Free radicals were first discovered more than 120 years ago by Gomberg1 and the first radical cross-couplings demonstrated by Kochi in the 1970s (ref. 2). In contrast to widely used polar cross-coupling chemistry to forge C(sp2)-C(sp2) bonds (such as Suzuki, Negishi and Kumada), radical cross-coupling is advantageous when applied to the coupling of saturated systems because of the mild conditions used and enhanced chemoselectivity associated with single-electron chemistry. The ability to use ubiquitous carbon-based fragments (such as carboxylic acids, alcohols, amines and olefins) in cross-coupling has greatly simplified access to various complex molecules3-9. Apart from these advantages, enantiospecific coupling reactions involving free radicals are unknown and generally believed to be challenging because of their near-instantaneous racemization (picosecond timescale)10. As a result, controlling the stereochemical outcome of radical cross-coupling can be achieved only on a case-by-case basis using bespoke chiral ligands11 or in a diastereoselective fashion guided by nearby stereocentres12. Here we show how readily accessible enantioenriched sulfonylhydrazides and low loadings of an inexpensive achiral Ni catalyst can be used to solve this challenge, thereby enabling enantiospecific, stereoretentive radical cross-coupling between enantioenriched alkyl fragments and (hetero)aryl halides without exogenous redox chemistry or chiral ligands. Calculations support the intermediacy of a unique Ni-bound diazene-containing transition state with C-C bond formation driven by loss of N2.

The rapid advent of high-throughput omics technologies has created an exponential growth in biological data, often outpacing our ability to derive molecular insights. Large-language models have shown a way out of this data deluge in natural language processing by integrating massive datasets into a joint model with manifold downstream use cases. Here we envision developing multimodal foundation models, pretrained on diverse omics datasets, including genomics, transcriptomics, epigenomics, proteomics, metabolomics and spatial profiling. These models are expected to exhibit unprecedented potential for characterizing the molecular states of cells across a broad continuum, thereby facilitating the creation of holistic maps of cells, genes and tissues. Context-specific transfer learning of the foundation models can empower diverse applications from novel cell-type recognition, biomarker discovery and gene regulation inference, to in silico perturbations. This new paradigm could launch an era of artificial intelligence-empowered analyses, one that promises to unravel the intricate complexities of molecular cell biology, to support experimental design and, more broadly, to profoundly extend our understanding of life sciences.

Ocean submesoscale (1-100 km) processes and their substantial impact on Earth's climate system have been increasingly emphasized in recent decades by high-resolution numerical models and regional observations1-11. However, the dynamics and energy associated with these processes, including submesoscale eddies and nonlinear internal waves, have never been observed from a global perspective. Where, when and how much do these submesoscale processes contribute to the large-scale ocean circulation and climate system? Here we show data from the recently launched Surface Water and Ocean Topography (SWOT) satellite12 that not only confirm the characteristics of submesoscale eddies and waves but also suggest that their potential impacts on ocean energetics, the marine ecosystem, atmospheric weather and Earth's climate system are much larger than anticipated. SWOT ushers in a new era of global ocean observing, placing submesoscale ocean dynamics as a critical element of the Earth's climate system.

The semiconductor industry is experiencing an accelerated transformation to overcome the scaling limits of the transistor and to adapt to new requirements in terms of data storage and computation, especially driven by artificial intelligence applications and the Internet of Things. In this process, new materials, devices, integration strategies and system architectures are being developed and optimized. Among them, memristive devices and circuits-memristors are two-terminal memory devices that can also mimic some basic bioelectronic functions-offer a potential approach to create more compact, energy-efficient or better-performing systems. The memristor industry is growing quickly, raising abundant capital investment, creating new jobs and placing advanced products in the market. Here we analyse the status and prospects of the memristor industry, focusing on memristor-based products that are already commercially available, prototypes with a high technological readiness level that might affect the market in the near future, and discuss obstacles and pathways to their implementation.

