Earth and Environmental Sciences

Mamenchisauridae is a group of long-necked non-neosauropodan eusauropod dinosaurs that were abundant in East Asia during the Middle to Late Jurassic, but their diversity and geographic distribution outside China remain poorly documented. Here we describe Uragasaurus kalasinensis gen. et sp. nov., a new sauropod dinosaur from the Phu Kradung Formation of northeastern Thailand. The new taxon is based on a well-preserved anterior dorsal vertebra exhibiting a distinctive combination of characters, including a unique Y-shaped configuration formed by the intraprezygapophyseal and single intraprezygapophyseal laminae and a camellate internal pneumatic structure within the centrum revealed by computed tomography (CT). Phylogenetic analyses recover the new taxon as an early-diverging member of Mamenchisauridae. This discovery represents the first formally named mamenchisaurid from Thailand and expands the known geographic distribution of the clade in Southeast Asia. The occurrence of this taxon in the Lower part of the Phu Kradung Formation also contributes to understanding faunal succession within the unit, supports an Upper Jurassic age for the lower part of the formation, and improves understanding of sauropod diversity in Southeast Asia during the Jurassic-Cretaceous transition.

Dual-atom catalysts (DACs) hold significant promise for advanced oxidation processes. However, their practical applications are often limited by sluggish electron transfer and low atomic utilization efficiency. Herein, we report that P-bridged Fe-Cu dual-atom catalysts (FeCu-NP-C) are precisely modulated by first-shell N and P ligands, forming a unique N3Fe-P1-CuN3 structure. The Fe-P-Cu bridging bond induces d-p-d gradient orbital coupling to establish a directional electron-transfer channel from the Cu site (electron donor) to the Fe site (electron acceptor) for enabling ultrafast pollutant degradation and bacterial inactivation. The FeCu-NP-C catalyst enables peroxymonosulfate activation to selectively generate high-valent iron-oxo species with a steady-state concentration of 5.80 × 10−5 mM, which is 100 times higher than that of Fe-NP-C. The FeCu-NP-C membrane reactor achieves a treatment capacity of 500 L of wastewater per gram of catalyst over 100 h, at an operational cost of USD 0.16 per tonne. This work provides deep insights into the bridge-mediated orbital interactions of DACs for water decontamination. P-bridged Fe-Cu dual-atom catalysts with N3Fe-P1-CuN3 structure enable directional electron transfer via d-p-d orbital coupling for efficient peroxymonosulfate activation. The FeCu-NP-C reactors achieve high-efficiency performance and long-term stability.

Extreme temperature variability (ETV) is a key dimension of climate risk for billions of residents. However, how urbanization shapes global ETV divergence remains unclear. Here, we assemble a 1950-2020 panel of 10,522 cities and demonstrate that ETV trajectories diverge by development status as measured by the human development index (HDI). ETV intensifies in high-development cities, whereas it slowly weakens in low-development cities, resulting in a current gap of roughly 1.66 °C. Decomposing ETV into event frequency and intensity reveals that cumulative ETV is driven mainly by intensity. Using an interpretable machine-learning framework, we find that aerosols are the urban factor most strongly associated with ETV after controlling for climate and geography. Blue-green space is consistently associated with lower ETV, whereas urban morphology has a smaller and context-dependent effect. These findings link global ETV inequality to urban governance and support targeted management that focusses on limiting volatility under climate risk.

