Browse Articles

Isolation of a Terminal Cobalt Nitride in a Metal–Organic Framework

Domain wall motion-driven magnetic convolutional accelerator

Abstract Modern computing powers applications from data analysis to artificial intelligence but now faces limitations. The slowdown of device scaling and the bottleneck between memory and processors motivate architectures that unify computation and data storage. Convolution is a core operation in learning, vision, and signal processing, yet its conventional implementation incurs high energy, high latency, and limited scalability. Magnetic systems that host spin textures, such as domain walls, offer dynamic behaviors that enable computation beyond traditional logic. Here we introduce a compute-in-memory platform that performs convolution by sequentially shifting magnetic domains and sensing the resulting signals. Information is written directly into domain patterns, processed through controlled motion, and read electrically, forming a nonvolatile structure suited for convolution tasks. This approach supports applications including Fourier analysis, neural networks, and image processing, achieving 10 3 to 10 5 improvements in area, energy, and throughput over existing technologies, marking a concrete advance in spintronic computing.

Subjective slow walking speed is associated with locomotive syndrome severity in 34,935 adults undergoing medical checkups

Synthesis of Heparan Sulfate Hexadecasaccharides and Their Molecular Interaction with Mycobacterial Heparin-Binding Hemagglutinin for the Detection of <i>Mycobacterium tuberculosis</i>

Precision phenotyping of type 2 diabetes in chinese populations using a variational autoencoder-informed tree model

Language of change in online narratives of recovery from disordered gaming

Magneto-Optical Readout of a Chiral Single-Molecule Magnet at Telecom Wavelengths

Synthetic gain for electron-beam spectroscopy

kNDVI reveals vegetation dynamics and hydro–edaphic controls in inner Mongolia (2000–2024)

Late-Stage Diversification of Native Tryptophan-Containing Peptides and Peptide Drugs through Nitrogen Atom Insertion

Loss of Fsr quorum sensing promotes biofilm formation and worsens outcomes in enterococcal infective endocarditis

Abstract Infective endocarditis (IE) is a severe heart infection caused predominantly by Gram-positive bacteria forming biofilm on heart valves. While biofilm formation is central to disease progression, the underlying bacterial mechanisms remain poorly understood. Here, we identify the Fsr quorum sensing (QS) system of Enterococcus faecalis as an unexpected negative regulator of biofilm and pathogenesis in IE. Using microfluidic and in vivo models, we show that blood flow prevents Fsr activation in early IE, with Fsr induction occurring only later, once bacteria form biofilm microcolonies and become shielded from flow. Deletion of Fsr promotes robust biofilm growth, driven partly through the downregulation of GelE and SprE proteases, reprograms metabolism by upregulating lrgAB to enhance pyruvate utilization, and increases gentamicin tolerance in vivo. Furthermore, we show that GelE cleaves the human pro-IL-1β into an active form, suggesting a species-specific mechanism for inflammation modulation by QS. In support of these findings, analysis of IE patient cohorts shows that naturally occurring Fsr-deficient E. faecalis strains are associated with prolonged bacteremia. Overall, our findings provide insights into how host blood flow impacts QS activation, which, in turn, regulates pathogenesis in IE, and highlight the Fsr QS as a potential determinant of clinical disease course.

School bullying predicts malevolent creativity in middle school students through anger and hostile attribution bias

Solvation-Structure Design of Multicomponent Eutectic Electrolytes Enabling Al-Rich Alloy Growth in Aqueous Aluminum-Ion Batteries

Sequence-based generative AI design of versatile tryptophan synthases

Abstract Enzymes are powerful and sustainable catalysts, but their widespread application is limited by the difficulty of identifying functional starting points for optimization, creating a major bottleneck in early- stage biocatalyst discovery. Designing libraries of such starting enzymes remains particularly challenging. Here, we use the GenSLM protein language model to generate novel β -subunit of tryptophan synthase (TrpB) enzymes that express in Escherichia coli and are both stable and catalytically active. Many generated TrpBs also display significant substrate promiscuity, outperforming their natural counterparts on non-native substrates. Some even surpass laboratory-evolved TrpBs. Comparison of the most-active and most-promiscuous generated TrpB to its closest natural homolog confirms that the enhanced versatility is absent from the natural enzyme, highlighting the creative potential of generative models. These results demonstrate that the generated TrpBs not only preserve natural structure and function but also acquire non-natural properties, establishing generative models as powerful tools for biocatalyst discovery and engineering.

Genetic diversity and transmission dynamics of SARS-CoV-2 in East Africa

When Core Orbitals Act as Valence Orbitals: Linear and Bent Geometries of <i>MX</i> ₂ for <i>X</i> a Halogen and <i>M</i> an Alkaline Earth

Chronically implantable μLED arrays for optogenetic cortical surface stimulation in mice

Abstract Cortical implants are a proven clinical neurotechnology with the potential to transform our understanding of cognitive processes. These processes rely on complex neuronal networks that are difficult to selectively probe or stimulate. Optogenetics offers cell-type specificity, but achieving the density and coverage required for chronic, high-resolution modulation remains a challenge. We fabricated 100-element μ LED arrays (200 μ m pixel pitch, 2 × 2 mm 2 footprint) coupled into chronically implantable systems for optogenetic stimulation of the mouse cortex. The μ LEDs remain stable for over 300 hours continuous operation time in vivo, allowing for months-long chronic experiments. Simultaneous electrophysiology recordings confirmed robust neuronal responses at low μ LED drive currents (&lt; 5 mA), minimising thermal effects. Here we show that our device can be chronically implanted in freely-behaving mice and drive behavioural outcomes from spatiotemporal, patterned, optogenetic stimulation of auditory cortical circuits.

Influence of a concrete lattice–vegetation composite revetment on levee slope stability based on an improved SWCC model

Reversible Photoswitching of Donor–Acceptor Stenhouse Adducts in Water

Variable temperature processing by plasmodesmata regulates robust bud dormancy release

Abstract Dormancy is a key mechanism in perennial plants in boreal and temperate regions, protecting buds from winter damage by repressing precocious bud break before spring onset. How plants robustly time dormancy release under fluctuating environments remains unknown. Here, we show that, rather than simply sensing cold duration, buds leverage warm spikes to sense winter progression and time dormancy release. This timing mechanism is mediated by previously unrecognized regulation of plasmodesmata by warm spikes acting through tree ortholog of FLOWERING LOCUS T ( FT1 ) and the gibberellic acid pathway. Our results reveal FT1 as a previously unrecognized, suppressor of callose levels and show that warm spikes repress cold induction of FT1 and GA pathway to suppress PD opening and dormancy release. Importantly, buds exhibit heterogeneity in bud break. This heterogeneity in bud break crucial for bet hedging is amplified under temperature fluctuations and is associated with the thermal responsiveness of plasmodesmata. Altogether, our work reveals dynamic plasmodesmata regulation as a crucial tissue-level mediator of variable temperature processing by buds, enabling robust adaptation of trees to seasonal changes.