Search

A collaboration system helps to integrate decisions between human experts and AI to optimize screening and triaging and to reduce clinicians’ workload.

Linking microscale cellular structures to macroscale features of the brain is required to fully understand its structure and function. Here, the authors present a resource which combines multi-contrast microscopy and MRI of a single whole macaque brain to facilitate multimodal analyses.

Med-PaLM, a state-of-the-art large language model for medicine, is introduced and evaluated across several medical question answering tasks, demonstrating the promise of these models in this domain.

We built an artificial neural network to control a biomechanically realistic virtual rodent, which, when trained to imitate real rats, predicts neural activity and variability across natural behaviours.

The severity of ulcerative colitis, and response to treatment, is highly variable. Here, the authors examine rectal gene expression signatures and faecal microbiomes of children and adults with the disease and provide new insights in to pathogenesis.

The process of protein crystallization is poorly understood and difficult to program through the primary sequence. Here the authors develop a computational approach to designing three-dimensional protein crystals with prespecified lattice architectures with high accuracy.

Observations from the Lucy spacecraft of the small main-belt asteroid (152830) Dinkinesh reveals unexpected complexity, with a longitudinal trough and equatorial ridge, as well as the discovery of the first contact binary satellite.

A sequencing chemistry that separates nucleotide identification from nucleotide incorporation achieves high accuracy.

Advances in electron cryo-microscopy hardware allow proteins to be studied at atomic resolution.

Deep learning-based methods to generate new molecules can require huge amounts of data to train. Skinnider et al. show that models developed for natural language processing work well for generating molecules from small amounts of training data, and identify robust metrics to evaluate the quality of generated molecules.

The authors propose Detect, a browser-based anomaly detection framework for diffusion magnetic resonance imaging tractometry data. The tool leverages normative microstructural brain features derived from healthy participants using deep autoencoders to detect anomalies at the individual level.

By leveraging paired microbiome and human gene expression data from pediatric patients who underwent hematopoietic cell transplantation, distinct lung–immune system–microorganism interactions were identified as important drivers of fatal lung injury.

In mammals, de novo DNA methyltransferase DNMT3B is essential for establishing DNA methylation patterns during embryonic development. Here the authors provide a structure-function characterization of homo-oligomeric DNMT3B, revealing the molecular basis for its vulnerability to ICF mutations.

Spectroscopy of a gravitationally lensed galaxy at a redshift of 2.7 with spatially resolved maps of two foreground damped Lyman α systems indicates a vast mass of neutral hydrogen gas, consistent with a star-forming region.

Nature Biotechnology asks a selection of researchers about the most exciting frontier in their field and the most needed technologies for advancing knowledge and applications.

Description of four new species of Xenoturbella and phylogenomic analyses, aligning Xenacoelomorpha as sister group to the rest of Bilateria, or as sister to Protostomia.

Breeding has increased crop productivity, but whether it has also changed phenotypic plasticity is unclear. Here, the authors find maize genomic regions selected for high productivity show reduced contribution to genotype by environment variation and provide evidence for regulatory control of phenotypic stability.

A protein has been engineered so that 24 identical copies self-assemble into a cube-shaped hollow cage 23 nm in diameter and containing a 130-Å-diameter inner cavity. This represents the largest and most porous structure of its type so far.

Design of a two-component protein array enables robust formation of complex large-scale ordered biologically active materials.

In contrast to their clinical success as inhibitors and targeting agents, antibodies have generally been ineffective as receptor agonists. Here, Romei et al. leverage a natural homotypic interface to tune antibody geometry, enabling optimization of agonist activity for multiple therapeutic targets.

Optical spectroscopy based on nonlinear effects usually requires coherent, ultrafast light sources such as femtosecond lasers. Here Turneret al.measure coherent multidimensional optical spectra of molecular systems using incoherent light.

Certain delafossite materials are the most conductive oxides known, for poorly understood reasons. This work elucidates this finding by uncovering a sublattice purification mechanism that enables ultrapure conductive planes even in impure crystals.

Through a community-driven competition, the PANDA challenge provides a curated diverse dataset and a catalog of models for prostate cancer pathology, and represents a blueprint for evaluating AI algorithms in digital pathology.

Emerging SARS-CoV-2 variants of concern were detected early and multiple cases of virus spread not captured by clinical genomic surveillance were identified using high-resolution wastewater and clinical sequencing.

Glucagon is hormone that signals via a dedicated g-protein coupled receptor, but downstream signaling is poorly understood. Here, Wu et al. uncover liver glucagon signaling using phosphoproteomics and define a role for the vesicle trafficking protein SEC22B in distinct metabolic actions.