Abstract Understanding the trophic ecology of top marine predators is crucial for assessing ecosystem structure and resilience. We investigated the feeding ecology of the common dolphin ( Delphinus delphis ) in the East Sea of Korea using stable isotope analysis of carbon (δ 13 C) and nitrogen (δ 15 N). Dolphin and prey muscle samples collected between 2020 and 2023 were analyzed to determine trophic position, habitat use, and seasonal dietary variation. δ 13 C values (−18.5‰ to −17.9‰) indicated consistent utilization of mid-shelf pelagic habitats, while significant seasonal shifts in δ 15 N values (12.3‰–13.1‰) suggested temporal variations in prey trophic levels. Bayesian mixing models identified Pacific herring ( Clupea pallasii ) as the dominant prey, contributing up to 56% in summer, whereas sand lance ( Ammodytes personatus ) and sandfish ( Arctoscopus japonicus ) were secondary and seasonally restricted prey. Isotopic niche analyses revealed broader trophic flexibility during periods of lower prey abundance (winter–fall) and narrower niches during summer, reflecting opportunistic but seasonally structured feeding behavior. These findings establish an isotopic baseline for D. delphis in Korean waters, highlighting its ecological role as a key mesopredator linking pelagic fish dynamics to higher trophic levels and underscoring the need for continued monitoring under accelerating climatic and anthropogenic changes in the East Sea ecosystem.

Urban overheating, which affects more than two billion people, is a growing problem for cities in arid and semi-arid regions. The majority of current research relies on two-dimensional satellite indices like Normalized Difference Vegetation Index (NDVI), which do not capture the vertical vegetation structures that shape thermal comfort at street level, despite the fact that urban greening has demonstrated promise as a cooling strategy. Quantitative cooling thresholds specific to hyper-arid climates are also lacking; these are the benchmarks that urban planners genuinely require in order to make well-informed design choices. To close these gaps, this study creates a predictive framework. XGBoost machine learning (R² = 0.84) and SHAP-based threshold analysis were used in conjunction with multi-dimensional vegetation metrics, including NDVI, Green View Index (GVI), and a recently developed Combined Greenery Index (CGI). The main case study is the 16 km² King Salman Garden that is being planned for Riyadh, Saudi Arabia. The analysis provides design benchmarks for cities with Köppen BWh climates by identifying critical non-linear thresholds, CGI ≥ 0.25 and GVI ≥ 25%, beyond which cooling effects increase sharply. The role of wind direction in green space planning is highlighted by model projections that show a mean daytime air temperature reduction of 3.6 °C ± 0.9 °C, extending 1.6 km downwind and 0.8 km upwind. Additionally, the findings identify an optimal range for irrigation efficiency between 65 and 75% of reference evapotranspiration (ET₀), which allows for 83% of maximum cooling with just 57% of full water use. This is a particularly pertinent finding for areas with limited water supplies. Although transferability is influenced by variations in urban form, local climate variability, and governance context, a comparative study with parks in Abu Dhabi, Dubai, and Phoenix reveals consistent threshold patterns across BWh climates. In order to maximize green infrastructure in arid cities, the framework provides urban planners with evidence-based tools.

After publication, concerns were raised regarding the XRD data presented in Figure 5. Specifically, the Fe 3 O 4 -NP and Fe 3 O 4 -PANI samples appear highly similar to each other. The raw data provided by the Authors was unable to address the concerns. The Editors therefore no longer have confidence in the reliability of the results and conclusions presented.

Accurate prediction of maximum dry density (MDD) and optimum moisture content (OMC) is critical for effective compaction control and earthwork design in geotechnical engineering. Conventional laboratory compaction tests are time-consuming and resource-intensive, motivating the adoption of reliable data-driven prediction models. In this study, a hybrid modeling framework integrating the Rao-1 metaheuristic optimization algorithm with Artificial Neural Network (ANN), Random Forest (RF), and Gradient Boosting (GB) models is proposed for predicting MDD and OMC. A dataset comprising 397 soil samples, characterized by gradation properties and Atterberg limits, is utilized for model development and validation. The Rao-1 algorithm is employed to optimize network weights and model hyperparameters, aiming to enhance convergence behavior and predictive accuracy. Comparative results reveal that Rao-1 optimization consistently enhances model performance across all algorithms and target variables. For MDD prediction, the ANN model achieves an increase in R 2 from 0.8512 to 0.9277, accompanied by a reduction in RMSE from 0.4327 to 0.2865. Similarly, the RF and GB models show notable improvements, with optimized R 2 values reaching 0.9176 and 0.9213, respectively. For OMC prediction, the Rao-1 optimized ANN exhibits the highest accuracy, improving R 2 from 0.8234 to 0.9245 and reducing RMSE from 0.4677 to 0.3071, while optimized RF and GB models also demonstrate substantial error reductions. Furthermore, SHapley Additive exPlanations (SHAP) and the Cosine Amplitude Method (CAM) were integrated to enhance model interpretability, enabling transparent evaluation of feature contributions and providing deeper insight into the influence of geotechnical parameters. Overall, the proposed Rao-1–based hybrid framework significantly enhances predictive accuracy and error minimization compared to conventional models. The results confirm the robustness and effectiveness of Rao-1 optimization in data-driven soil compaction modeling, offering a practical decision-support tool for process innovation and the preliminary estimation of compaction parameters, rather than replacing standardized laboratory testing.

