Search

Metal–support interfaces play a crucial role in dictating the catalytic behavior of heterogeneous systems. Here, the authors reveal a unique looping metal–support interaction in NiFe–Fe₃O₄ catalysts, where spatially separated yet coupled redox cycles boost hydrogen oxidation, providing new insights for the design of efficient heterogeneous catalysts.

A new artificial intelligence model, DeepSeek-R1, is introduced, demonstrating that the reasoning abilities of large language models can be incentivized through pure reinforcement learning, removing the need for human-annotated demonstrations.

Neural tissue is a key signalling centre for fundic–antral patterning in development of gastric organoids featuring asymmetric PDX1 patterning.

DeepRETStroke is a deep learning system using retinal photographs to detect silent brain infarction and predict future stroke events.

The evolution of superconductivity in LiFeAs with respect to pressure or strain remains elusive. Here, the authors observe different response of superconducting states due to different orientations of local wrinkles on the surface of LiFeAs.

Adenoids and tonsils in children can become hypertrophic because of multiple infections. Here the authors use single cell transcriptomics to assess the immune cell composition of these hypertrophic organs and find B, T and NK cells and exhausted memory CD4 T cells indicating immune changes in these organs associated with respiratory infection.

The brain and body are necessarily connected. Here the authors show that brain blood flow and electrical activity are coupled with systemic physiological changes in the body.

Understanding and controlling morphogenesis is vital for biology and organoid technology. Here, the authors report an efficient biomechanical system to generate gut spheroids, and reveal instability-driven morphogenetic transitions with computational models.

The stochastic features of memristors make them suitable for computation and probabilistic sampling; however, implementing these properties in hardware is extremely challenging. Lin et al. introduce an approach that leverages the cycle-to-cycle read variability of memristors as a physical random variable for in situ, real-time random number generation, and demonstrate it on a risk-sensitive reinforcement learning task.

A double-layer-coated slow-release fertilizer (DCSRF) was prepared with urea as the core, citric acid-modified starch and polyvinyl alcohol as the inner coating, and sodium alginate as the outer coating. Compared with the single-layer coated fertilizer (SCSRF), the DCSRF demonstrated superior slow-release effects, releasing 69.2% of the nutrients within 38 days in water. Additionally, adding 2% DCSRF to soil increased its maximum water-holding capacity by 8%. Pot trials revealed that the DCSRF considerably improved target crop growth, including tiller number, plant height, root depth, and leaf width, outperforming the other methods.

Ferroelectric phase stability is a pivotal challenge for fluorite-structure ferroelectrics. Using electron microscopy, a ferroelastically protected reversible polar-to-non-polar phase transition in ZrO₂ is observed and the critical strain state to break the reversibility is measured.

Mpox virus has spread globally since 2022 with cases in many previously non-endemic countries including China. Here, the authors describe the genomic epidemiology of the first local outbreak in Shenzhen, China, including 92 cases reported from June–Oct 2023.

Zhang et al. identify a unique population of cardiac-resident macrophages that could be a potential therapeutic strategy for sepsis-induced cardiomyopathy.

Functional magnetic resonance imaging signals linked to heart rate and respiration reveal age-related alterations in vascular physiology and autonomic function.

Interpretable machine learning offers a powerful method for accelerating catalyst design. Here the authors report an interpretable descriptor model to unify activity and selectivity prediction for multiple electrocatalysis using only easily accessible intrinsic properties.

The application of quantum computing to computational chemistry faces various experimental and theoretical challenges. Now, a quantum simulation on a noisy quantum device has achieved chemical accuracy for small H₂ and LiH molecules.

Previous studies have shown that high alcohol-producing Klebsiella pneumoniae (HiAlc Kpn) in the intestinal microbiome could be one of the causes of non-alcoholic fatty liver disease (NAFLD). Here, the authors show the effectiveness of phage in mice with HiAlc Kpn-induced NAFLD indicating phage therapy targeting gut microbiota may be an alternative to antibiotics, with potential efficacy and safety.

Despite the discovery of Majorana zero modes (MZM) in iron-based superconductors, sample inhomogeneity may destroy MZMs during braiding. Here, authors observe MZM in impurity-assisted vortices due to tuning of the bulk Dirac fermions in a homogeneous superconductor LiFeAs.

Cryo-electron microscopy structures and biochemical analyses provide insight into how short prokaryotic Argonaute proteins are assembled and activated, and reveal that oligomerization has a key role in driving catalytic activity.

Kagome materials host 2D planes which give rise to kagome physics, but these are typically embedded in the bulk. Huang et al. demonstrate a strategy for generating surface kagome electronic states by vertical p-d electronic hybridization between surface atoms and the buried Co kagome network in Co₃Sn₂S₂.

A scalable approach is provided for preparing and verifying intermediate-scale genuine entanglement on a 66-qubit superconducting quantum processor.

Agriculture dependent on snowmelt will face serious challenges under climate change, which increases risks for countries that import these crop products. Food security and livelihoods in countries heavily exposed to global food trade may be vulnerable even though domestic production is not affected.

Two-dimensional transition metal carbides and nitrides, known as MXenes, are currently considered as energy storage materials. A generic Lewis acidic etching route for preparing high-rate negative-electrode MXenes with enhanced electrochemical performance in non-aqueous electrolyte is now proposed.

In the presence of stress stimuli, the endoplasmic reticulum either adapts the protein synthesis or triggers an apoptotic response, but the mechanisms underlying this decision are unknown. Kaufman and colleagues show that the ER stress response factors ATF4 and CHOP increase protein synthesis, which in turn induces oxidative stress and increased ATP consumption, leading to cell death during chronic ER stress.

A transmission-matrix-based framework is proposed to establish the intrinsic link between angular diversity and spatial footprint in multi-channel systems and to derive a bound on the thickness of nonlocal metasurfaces.

In this Perspective, the authors discuss the idea of targeting histone modifications in bladder cancer, providing a comprehensive overview of findings from preclinical and translational studies. Clinical trials in which the efficacy of molecules targeting histone modifiers was assessed in bladder cancer are also discussed.

In this protocol, the authors describe the design, fabrication, purification, characterization and potential biomedical applications of a self-assembling TDN-based multifunctional delivery system.