Sensitive methods for detection of cell-free RNA (cfRNA) could facilitate non-invasive gene expression profiling and monitoring of diseases1-6. Here we describe RARE-seq (random priming and affinity capture of cfRNA fragments for enrichment analysis by sequencing), a method optimized for cfRNA analysis. We demonstrate that platelet contamination can substantially confound cfRNA analyses and develop an approach to overcome it. In analytical validations, we find RARE-seq to be approximately 50-fold more sensitive for detecting tumour-derived cfRNA than whole-transcriptome RNA sequencing (RNA-seq), with a limit of detection of 0.05%. To explore clinical utility, we profiled 437 plasma samples from 369 individuals with cancer or non-malignant conditions and controls. Detection of non-small-cell lung cancer expression signatures in cfRNA increased with stage (6 out of 20 (30%) in stage I; 5 out of 8 (63%) in stage II; 10 out of 15 (67%) in stage III; 80 out of 96 (83% sensitivity) in stage IV at 95% specificity) and RARE-seq was more sensitive than tumour-naive circulating tumour DNA (ctDNA) analysis. In patients with EGFR-mutant non-small-cell lung cancer who developed resistance to tyrosine kinase inhibitors, we detected both histological transformation and mutation-based resistance mechanisms. Finally, we demonstrate the potential utility of RARE-seq for determination of tissue of origin, assessing benign pulmonary conditions and tracking response to mRNA vaccines. These results highlight the potential value of ultrasensitive cfRNA analysis and provide proof of concept for diverse clinical applications.

During classical non-homologous end joining (cNHEJ), DNA-dependent protein kinase (DNA-PK) encapsulates free DNA ends, forming a recruitment platform for downstream end-joining factors including ligase 4 (LIG4)1. DNA-PK can also bind telomeres and regulate their resection2-4, but does not initiate cNHEJ at this position. How the end-joining process is regulated in this context-specific manner is currently unclear. Here we show that the shelterin components TRF2 and RAP1 form a complex with DNA-PK that directly represses its end-joining function at telomeres. Biochemical experiments and cryo-electron microscopy reveal that when bound to TRF2, RAP1 establishes a network of interactions with KU and DNA that prevents DNA-PK from recruiting LIG4. In mouse and human cells, RAP1 is redundant with the Apollo nuclease in repressing cNHEJ at chromosome ends, demonstrating that the inhibition of DNA-PK prevents telomere fusions in parallel with overhang-dependent mechanisms. Our experiments show that the end-joining function of DNA-PK is directly and specifically repressed at telomeres, establishing a molecular mechanism for how individual linear chromosomes are maintained in mammalian cells.

Protein misfolding diseases, including α1-antitrypsin deficiency (AATD), pose substantial health challenges, with their cellular progression still poorly understood1-3. We use spatial proteomics by mass spectrometry and machine learning to map AATD in human liver tissue. Combining Deep Visual Proteomics (DVP) with single-cell analysis4,5, we probe intact patient biopsies to resolve molecular events during hepatocyte stress in pseudotime across fibrosis stages. We achieve proteome depth of up to 4,300 proteins from one-third of a single cell in formalin-fixed, paraffin-embedded tissue. This dataset reveals a potentially clinically actionable peroxisomal upregulation that precedes the canonical unfolded protein response. Our single-cell proteomics data show α1-antitrypsin accumulation is largely cell-intrinsic, with minimal stress propagation between hepatocytes. We integrated proteomic data with artificial intelligence-guided image-based phenotyping across several disease stages, revealing a late-stage hepatocyte phenotype characterized by globular protein aggregates and distinct proteomic signatures, notably including elevated TNFSF10 (also known as TRAIL) amounts. This phenotype may represent a critical disease progression stage. Our study offers new insights into AATD pathogenesis and introduces a powerful methodology for high-resolution, in situ proteomic analysis of complex tissues. This approach holds potential to unravel molecular mechanisms in various protein misfolding disorders, setting a new standard for understanding disease progression at the single-cell level in human tissue.