Grade-control structures (GCSs) are widely used in river engineering to stabilize bed elevation, regulate sediment movement, and control channel degradation. However, conventional impermeable GCSs often generate plunging jets that intensify local scour downstream, threatening structural stability and altering channel morphology. This study aims to evaluate the morphodynamic performance of gabion grade-control structures and to quantify the effects of gabion porosity and tailwater depth on downstream scour mitigation. A series of clear-water laboratory experiments was conducted in a recirculating tilting flume, comparing a solid GCS with gabion models of three porosities (n = 0.38, 0.45, and 0.50) under unit discharges ranging from 0.023 to 0.043 m 2 /s and different tailwater depths. Temporal scour evolution and equilibrium bed morphology were measured using high-resolution bed profiling. The results showed that gabion configurations consistently reduced maximum scour depth and scour length relative to the solid structure, with reductions of up to about 38% and 44%, respectively, under the tested free-overfall conditions. Increasing porosity improved scour mitigation, although the incremental benefit became small beyond n ≈ 0.45. Greater tailwater depth further reduced scour for both structure types, and the mitigation effect was stronger when combined with gabion porosity, indicating that downstream submergence and structural permeability act together to control scour development. Gabion cases also reached equilibrium more rapidly than the solid configuration and promoted more localized deposition near the structure toe. Statistical analysis identified the densimetric Froude number as the dominant driver of scour growth, whereas tailwater depth and porosity acted as significant mitigating variables. Empirical relationships developed for relative scour depth and length showed high goodness-of-fit within the tested laboratory domain (R 2 > 0.95). Overall, the findings indicate that gabionized GCSs can improve downstream scour performance compared with impermeable structures, while the proposed equations provide laboratory-based screening tools for assessing the combined influence of porosity and tailwater depth within the calibrated range.

Accurate prediction of CO 2 injection profiles is essential for optimizing injection strategies in heterogeneous reservoirs. However, repeated compositional reservoir simulations under multiple geological and operational scenarios remain computationally prohibitive for rapid optimization and multi-scenario analysis. In addition, conventional machine learning models often struggle to jointly capture nonlinear temporal responses and static geological heterogeneity. This study proposes a deep learning surrogate framework for one-step-ahead prediction of layer-wise CO 2 injection profiles. A large-scale dataset was generated using the ECLIPSE compositional simulator under multiple geological and operational scenarios and converted into supervised time-series samples. The proposed framework integrates Bidirectional Long Short-Term Memory (Bi-LSTM), a self-attention mechanism, and Feature-wise Linear Modulation (FiLM). Bi-LSTM extracts temporal dependencies from historical injection-rate sequences, attention adaptively weights influential historical time steps, and FiLM incorporates static geological attributes by modulating learned temporal representations. Quantitative evaluations over five independent runs demonstrate that the proposed model outperforms conventional LSTM-based baselines, with an MAE of 14.28 ± 0.41 m 3 /d and an R 2 of 0.9915 ± 0.0024. Ablation studies further verify the complementary predictive contributions of all three core modules. The model also maintains stable predictive performance under different injection regimes, including continuous gas injection, Water-Alternating-Gas injection, and shut-in operations. This framework provides an efficient data-driven surrogate for accelerating multi-scenario evaluation and supporting layer-wise injection allocation optimization in CO 2 -EOR and CCUS applications.

