pathogen DNA from animals elucidates the origin and evolution of infectious diseases

Grey wolf optimized neural network with hybrid feature extraction for heart disease prediction

Abstract With an increasing mortality rate due to heart diseases, there is a critical need for early and reliable cardiovascular disease prediction. However, when it comes to health records data, traditional approaches have primarily utilized a single prediction model. This study proposes an advanced approach that leverages Support Vector Machine (SVM) in independent parallel streams with Convolutional Neural Networks (CNN) for extracting features, combined with an ANN enhanced via Grey Wolf Optimization (GWO) for predictive classification. First, the input data is passed into the SVM and CNN simultaneously to extract relevant and important features by each method. Both extractors receive identically preprocessed inputs simultaneously; their outputs are concatenated post-extraction. The SVM extracts probability-based discriminative features reflecting class separation confidence, while CNN extracts hierarchical convolutional features capturing local patterns and non-linear attribute interactions. Then, the attributes extracted by SVM and the attributes extracted by CNN are concatenated to get a comprehensive set of features. The concatenation enriches the feature space with diverse representational characteristics: SVM contributions provide regularized decision confidence scores, while CNN contributions offer multi-scale pattern descriptors. This heterogeneous feature integration enables the ANN to learn more robust classification boundaries by accessing both discriminative and descriptive feature domains simultaneously. Second, the concatenated features are sent to the ANN for prediction. The GWO algorithm is used to configure ANN parameters like channel size, drop-out rate, and learning rate. The fitness function maximizes stratified 5-fold cross-validation accuracy. The experiments were conducted on the Heart Disease dataset collection, collected from the UCI Machine Learning Repository (Cleveland, Hungarian, Switzerland, and Long Beach), which includes 14 variables for heart disease prediction. All results represent mean ± standard deviation from nested cross-validation. The results demonstrate that the proposed SVM–CNN+ANN-GWO model achieves a high classification accuracy of 91.80% ± 1.2% with the Cleveland, 91.53% ± 1.5% with Hungarian, 96.08% ± 0.8% with Switzerland, and 87.69% ± 2.1% with VA Long Beach databases, respectively. Statistical significance was confirmed via paired Wilcoxon tests ( $$p &lt; 0.05$$ ). It outperforms existing baseline approaches and other recent approaches from previous studies based on predictive accuracy. It achieves promising performance in heart disease prediction across all four databases and hence generalizes well on different datasets.

AI may raise the bar and thin the pipeline