Search

Teresa Lambe is working with AstraZeneca to give the world a shot against the SARS-CoV-2 coronavirus.

Researchers are arguing over whether ‘disruptive’ or ‘novel’ science is waning – and how to remedy the problem.

Papers from 2015 are a tougher read than some from the nineteenth century — and the problem isn't just about words, says Philip Ball.

This study examines long-term changes in species richness across tropical forests in the Andes and Amazon. Hotter, drier and more seasonal forests in the eastern and southern Amazon are losing species, while Northern Andean forests are accumulating species, acting as a refuge for climate-displaced species.

Large-scale computational and in vitro analyses identify commensal type III secretion systems and substrates in the human gut microbiome that can interact with human proteins to modulate immune pathways.

Satellite observations over the Greenland Ice Sheet reveal a destructive mode of meltwater drainage whereby a subglacial flood induced by the rapid drainage of a subglacial lake burst through the surface, fracturing the ice sheet.

Legislative efforts aim to tackle rising incidence of disease.

Carbapenems are last-resort antibiotics, but resistance is rising due to hydrolyzing enzymes called carbapenemases. The authors present a machine learning algorithm and web-app to rapidly predict carbapenemase-production in Enterobacterales.

Disease heterogeneity complicates precision medicine, which focuses on single conditions and ignores shared mechanisms. Here the authors introduce ‘pan-disease’ analysis using a deep learning model on multi-organ data, identifying 11 AI-derived biomarkers that reveal new therapeutic targets and pathways, enhancing patient stratification for disease risk monitoring and drug discovery.

Authors perform an analysis of the patient data and risk factors to evaluate unfavorable outcomes and adverse events in adults with pulmonary tuberculosis treated with a 4-month rifapentine based regimen. Low rifapentine exposure was the most clinically significant risk factor for treatment failure and tuberculosis relapse.

The young discipline of sports science is finding ways to stretch the boundaries of human biology.

Examples of materials with non-trivial band topology in the presence of strong electron correlations are rare. Now it is shown that quantum fluctuations near a quantum phase transition can promote topological phases in a heavy-fermion compound.

Human genetic variation influences instability of DNA repeats that commonly expand as we age.

Ion diffusion region is an indicator of active magnetic reconnection, but it had not been detected in Jupiter’s magnetosphere previously. Here, the authors show a magnetic reconnection event in Jupiter’s inner magnetosphere that presents the detection of an ion diffusion region.

CONCORD generates denoised cell encodings that preserve key biological structures in single-cell datasets.

This study demonstrates that 16p11.2 hemideletion, a genetic deletion linked to neurodevelopmental disorders, affects male and female mice differently, altering specific striatal neuronal circuits and driving sex-specific behavioral changes.

Accurately measuring dietary intake has long been a challenge in nutrition research. Integrating emerging tools with multisampling strategies and a dietary assessment methodology aligned with the research aims enables a more objective and comprehensive evaluation of dietary behaviours—and a deeper understanding of diets’ impact on human and planetary health.

Here, in a randomized clinical trial, the authors report that an experimental young child formula containing a combination of Limosilactobacillus reuteri DSM 17938 and galacto-oligosaccharides enhances bone and muscle development in early life through modulation of the microbiome.

SmartEM is a ‘smart’ pipeline for electron microscopy-based data acquisition for connectomics. In order to efficiently image large datasets, the approach involves imaging at short pixel dwell times and identifying problematic regions that are then imaged with longer dwell times and therefore higher quality.

An architecture inspired by Hopfield networks based on a programmable, stable, room-temperature optoelectronic oscillator-based photonics Ising machine is introduced that can be used to efficiently address optimization and combinatorics problems.

For many young people with the condition, screen interactions are especially hard to resist — and intensify the mental-health challenges they face.

Detection of somatic mutations in cell-free DNA is challenging due to low variant allele frequencies and extensive DNA fragmentation. Here, the authors use longitudinal samples from colorectal and breast cancer patients to clarify performance limits of current approaches and support application of cfDNA analyses in cancer liquid biopsies.

Electrochemical hydrogenation drives a reversible conductor–insulator transition in graphene. Authors show that it is 10⁶× faster than other methods and tunable by isotope effects and lattice corrugations, enabling ionic control of 2D electronics.

Two researchers describe how body size and having a disability can challenge perceptions of what a scientist should look like.

The authors propose a Generalized Latent Equilibrium framework for fully local credit assignment in physical, dynamical neuronal networks such as the brain. By exploiting dendritic structure and prospective coding in cortical neurons, it enables an online approximation of backpropagation through space and time.

Active fluids, such as bacterial suspensions, exhibit chaotic flows at low Reynolds number - a phenomenon known as active turbulence. Here, the authors show a discontinuous transition from laminar to chaotic flows in unconfined active nematics.

Amorphous phases in alkali-activated slag cement are explored using spatially resolved synchrotron X-ray techniques. Local atomic structures of individual phases show the impact of magnesium content on resistance to CO₂-induced degradation.

The selective autophagy receptor p62/SQSTM1 is organized in a structured double helical filament in vitro, the cryo-electron tomographic images in ATG5 knockdown conditions reveal Ca-rich layers of p62 oligomers enwrapping lipid droplets in situ.

Children with dyslexia show significant differences in Visual Word Form Area size and specialization compared to typical readers, suggesting enduring neural characteristics exist even after targeted intervention increases reading ability scores.

This study demonstrates that the fastest astrophysical jets are locked to a fixed axis, whereas slower jets can precess or wobble. The findings imply that the most relativistic jets are tied to the spin axis of the black hole that launched them.

Through next-generation spectral analysis, scientists have uncovered an evolutionary path for Wolf–Rayet stars in metal-poor environments. Characterized by hard ionizing radiation, these stars challenge current assumptions about massive star evolution.

Functional material design can be enhanced by taking inspiration from nature. This Primer describes how micropatterns inspired by the natural world can be designed, fabricated and used to solve technical challenges.

Behavioural and psychosocial interventions can effectively reduce the risk of cardiovascular disease. In this Perspective, Burg et al. identify the elements and actions needed for a broad implementation of these interventions in standard cardiovascular care.

Pl@ntBERT is a language-based AI model that learned the ‘syntax’ of plant assemblages, predicting likely species and inferring habitats by modelling biotic relationships.

Actionable goals are needed to guide the world towards what needs to happen most quickly: shifting economies to clean energy sources.

A scientific foundation is required to establish nations’ responsibilities in a hotter ‘overshoot’ world.

Regional, place-based biodiversity information is used to comprehensively map and quantify biodiversity intactness of sub-Saharan Africa to inform national and global sustainability policies and planning.

The International Brain Laboratory presents a brain-wide electrophysiological map obtained from pooling data from 12 laboratories that performed the same standardized perceptual decision-making task in mice.

Nimbus, a deep learning model, uses a large multiplexed imaging dataset to predict the likelihood of marker positivity in single cells.

From education to pandemic preparedness and public health, the past year has seen huge stress put on US science.