Low-cost sensors (LCS) are increasingly used to enhance air quality monitoring by enabling dense measurement networks; however, their performance across particle size ranges and under realistic indoor-emission conditions remains uncertain, particularly in low-and middle-income settings. This study evaluated the size-resolved performance of two widely used LCS (Plantower-PMS7003 and Sensirion-SPS30) in contrasting low- and high-PM indoor environments representative of real-world conditions in India. The study’s low-PM environment exhibited 24-hour averaged concentrations of 37 ± 6.2 µg/m 3 for PM 2.5 and 39 ± 6.7 µg/m 3 for PM 10 . The corresponding concentrations were 67 ± 15.9 µg/m 3 and 119 ± 43.6 µg/m 3 , respectively, in a high-PM environment. Triplicate-LCS were collocated with a reference optical particle counter (Grimm-1.109). Analyses were done for particle number concentrations (PNC) across size bins (0.3–10.0 μm) and PM-mass fractions (PM 1 , PM 2.5 , PM 10 , intermediate fractions). Intra-sensor correlation remained high, whereas in high-PM conditions inter-sensor correlation declined but improved with hourly-averaging. Coefficient of Variation was < 30% for PM-mass and smaller PNC-bins (≤ 1.0–2.5 μm), while > 43% for larger-bins, indicating poor reproducibility. Accuracy metrics were highest for PM 1 (R 2 ~ 0.90–0.95) and decreased with increasing particle-size (PM 10 -R 2 ~ 0.28–0.68). Strong PNC-correlations were observed for 0.3–0.5 μm particles and 0.5–1.0 μm (R 2 ~ 0.72) but decreased substantially for > 1.0 μm particles (R 2 < 0.36). The evaluated sensors did not capture the dominant contribution of particles < 0.30 μm under either low- or high-PM conditions. These findings demonstrate that LCS effectively capture fine particles but shows limitations for coarse-intermediate fractions, constraining size-resolved assessments and cumulative-PM estimates based on proprietary conversions. This study provides insights to improve LCS deployment in monitoring networks and supports high-resolution exposure-assessment in resource-constrained environments.

Lassen Volcanic National Park (LVNP) contains a variety of environments such as streams, acid-sulfate and chloride-rich hot springs, fumaroles, and mud pots. This variation in geochemistry hosts a high diversity of microorganisms. However, published research on microbial communities in hydrothermally active environments at LVNP is limited, particularly using Illumina sequencing. To address this gap, we analyzed prokaryotic (16S rRNA gene) and microeukaryotic (18S rRNA gene) amplicon sequence variants (ASVs) from biofilms, mineral precipitates, sediments, spring fluids, and stream water from five thermal areas in LVNP. Prokaryotic diversity decreased with increasing temperature (28.3-90.1 °C), and the archaeal class Thermoprotei dominated the high-temperature samples. Fungal taxa within Ascomycota were the most dominant microeukaryotic group across LVNP, while the sulfur-oxidizing red alga, Cyanidioschyzon merolae, was the single most abundant microeukaryotic ASV. Cyanobacteriia and Chloroflexia were most prevalent in biofilm and sediment samples at temperatures below 72.3 °C in predominantly circumneutral conditions. Microbial network analysis revealed distinct correlations among microorganisms across five main clusters, reflecting FAPROTAX-derived functional potential for phototrophy, chemoheterotrophy, and sulfur and nitrogen cycling. These interactions point to niche separation for microbial communities, likely shaped by site- and substrata-specific environmental factors. Collectively, these findings characterize LVNP as a steep thermal and geochemical gradient in which microbial phototrophic biofilm networks, thermoacidiphilic archaea, and acid- and temperature-tolerant fungi occupy distinct microenvironments.

