Epstein-Barr virus (EBV) infects approximately 90-95% of the global population1,2 and persists in B cells as a lifelong infection3. Previous EBV infection is associated with autoimmune and neoplastic disease4. Still, the biological basis of host control during EBV persistence remains unclear. Here we report the identification of non-genetic and genetic factors that are associated with EBV control during persistent infection. Using blood-based genome sequence data from 486,315 UK Biobank and 336,123 All of Us participants, we identified short-read pairs mapping to the EBV genome in 16.2% and 21.8% of individuals, respectively. EBV read detection (EBVread+) reflects increased viral load in blood cells, as shown by orthogonal measurements, and was associated with HIV infection, immunosuppressive drug intake and current smoking. Genome-wide analyses of EBVread+ identified strong associations at the major histocompatibility complex (MHC), including 54 independent human leukocyte antigen (HLA) alleles of MHC classes I and II, and at 27 genomic regions outside MHC. Epistasis with distinct HLA alleles of MHC class I was observed at the ERAP2 locus. Analysis of individuals with EBV-associated diseases4 revealed a higher polygenic burden of EBVread+ for HLA alleles at MHC class I in multiple sclerosis (driven by HLA-A*02:01) and at MHC class II in rheumatoid arthritis. Phenome-wide analyses identified a polygenic overlap of EBVread+ with inflammatory bowel disease, hypothyroidism and type 1 diabetes. Our study establishes by-products of human genome sequencing as a surrogate marker of EBV viral load. This will facilitate investigation and treatment for EBV and other persistent viral infections.

Anthropogenic nitrogen (N) pollution is a cause of eutrophication globally1. However, recent datasets indicate that some ecosystems may be experiencing widespread oligotrophication-declining N availability-which is suggested to be a response to elevated atmospheric carbon dioxide (CO2)2. Plant N isotope (δ15N) chronologies have served as primary evidence for oligotrophication, but there is wide disagreement whether rising CO2 or temporal changes in N deposition explain these patterns3-6. Here we construct δ15N tree-ring chronologies using archived samples from Sweden's 23.5-million-hectare forest area from 1961 to 2018. The study area spans a 1,500-km latitudinal distance where N deposition varies fourfold, but where rising CO2 is spatially uniform. Our data show declining δ15N chronologies throughout Sweden, including forests in the far north where atmospheric N deposition rates are very low. Linear mixed-effects models showed that rising CO2 is the strongest predictor of δ15N values, whereas N deposition variables, temperature and forest basal area had lower explanatory power. Our findings suggest that elevated atmospheric CO2 is causing oligotrophication in boreal forests, which has implications for predicting their future role as sinks in the global carbon cycle7-9.

The question of whether large language models (LLMs) can exhibit moral capabilities is of growing interest and urgency, as these systems are deployed in sensitive roles such as companionship and medical advising, and will increasingly be tasked with making decisions and taking actions on behalf of humans. These trends require moving beyond evaluating for mere moral performance, the ability to produce morally appropriate outputs, to evaluating for moral competence, the ability to produce morally appropriate outputs based on morally relevant considerations. Assessing moral competence is critical for predicting future model behaviour, establishing appropriate public trust and justifying moral attributions. However, both the unique architectures of LLMs and the complexity of morality itself introduce fundamental challenges. Here we identify three such challenges: the facsimile problem, whereby models may imitate reasoning without genuine understanding; moral multidimensionality, whereby moral decisions are influenced by a range of context-sensitive relevant moral and non-moral considerations; and moral pluralism, which demands a new standard for globally deployed artificial intelligence. We provide a roadmap for tackling these challenges, advocating for a suite of adversarial and confirmatory evaluations that will enable us to work towards a more scientifically grounded understanding and, in turn, a more responsible attribution of moral competence to LLMs.

Long-term preservation of digital information is vital for safeguarding the knowledge of humanity for future generations. Existing archival storage solutions, such as magnetic tapes and hard disk drives, suffer from limited media lifespans that render them unsuitable for long-term data retention1-3. Optical storage approaches, particularly laser writing in robust media such as glass, have emerged as promising alternatives with the potential for increased longevity. Previous work4-16 has predominantly optimized individual aspects such as data density but has not demonstrated an end-to-end system, including writing, storing and retrieving information. Here we report an optical archival storage technology based on femtosecond laser direct writing in glass that addresses the practical demands of archival storage, which we call Silica. We achieve a data density of 1.59 Gbit mm-3 in 301 layers for a capacity of 4.8 TB in a 120 mm square, 2 mm thick piece of glass. The demonstrated write regimes enable a write throughput of 25.6 Mbit s-1 per beam, limited by the laser repetition rate, with an energy efficiency of 10.1 nJ per bit. Moreover, we extend the storage ability to borosilicate glass, offering a lower-cost medium and reduced writing and reading complexity. Accelerated ageing tests on written voxels in borosilicate suggest data lifetimes exceeding 10,000 years.

In the past decade, moiré materials have revolutionized how we engineer and control quantum phases of matter1,2. They are versatile platforms for strongly correlated electronic phenomena3,4 and support new ferroelectric5,6, magnetic7 and superconducting states8. Among incommensurate materials9, moiré materials are aperiodic composite crystals10,11 whose long-wavelength superlattices enable tunable properties without chemically modifying their layers. So far, nearly all reports of moiré materials have investigated van der Waals heterostructures assembled far from thermodynamic equilibrium (T < 150 °C)1,2. Here we introduce a conceptually new approach to synthesizing high-mobility moiré materials in thermodynamic equilibrium. We report a new family of foliated superlattice materials (Sr6TaS8)1+δ(TaS2)8 that are exfoliatable, incommensurate-lattice, van der Waals crystals. Lattice mismatches between alternating layers generate moiré superlattices, analogous to 2D moiré heterobilayer superlattices, which are coherent throughout these crystals and tunable through synthesis conditions without altering their chemical composition. Quantum oscillation measurements map the complex Fermiology of these moiré metals12-14, showing that the Fermi surface of the structurally simplest moiré metal comprises more than 40 distinct cross-sectional areas. This is naturally understood by proposing that these bulk moiré metals encode electronic properties of higher-dimensional superspace crystals in ways paralleling well-established crystallographic methods for incommensurate lattices15,16. More broadly, our work demonstrates a scalable synthesis approach potentially capable of producing large-area moiré materials for electronics applications and evidences a new material design concept for accessing phenomena proposed in higher dimensions17-21.