Methyl-coenzyme M reductase (MCR) is the enzyme responsible for nearly all biologically generated methane1. Its active site comprises coenzyme F430, a porphyrin-based cofactor with a central nickel ion that is active exclusively in the Ni(I) state2,3. How methanogenic archaea perform the reductive activation of F430 represents a major gap in our understanding of one of the most ancient bioenergetic systems in nature. Here we purified and characterized the MCR activation complex from Methanococcus maripaludis. McrC, a small subunit encoded in the mcr operon, co-purifies with the methanogenic marker proteins Mmp7, Mmp17, Mmp3 and the A2 component. We demonstrated that this complex can activate MCR in vitro in a strictly ATP-dependent manner, enabling the formation of methane. In addition, we determined the cryo-electron microscopy structure of the MCR activation complex exhibiting different functional states with local resolutions reaching 1.8-2.1 Å. Our data revealed three complex iron-sulfur clusters that formed an electron transfer pathway towards F430. Topology and electron paramagnetic resonance spectroscopy analyses indicate that these clusters are similar to the [8Fe-9S-C] cluster, a maturation intermediate of the catalytic cofactor in nitrogenase. Altogether, our findings offer insights into the activation mechanism of MCR and prospects on the early evolution of nitrogenase.

Impaired differentiation is a hallmark of myeloid malignancies1,2. Therapies that enable cells to circumvent the differentiation block, such as all-trans retinoic acid (ATRA) and arsenic trioxide (ATO), are by and large curative in acute promyelocytic leukaemia3, but whether 'differentiation therapy' is a generalizable therapeutic approach for acute myeloid leukaemia (AML) and beyond remains incompletely understood. Here we demonstrate that simultaneous inhibition of the histone demethylase LSD1 (LSD1i) and the WNT pathway antagonist GSK3 kinase4 (GSK3i) robustly promotes therapeutic differentiation of established AML cell lines and primary human AML cells, as well as reducing tumour burden and significantly extending survival in a patient-derived xenograft mouse model. Mechanistically, this combination promotes differentiation by activating genes in the type I interferon pathway via inducing expression of transcription factors such as IRF7 (LSD1i) and the co-activator β-catenin (GSK3i), and their selective co-occupancy at targets such as STAT1, which is necessary for combination-induced differentiation. Combination treatment also suppresses the canonical, pro-oncogenic WNT pathway and cell cycle genes. Analysis of datasets from patients with AML suggests a correlation between the combination-induced transcription signature and better prognosis, highlighting clinical potential of this strategy. Collectively, this combination strategy rewires transcriptional programs to suppress stemness and to promote differentiation, which may have important therapeutic implications for AML and WNT-driven cancers beyond AML.

DNA sequence-specific transcription factors (TFs) modulate transcription and chromatin architecture, acting from regulatory sites in enhancers and promoters of eukaryotic genes1,2. How multiple TFs cooperate to regulate individual genes is still unclear. In yeast, most TFs are thought to regulate transcription via binding to upstream activating sequences, which are situated within a few hundred base pairs upstream of the regulated gene3. Although this model has been validated for individual TFs and specific genes, it has not been tested in a systematic way. Here we integrated information on the binding and expression targets for the near-complete set of yeast TFs and show that, contrary to expectations, there are few TFs with dedicated activator or repressor roles, and that most TFs have a dual function. Although nearly all protein-coding genes are regulated by one or more TFs, our analysis revealed limited overlap between TF binding and gene regulation. Rapid depletion of many TFs also revealed many regulatory targets that were distant from detectable TF binding sites, suggesting unexpected regulatory mechanisms. Our study provides a comprehensive survey of TF functions and offers insights into interactions between the set of TFs expressed in a single cell type and how they contribute to the complex programme of gene regulation.