In the construction of permeable roads in collapsible loess regions, a critical technical conflict exists between the water permeability requirement of road surfaces for sponge city development and the collapsibility of loess. Traditional pavement treatments impair permeability or trigger loads worsening loess collapse. Low-density, permeable and high-bearing lightweight soil suits pedestrian and non-motor roads. Based on the mechanism of loess collapsibility, this study proposes a novel foundation treatment approach: lightweight soil replacement is adopted to reduce self-weight stress and overburden pressure, thereby mitigating the collapsibility of subgrades for pedestrian and non-motor vehicle roads. The effectiveness of this treatment method is validated via multi-stratum centrifugal model tests. The research indicates that the regional correction coefficient β 0 value obtained from the field immersion test was 1.56, whereas the corresponding β 0 value obtained from the multi-stratum centrifugal model test was 1.78, with a discrepancy of 0.22 between the two values. It is verified that the multi-stratigraphic centrifugal model test method can be used to obtain consistent results with the field immersion test, while possessing the distinct advantages of low cost and short test duration. Following 1.5 m of soil replacement, the site collapse value under overburden pressure was calculated to decrease by 22.73% based on laboratory tests, whereas the collapse value under overburden pressure measured by the centrifugal model test was reduced by 23.04%. These results demonstrate that lightweight soil replacement method can effectively reduce the site collapsibility. Field immersion test and the multi-stratigraphic centrifugal model test were conducted on self-weight collapsible loess in the Weicheng District of Xianyang City. The regional correction coefficient β 0 was determined, ranging from 1.56 to 1.78. This empirically derived range is more representative of the site conditions in the northern loess tableland of Xianyang than the value of 0.90 specified in the current standard for loess engineering in the Guanzhong region. Therefore, it provides a technically justified basis for selecting β 0 in high-grade loess plains across Guanzhong region in Shaanxi Province, China.

Landslide hazard assessment is a combined outcome of landslide susceptibility analysis, estimated frequency and magnitude of potential landslides. The objective of the present study is to assess landslide hazard along a section of National Highway 10 (NH 10), the lifeline of the Darjeeling-Sikkim Himalayas. To achieve this, landslide susceptibility in terms of spatial probability was first estimated by establishing the relationship between existing landslide scars and 11 geoenvironmental factors using a machine learning-based Bagging model. Secondly, after analysing landslide events and past rates of occurrence, the mean recurrence interval was computed for individual slope facets and fitted to the Poisson probability model to estimate the exceedance probability of landslides over time. In the third stage, landslide area-frequency analysis using the probability density function (PDF) and cumulative distribution function (CDF) was carried out to estimate landslide size/magnitude probabilities. Finally, after integrating spatial, temporal and magnitude probabilities, the landslide hazard was assessed for different return periods and magnitudes. It was found that annual exceedance probabilities reach up to 0.27 in the study area. The CDF shows that future landslides of ≥ 1000 m 2 and ≥ 10,000 m 2 in size have probabilities of 0.82 and 0.32, respectively. The hazard analysis depicts that over the next 10 years, there is a 37% chance that future landslides will be ≥ 1000 m 2 , and a 30% likelihood that they will be ≥ 10,000 m 2 . Moreover, Kalijhora, 7 th Mile, Birik Dara, Durbin Dara, and 29 th Mile is identified as the most hazardous sites.

Polystyrene microplastics (PS MPs) are widely distributed in aquatic environments and pose ecological and potential human-health risks. This study investigated the coagulation-assisted removal of PS MPs using MWCNTs@NH 2 in combination with okra extract as a natural coagulant aid. A Box–Behnken design was used to evaluate the effects of MWCNTs@NH 2 dose (0.2–1 g/L), initial PS concentration (50–400 mg/L), and pH (4–10), while maintaining the okra extract concentration at 0.5% (w/v). The maximum removal efficiency of 80.5% was achieved at 0.6 g/L MWCNTs@NH 2 , 400 mg/L PS, and pH 10. The removal mechanism involved partial charge neutralization, interparticle bridging by okra polysaccharides, hydrophobic interaction between PS and the graphitic CNT backbone, and floc enmeshment. The combined use of MWCNTs@NH 2 and okra extract reduced the required nanomaterial dose and improved floc formation. The proposed hybrid coagulation system offers a natural-coagulant-assisted strategy for PS MP removal, although further validation in real water matrices and assessment of residual nanotube release are required.