Cells exhibit exceptional chemical sensitivity, yet we haven’t fully understood how they achieve it. Here the authors consider the mutual information between signals and two coupled sensors as a proxy for sensing performance and show its optimisation depending on noise level and signal statistics.

Thermal triggering of rock exfoliation has long been discounted as relevant to the evolution of rock domes. Here, the authors documented and measured recent fracturing events in California, USA to show that hot summer periods can lead to thermal stresses and cause seemingly spontaneous rock exfoliation.

It is unclear whether climate driven phenological shifts of tundra plants are consistent across the plant growing season. Here the authors analyse data from a network of field warming experiments in Arctic and alpine tundra, finding that warming differentially affects the timing and duration of reproductive and vegetative phenology.

Antibody-drug conjugates targeting high expression receptors can suffer from poor tumour penetration. Here, the authors use unconjugated antibody to improve the penetration of an antibody-dye conjugate in a clinical study, supporting further clinical investigation of the co-administration strategy.

Fibroblastic reticular cells support lymph-node function and adaptive immunity. McGeachy and colleagues show that the cytokine IL-17 is needed to trigger metabolic changes required for the proliferation and survival of these cells in reactive lymph nodes.

Directed evolution is a process of mutation and artificial selection to breed biomolecules with new or improved activity. Here the authors develop a directed evolution platform (PROTein Evolution Using Selection; PROTEUS) that enables the generation of proteins with enhanced or novel activities within a mammalian context.

A conserved network of bacteria and fungi underpins cadaver decomposition despite selection effects of location, climate and season.

A Free Ocean Carbon Enrichment study on the Great Barrier Reef finds that elevated carbon dioxide impairs net calcification of living corals and may accelerate dissolution of dead corals.

Visual input received by photoreceptors is relayed to retinal ganglion cells (RGCs), which have selectivity for inputs of certain orientations. Here, the authors show that gap junction-mediated input onto one type of RGC contributes to its orientation selectivity.

Analysis of whole-genome sequences of 25,465 individuals of European ancestry shows that rare variants contribute substantially to the heritability of height and body mass index.

Identifying a complex panel of bias dimensions to be evaluated, a framework is proposed to assess how prone large language models are to biased reasoning, with possible consequences on equity-related harms, and is applied to a large-scale and diverse user survey on Med-PaLM 2.

Halophilic organisms thrive in high salt conditions and express proteins that display desirable characteristics for industrial applications. Here, the authors use a rational design approach to transform wild-type carbonic anhydrase into a strongly halophilic enzyme.

How microglia sense amyloid plaques in Alzheimer’s disease has remained mysterious. Lemke and colleagues report that TAM receptor kinases are absolutely required for normal microglial recognition of, response to and phagocytosis of Aβ plaques. Surprisingly, TAM-mediated microglial phagocytosis of Aβ material does not constrain, but rather promotes, the formation of dense-core plaques.

Deep-learning models trained on external eye photographs can detect diabetic retinopathy, diabetic macular oedema and poor blood glucose control more accurately than models relying on demographic and medical history data.

Targeted protein degradation (TPD) is a key modality for drug discovery. Here the authors present the discovery and analysis of reversible DCAF1-PROTACs, which show efficacy in cellular environments resistant to VHL-PROTACs or with acquired resistance to CRBN-PROTACs.

The mechanisms underlying drought-induced tree mortality are not fully resolved. Here, the authors show that, across multiple tree species, loss of xylem conductivity above 60% is associated with mortality, while carbon starvation is not universal.

A phase-separation mechanism is proposed for the dominance of the surface tension effect over the solute effect in the observed activation of ultrafine cloud condensation nuclei.

By comparing the genome-wide profile of H4K16ac in AD with younger and elder controls, the authors propose a mechanism for how age is a risk factor for AD: a histone modification, whose accumulation is associated with aging, is dysregulated in AD.

Metal-catalysed prebiotic reactions have been proposed as forerunners of modern metabolism. Now, an abiotic pathway resembling the reverse tricarboxylic acid cycle has been shown to proceed without metal catalysis. The reaction of glyoxylate and pyruvate produces a series of α-ketoacid tricarboxylic acid analogues, and provides a route to generate α-amino acids by transamination.

The growth factor NELL-1 induces bone formation during development, but its role in osteoporosis is unknown. This study shows that NELL-1 binding to integrin ß1 induces Wnt/ß-catenin signalling in the bone and restores bone mineral density in osteoporotic mice and sheep, suggesting the therapeutic potential of NELL-1 for the treatment of bone loss.