Proliferating hepatocytes often undergo ductal metaplasia to balance the energy trade-off between cellular functions and replication, hindering the expansion of human adult hepatocytes with functional competency1. Here we demonstrate that the combined activation of Wnt and STAT3 signalling enables long-term self-renewal of human adult hepatocyte organoids. YAP activation facilitates hepatocyte proliferation but commits it towards the biliary duct lineage. By contrast, STAT3 activation by oncostatin M induces hepatocyte proliferation while counteracting ductal metaplasia and maintaining the hepatic identity. Xenotransplanted hepatocyte organoids repopulate the recipient mouse liver and reconstitute the metabolic zonation structure. Upon niche factor removal and hormone supplementation, hepatocyte organoids form cord-like structures with bile canalicular networks and exhibit major liver metabolic functions comparable to those of in vivo hepatocytes. Hepatocyte organoids are amenable to gene editing, prompting functional modelling of inherent metabolic liver diseases. The new culture system offers a promising avenue for developing therapeutic strategies against human liver diseases.

Oryza rufipogon, the wild progenitor of Asian cultivated rice Oryza sativa, is an important resource for rice breeding1. Here we present a wild-cultivated rice pangenome based on 145 chromosome-level assemblies, comprising 129 genetically diverse O. rufipogon accessions and 16 diverse varieties of O. sativa. This pangenome contains 3.87 Gb of sequences that are absent from the O. sativa ssp. japonica cv. Nipponbare reference genome. We captured alternate assemblies that include heterozygous information missing in the primary assemblies, and identified a total of 69,531 pan-genes, with 28,907 core genes and 13,728 wild-rice-specific genes. We observed a higher abundance and a significantly greater diversity of resistance-gene analogues in wild rice than in cultivars. Our analysis indicates that two cultivated subpopulations, intro-indica and basmati, were generated through gene flows among cultivars in South Asia. We also provide strong evidence to support the theory that the initial domestication of all Asian cultivated rice occurred only once. Furthermore, we captured 855,122 differentiated single-nucleotide polymorphisms and 13,853 differentiated presence-absence variations between indica and japonica, which could be traced to the divergence of their respective ancestors and the existence of a larger genetic bottleneck in japonica. This study provides reference resources for enhancing rice breeding, and enriches our understanding of the origins and domestication process of rice.

The interplay between nontrivial band topology and layered antiferromagnetism in MnBi2Te4 has opened a new avenue for exploring topological phases of matter1-4. The quantum anomalous Hall effect5 and axion insulator state6 have been observed in odd and even number layers of MnBi2Te4, and the quantum metric nonlinear Hall effect7,8 has been shown to exist in this topological antiferromagnet. The rich and complex antiferromagnetic spin dynamics in MnBi2Te4 is expected to generate new quantum anomalous Hall phenomena that are absent in conventional ferromagnetic topological insulators, but experimental observations are still unknown. Here we fabricate a device of 7-septuple-layer MnBi2Te4 covered with an AlOx capping layer, which enables the investigation of antiferromagnetic quantum anomalous Hall effect over wide parameter spaces. By tuning the gate voltage and perpendicular magnetic field, we uncover a cascade of quantum phase transitions that can be attributed to the influence of complex spin configurations on edge state transport. Furthermore, we find that an in-plane magnetic field enhances both the coercive field and the exchange gap of the surface state, in contrast to that in the ferromagnetic quantum anomalous Hall state. Combined with numerical simulations, we propose that these peculiar features arise from the spin flip and flop transitions that are inherent to a van der Waals antiferromagnet. The versatile tunability of the quantum anomalous Hall effect in MnBi2Te4 paves the way for potential applications in topological antiferromagnetic spintronics9,10.

Despite the widespread use of mRNA vaccines against COVID-19, little is known about the metabolism of therapeutic RNAs. Here we use nanopore sequencing1-3 to analyse individual therapeutic mRNA molecules, focusing on their poly(A) tails. We show that the Moderna mRNA-1273 vaccine4 has a poly(A) tail of around 100 nucleotides, followed by an mΨCmΨAG sequence. In cell lines, mRNA-1273 undergoes rapid degradation initiated by mΨCmΨAG removal, followed by CCR4-NOT-mediated deadenylation. However, in medically relevant preclinical models, particularly in macrophages, mRNA-1273 poly(A) tails are extended to up to 200 nucleotides by the TENT5A poly(A) polymerase5-7, which is induced by the vaccine. Re-adenylation, which stabilizes target mRNAs, is consistently observed in synthetic mRNAs that encode proteins targeted to the endoplasmic reticulum, such as ovalbumin or antigens from Zika virus8 or the malaria parasite9. The extent of re-adenylation varies: the BioNTech-Pfizer BNT162b2 vaccine10 shows less potent re-adenylation than mRNA-1273, which correlates with a smaller proportion of membrane-associated BNT162b2. This highlights the crucial role of spatial accessibility to ER-resident TENT5A in determining re-adenylation efficiency. In vivo, TENT5A is expressed in immune cells that take up mRNA vaccine, and TENT5A deficiency reduces specific immunoglobulin production for mRNA vaccines after immunization in mice. Overall, our findings reveal a principle for enhancing the efficacy of therapeutic mRNAs, paving the way for improvement.