The EAT-Lancet diet (ELD) is recognized as a sustainable dietary pattern that supports both human health and environmental well-being. However, the relationship between the ELD and infant growth metrics is still poorly understood. This study explored the association between the ELD, plant-based dietary indices, and various infant growth parameters, including weight, length, head circumference, and ponderal index. We recruited 326 healthy mother-infant pairs for this study. The mothers' dietary intake was assessed using a semi-quantitative food frequency questionnaire (FFQ), and the ELD was calculated based on the Kesse-Guyot score. The infants' measurements, including weight, length, head circumference, and ponderal index, were taken between 2 and 6 months old. The average age of mothers with the highest adherence to the ELD was 28.90 years (± 6.06), while those with the highest adherence to the plant-based diet index (PDI) averaged 29.31 years (± 5.52). Infants of mothers in the highest ELD adherence group (third tertile) had a 13.8% increase in head circumference compared to those in the lowest group (first tertile) [OR: 1.138, 95% Cl: 1.026, 1.262, p = 0.015]. There was no significant correlation between plant-based diet indices and infant growth metrics. Our findings indicate that ELD adherence was associated with increased infant head circumference. In contrast, plant-based dietary indices did not exhibit a similar relationship. This suggests that ELD may offer distinct advantages in promoting infant growth, possibly due to the inclusion of small amounts of animal-based products that enhance the safety of the maternal diet compared to strictly plant-based dietary patterns. Future research should explore these relationships using various indicators for measuring infant growth.

<title>Abstract</title> This study presents a comprehensive molecular dynamics (MD) investigation of braid-reinforced hollow fiber membranes (BRHFMs) to elucidate the interfacial and mechanical behaviors of polymeric composites composed of cellulose acetate (CA) and polyacrylonitrile (PAN). The analysis focuses on three representative configurations—homogeneous (CA/CA-II), semi-heterogeneous (PAN|CA/CA-I), and heterogeneous hybrid (CA/PAN-III)—to evaluate their interfacial energies, adhesion mechanisms, and tensile responses. The calculated interfacial energies of − 1.20 eV, − 1.45 eV, and − 1.61 eV for CA/CA-II, PAN|CA/CA-I, and CA/PAN-III, respectively, reveal that chemical homogeneity promotes stronger interfacial bonding, whereas polarity mismatches between functional groups (–OH, –OCOCH₃, and –CN) weaken adhesion and increase diffusivity at the interface.Mechanical testing through MD tensile simulations further demonstrates that the CA/PAN-III composite exhibits pronounced stress fluctuations and higher local interfacial activity. At the same time, the CA/CA-II system maintains the highest cohesive stability and elastic modulus due to structural uniformity. The CA/PAN-III hybrid achieves an optimal balance between flexibility and strength, indicating its suitability for water treatment membranes requiring both mechanical resilience and interfacial durability. These findings provide molecular-level insight into how polymer compatibility governs the performance of BRHFMs and offer valuable guidelines for designing next-generation high-strength composite membranes.

Invasive species can have devastating effects when introduced into remote island ecosystems, and a fundamental aspect of this concerns the diet of these exotic taxa. Here, we employed a DNA metabarcoding approach to determine the diet of the lizard Agama picticauda on Réunion Island, where it was introduced in 1995. Two separate markers were used to identify dietary components: COI for animals and trnL for plants. The arthropod aspect was notably conservative, with the agama continuing to predominantly consume ants, as they do in their native range. A variety of other invertebrates were also preyed upon, the vast majority being introduced species. For plants, again a wide variety was detected, and while most could not be identified fully, it seems that agamas are deliberately consuming many species, rather than accidentally ingesting them along with targeted invertebrates. Agamas may play a role in seed dispersal of invasive plant species. We also detected some nematode groups, although with limited comparative sequences, these could not be identified to the species level. Several invertebrate records appear to be new for Réunion Island, highlighting how reptiles can be considered as excellent biodiversity samplers, with barcoding diet studies providing novel data on poorly known invertebrate groups. The minimal identification of endemic prey items may reflect the fact that agamas are still predominantly occupying anthropogenically disturbed parts of the island. Our study therefore provides baseline data that can be used to determine the impact of this introduced lizard as it spreads through the ecosystem.

Acute myeloid leukaemia (AML) is an aggressive blood cancer characterized by the unregulated proliferation of immature myeloblasts. Gene mutations have been shown to have a large effect on pathogenesis, inter-tumour heterogeneity and clinical outcomes in AML1–8; however, the role of epigenetic alterations in these respects has been investigated less extensively. Here we use ATAC-seq (assay for transposase-accessible chromatin with sequencing) in a cohort of 1,563 individuals with a recent diagnosis of AML (the ‘eCHROMA’ cohort) to show that AML can be classified into 16 subgroups on the basis of chromatin accessibility profiles. Multiomics analyses of gene mutations, the transcriptome, DNA methylation and histone marks show that these ATAC subgroups exhibit distinct driver mutations, differentiation states, gene expression, DNA methylation and super-enhancer profiles, and are also associated with clinical outcomes. These findings were validated in independent cohorts. Single-cell ATAC sequencing reveals that all leukaemic cells in each subgroup share a common chromatin accessibility profile, which suggests that subgroup-specific epigenomic fingerprints underlie the ATAC-based classification. Mechanistically, the subgroups have distinct gene-regulatory networks that are driven by the activities of key transcription factors in haematopoiesis, and in which subgroup-specific super-enhancers have a pivotal role. Multiomics single-cell analysis further reveals deregulated trajectories of differentiation coupled with chromatin accessibility and gene expression. Notably, ATAC subgroups have an independent prognostic effect, compared with genomic classification, and are associated with particular drug sensitivities. In summary, ATAC-based chromatin profiling, combined with multiomics data, provides insights into AML pathogenesis beyond genomics and constitutes a valuable resource for AML research. An ATAC-seq-based approach is used to classify acute myeloid leukaemia (AML) into 16 epigenomic subgroups, and provides insight into the role of non-genetic mechanisms in determining pathogenesis, clinical behaviour and drug sensitivity in this disease.

The short-term and long-term effects of genotoxic pre-transplant conditioning remain barriers to the broader application of haematopoietic stem/progenitor cell (HSPC) transplantation and gene therapies1–4. Although monoclonal antibodies targeting KIT have been proposed as alternatives to chemotherapy or radiotherapy5–7, their pharmacokinetics hinder clinical applications owing to the risk of depleting transplanted HSPCs. Here, to address this issue, we identified amino acid changes in the extracellular domain of KIT that disrupt the binding of two therapeutic monoclonal antibodies8,9, which impair stem cell factor (SCF)-mediated signalling without affecting KIT expression or functionality. We exploited adenine base editing10 or prime editing11 to efficiently introduce these mutations in HSPCs and combined them with the disruption of the BCL11A erythroid enhancer to promote expression of fetal haemoglobin (HbF)12,13, a therapeutic approach for several haemoglobinopathies. This strategy enables in vivo co-selection of gene-engineered cells to reach the threshold required to provide therapeutic benefit in patients affected by sickle cell disease and β-thalassaemia. We show progressive enrichment of KIT plus BCL11A multiplex-edited haematopoiesis under selective pressure with KIT monoclonal antibody, in vitro and in vivo. We report that extended treatment with anti-KIT regimens leads to superior in vivo enrichment while avoiding clonal selection, as assessed by a lentiviral barcoded library. Finally, by overcoming the limitations of monoclonal antibody pharmacokinetics, epitope editing enables novel haematopoietic replacement regimens that are not limited by on-target graft elimination, allowing prolonged immune-based conditioning that maximizes haematopoietic niche clearance without chemo-radiotherapy or monoclonal antibody wash-out. Epitope editing of KIT enables antibody-based, non-genotoxic conditioning that selectively enriches therapeutic BCL11A-edited haematopoietic stem/progenitor cells, supports durable engraftment, preserves clonal diversity and enhances induction of fetal haemoglobin, a therapeutic approach for conditions such as sickle cell disease and β-thalassemia.