Structural disorder within materials gives rise to fascinating phenomena, attributed to the intricate interplay of their thermodynamic and electrochemical properties1,2. Oxygen-redox (OR) electrochemistry offers a breakthrough in capacity limits, while inducing structural disorder with reduced electrochemical reversibility3-5. The conventional explanation for the thermal expansion of solids relies on the Grüneisen relationship, linking the expansion coefficient to the anharmonicity of the crystal lattice6. However, this paradigm may not be applicable to OR materials due to the unexplored dynamic disorder-order transition in such systems7,8. Here we reveal the presence of negative thermal expansion with a large coefficient value of -14.4(2) × 10-6 °C-1 in OR active materials, attributing this to thermally driven disorder-order transitions. The modulation of OR behaviour not only enables precise control over the thermal expansion coefficient of materials, but also establishes a pragmatic framework for the design of functional materials with zero thermal expansion. Furthermore, we demonstrate that the reinstatement of structural disorder within the material can also be accomplished through the electrochemical driving force. By adjusting the cut-off voltages, evaluation of the discharge voltage change indicates a potential for nearly 100% structure recovery. This finding offers a pathway for restoring OR active materials to their pristine state through operando electrochemical processes, presenting a new mitigation strategy to address the persistent challenge of voltage decay.

Potatoes were first brought to Europe in the sixteenth century1,2. Two hundred years later, one of the species had become one of the most important food sources across the entire continent and, later, even the entire world3. However, its highly heterozygous, autotetraploid genome has complicated its improvement since then4-7. Here we present the pan-genome of European potatoes generated from phased genome assemblies of ten historical potato cultivars, which includes approximately 85% of all haplotypes segregating in Europe. Sequence diversity between the haplotypes was extremely high (for example, 20× higher than in humans), owing to numerous introgressions from wild potato species. By contrast, haplotype diversity was very low, in agreement with the population bottlenecks caused by domestication and transition to Europe. To illustrate a practical application of the pan-genome, we converted it into a haplotype graph and used it to generate phased, megabase-scale pseudo-genome assemblies of commercial potatoes (including the famous French fries potato 'Russet Burbank') using cost-efficient short reads only. In summary, we present a nearly complete pan-genome of autotetraploid European potato, we describe extraordinarily high sequence diversity in a domesticated crop, and we outline how this resource might be used to accelerate genomics-assisted breeding and research.

The formation of accessible chromatin around DNA double-strand breaks is essential for their efficient repair1. Although the linker histone H1 is known to facilitate higher-order chromatin compaction2,3, the mechanisms by which H1 modifications regulate chromatin relaxation in response to DNA damage are unclear. Here we show that CTP synthase 1 (CTPS1)-catalysed deamidation of H1 asparagine residues 76 and 77 triggers the sequential acetylation of lysine 75 following DNA damage, and this dual modification of H1 is associated with chromatin opening. Mechanistically, the histone acetyltransferase p300 showed a preference for deamidated H1 as a substrate, establishing H1 deamidation as a prerequisite for subsequent acetylation. Moreover, high expression of CTPS1 was associated with resistance to cancer radiotherapy, in both mouse xenograft models and clinical cohorts. These findings provide new insights into how linker histones regulate dynamic chromatin alterations in the DNA damage response.