Physiological host factors, such as the gut microbiome and obesity, independently influence anti-tumour immunity and responses to immune checkpoint inhibitors (ICIs)1, with high body mass index (BMI) having an unexpected link with greater ICI efficacy2–6. However, how these factors interact across diverse dietary contexts remains unclear. Here, using 12 mouse diet models that reflect a spectrum of obesity biology, we characterize diet-driven metabolic, immune and gut microbiota features associated with ICI sensitivity. We find that obesity-associated ICI responses are poorly correlated with metabolic dysfunction and are instead dependent on the diet–gut axis. Obesogenic diets promote a robust and persistent gut microbial ecosystem that is capable of restoring ICI sensitivity following a short-term diet switch or fecal microbiota transplants (FMTs) from non-responder models. Monocolonization of germ-free mice with favourable bacteria such as Lactobacillus johnsonii, together with an obesogenic diet, synergistically promotes tumour regression through an enrichment of microbiota-derived aromatic amino acid metabolites. Moreover, human-to-mouse FMT from donors with a high BMI enhanced ICI efficacy compared with donors with a normal BMI, and an obesogenic diet restored sensitivity following FMT from a non-responder patient. Our study provides insight on epidemiological associations between BMI and ICI efficacy, and suggests that immunomodulatory synergy between diet and the gut microbiota could be leveraged to improve ICI outcomes and FMT interventions. Diet shapes obesity-associated therapeutic responses to immune checkpoint inhibitors through gut microbial metabolism and host anti-tumour immunity, demonstrated in mouse custom-diet models and human-to-mouse fecal microbiota transplantation experiments.

Hydrogels are widely used in biomedical interfaces, in which effective gas exchange (for example, O2, CO2) within a water-rich environment is essential. However, hydrogels show intrinsically limited air exchange efficiency, owing to the low solubility (C) and diffusivity (D) of non-polar gases in the polar water medium1. This limitation poses a substantial bottleneck in long-term applications, such as wearable health monitors2–7 and tissue engineering8–12. Existing methods13–16 to enhance air permeability suffer from poor robustness and/or an inherent trade-off between permeability and water content (for example, <50 vol%). Here we introduce a viscoelastic phase separation17 (VPS)-enabled strategy to create a non-collapsible, air-rich network in high-water-content hydrogels, achieving a record-high oxygen permeability of 185 barrer with 70 vol% water—a tenfold increase compared with pristine hydrogels. VPS, a ubiquitous phenomenon in soft matter, is used to drive hydrophobic, dry gas particles within a hydrophilic, wet medium into a thin, stable three-dimensional network. This approach allows the facile and scalable fabrication of air-permeable hydrogels across diverse chemistries and form factors. Physiological tests over a 10-day continuous wear condition confirmed their effectiveness in preventing fluid accumulation and maintaining skin health. This strategy paves the way for hydrogels in long-term biomedical applications in which efficient and sustained air exchange becomes critical. Viscoelastic phase separation is used to fabricate non-collapsible, air-rich networks in high-water-content hydrogels containing silica aerogel beads, allowing air to permeate through the material and enabling a tenfold increase in oxygen permeability over pristine hydrogels.

Cancer genomes are unstable, often experiencing gains or losses of whole chromosomes or chromosome arms — changes known as aneuploidies. Experiments in mouse models of breast cancer reveal that these instabilities harbour one or two prevalent genes that drive cancer and that bypass the need to accumulate aneuploidies; they also require an intact microenvironment to produce their effects. Altered expression of many drivers identified in the in vivo CRISPR screen did not alter cell behaviour in vitro.