The pursuit of non-volatile memory with program speeds below one nanosecond, beyond the capabilities of non-volatile flash and high-speed volatile static random-access memory, remains a longstanding challenge in the field of memory technology1. Utilizing fundamental physics innovation enabled by advanced materials, series of emerging memories2-5 are being developed to overcome the speed bottleneck of non-volatile memory. As the most extensively applied non-volatile memory, the speed of flash is limited by the low efficiency of the electric-field-assisted program, with reported speeds6-10 much slower than sub-one nanosecond. Here we report a two-dimensional Dirac graphene-channel flash memory based on a two-dimensional-enhanced hot-carrier-injection mechanism, supporting both electron and hole injection. The Dirac channel flash shows a program speed of 400 picoseconds, non-volatile storage and robust endurance over 5.5 × 106 cycles. Our results confirm that the thin-body channel can optimize the horizontal electric-field (Ey) distribution, and the improved Ey-assisted program efficiency increases the injection current to 60.4 pA μm-1 at |VDS| = 3.7 V. We also find that the two-dimensional semiconductor tungsten diselenide has two-dimensional-enhanced hot-hole injection, but with different injection behaviour. This work demonstrates that the speed of non-volatile flash memory can exceed that of the fastest volatile static random-access memory with the same channel length.

The stability of hazardous volcanic systems is strongly influenced by the uppermost magma storage depth and volatile exsolution1-3. Despite abundant evidence for an upper crustal magma reservoir beneath Yellowstone caldera4-7, its depth and the properties at its top have not been well constrained. New controlled-source seismic imaging illuminates a sharp reflective cap of the magma reservoir approximately 3.8 km beneath the northeastern caldera. Magma ascent to such low pressure is expected to drive volatile exsolution and potentially localized accumulation of bubbles near the top of the reservoir8,9, but this process typically remains hidden in contemporary volcanic systems. P-wave and P-to-S-wave reflections from the sharp top of the Yellowstone magma reservoir indicate that a mixture of supercritical fluid and magma fills the pore space at the cap of the approximately 3-8-km-deep low-shear-velocity layer imaged by seismic tomography6,7. The results are consistent with partial retention of bubbles exsolved from an upper crustal reservoir with ongoing magma supply from a volatile-enriched mantle source. Bubble accumulation can eventually lead to reservoir instability2,8, but the bubble volume fraction seismically estimated at the top of the reservoir today is lower than typical estimates of pre-eruptive conditions for rhyolites1,10,11, and measurements of the hydrothermal system document high fluxes of magmatic volatiles escaping to the surface12-15. We infer that the magma reservoir is in a stable state of efficient bubble ascent into the hydrothermal system on the basis of estimates that it is a crystal-rich (less than 30% porosity) reservoir for which dynamic modelling favours channelized bubble escape that prevents instability8.

The axion is a hypothetical fundamental particle that is conjectured to correspond to the coherent oscillation of the θ field in quantum chromodynamics1,2. Its existence would solve multiple fundamental questions, including the strong CP problem of quantum chromodynamics and dark matter, but the axion has never been detected. Electrodynamics of condensed-matter systems can also give rise to a similar θ, so far studied as a static, quantized value to characterize the topology of materials3-5. Coherent oscillation of θ in condensed matter has been proposed to lead to physics directly analogous to the high-energy axion particle-the dynamical axion quasiparticle (DAQ)6-23. Here we report the observation of the DAQ in MnBi2Te4. By combining a two-dimensional electronic device with ultrafast pump-probe optics, we observe a coherent oscillation of θ at about 44 gigahertz, which is uniquely induced by its out-of-phase antiferromagnetic magnon. This represents direct evidence for the presence of the DAQ, which in two-dimensional MnBi2Te4 is found to arise from the magnon-induced coherent modulation of the Berry curvature. The DAQ also has implications in light-matter interaction and coherent antiferromagnetic spintronics24, as it might lead to axion polaritons and electric control of ultrafast spin polarization6,15-20. Finally, the DAQ could be used to detect axion particles21-23. We estimate the detection frequency range and sensitivity in the millielectronvolt regime, which has so far been poorly explored.