The Arctic is undergoing rapid warming, resulting in retreating sea ice and glaciers1, yet how cryospheric changes propagate into the deep ocean remains poorly understood2. Here we identify a climate-driven mechanism linking accelerating glacier disintegration to an increase in deep-sea hard-bottom habitats far beyond calving fronts. Seafloor observations in Fram Strait show a localized increase in the density and patchiness of dropstones delivered by debris-laden icebergs. At the same time, four decades of shipboard records show that the occurrence of icebergs increased abruptly in the early 2000s. Backtracking links these icebergs to the main outlet glaciers in northeast Greenland and the Russian High Arctic. In northeast Greenland, the timing of glacier destabilization coincides with this rise, whereas sparse satellite coverage in the Russian sector limits temporal attribution despite indications of enhanced glacier activity. A model sensitivity study shows that, apart from intensified calving, a more dynamic sea ice cover enhances downstream transport of glacial ice. Along these pathways, increased iceberg activity could reshape deep-sea habitats through enhanced melt and associated lithogenic input, and elevate navigational hazards as maritime traffic expands in the Arctic. Although modest compared with the iceberg discharges of Pleistocene Heinrich events, this mechanism provides a modern analogue of long-range cryospheric influence on the seafloor in a warming climate.
Antarctica is an extreme region in many ways and long term marine ice monitoring shows 2025 was the most of those extremes. Seasonal sea ice duration was observed daily at Rothera Research Station, Antarctica since 1985 and seabed markers were monitored for iceberg scour hits since 2002. Two decades of data showed that sea ice duration showed a strong inverse relationship with iceberg scouring. In 2025 sea ice duration was zero days, a record low and 68% of seabed markers were hit by icebergs, a record high. Long term monitoring data showed that ice scour responses to benthos were most significant after a two year lag period. Sustained seasonal sea ice losses in a warming world suggest these extremes may become normal in both levels in the coming decades and assemblages may take decades to recover and became pioneer dominated.
Business analysts and non-technical users need insights from enterprise data lakes but lack SQL expertise to query them directly. While large language models (LLMs) can translate natural language to SQL, existing text-to-SQL approaches face critical limitations: severe SQL injection vulnerabilities, inability to leverage data-lake-specific features like time-travel queries, and inconsistent metric definitions across organizations. We present the LangChain Iceberg Toolkit, enabling users to query Apache Iceberg data lakes through natural language conversations with LLM agents, no SQL knowledge required. Users ask questions in plain English (e.g., "What was revenue last quarter?"), and the system automatically: (1) interprets intent using LLMs, (2) selects appropriate tools from a YAML-based semantic layer mapping business terms to data structures, (3) executes queries through a hybrid architecture combining PyIceberg's type-safe API (for security) with DuckDB's SQL engine (for complex analytics), and (4) returns formatted answers with business context. Our evaluation demonstrates 100% success across 100 systematically designed queries leveraging semantic layer integration for consistent metric definitions. Critically, in direct comparison against a schema-aware text-to-SQL baseline on the same query set, our system achieves a 33 percentage-point accuracy improvement (100% vs. 67%) while reducing SQL injection attack success rate from the 99% reported in prior text-to-SQL research to 0% across both execution paths. End-to-end query latency averages 2.6 seconds on 15.1M records, with partition pruning eliminating 90%+ of scanned data files. The hybrid execution architecture prevents SQL injection vulnerabilities through type-safe query construction for simple queries and controlled, pre-validated SQL execution for complex analytics. Users receive data insights through conversational interfaces without writing SQL, understanding schemas, or knowing technical implementation details. We provide a production-ready, open-source implementation demonstrating practical viability for democratizing enterprise data access.
Despite public campaigns, pre-hospital delays in acute ischemic stroke remain substantial, with only modest improvement in onset-to-door times. We ran a nationwide cross-sectional survey on ischemic stroke knowledge across four domains: conceptual knowledge, symptom recognition, awareness of time-dependent treatment, and intended emergency action (immediate EMS/112 call). A composite score (≥ 3 correct) was analysed using contingency tests and multivariable logistic regression. Among 1,769 participants, 71% were female, 60% university-educated, 35% were > 40 years, 9% worked in healthcare, and 47% reported a family history of stroke. Only 25% identified "cerebral stroke" and "cerebral ischemia" as synonyms, and 52% rated stroke as severe but ischemia as milder; accuracy was higher in healthcare workers (45% vs 23%, p < 0.001) and higher education (29% vs 22%, p = 0.002). One quarter failed to identify unilateral paralysis as the cardinal symptom, with lower recognition in those aged ≥ 41 years (70-73% vs ~ 77-79%, p = 0.019). Treatment awareness was high (84%) and associated with higher education/healthcare-related exposure (p = 0.007). Intended EMS/112 call was reported by 85% and associated with younger age (p = 0.004) and indirect stroke-care exposure (p = 0.001). Overall, 65% achieved a high composite score; healthcare employment independently predicted a high score (OR 2.02, 95% CI 1.28-3.19), whereas age and educational level were not independently associated. Findings support an iceberg model of stroke literacy, with surface awareness masking deeper, phase-specific gaps; education should target stroke severity misconceptions and symptom recognition, particularly in older adults.
Spinal cord injury (SCI) has traditionally been viewed as a focal trauma to the central nervous system, with research focusing primarily on the disruption of motor and sensory pathways. However, emerging evidence reveals that SCI rapidly induces systemic changes that extend far beyond the spinal cord. This review synthesizes current evidence to conceptualize SCI as a comprehensive systemic disorder, utilizing the "neuro-immune-metabolic network" as a central explanatory framework. Following the initial injury, the disruption of descending autonomic pathways contributes to widespread physiological dysregulation, driving peripheral organ dysfunction such as cardiovascular aberrations and splenic atrophy. Concurrently, systemic inflammation and metabolic disturbances, characterized by the release of pro-inflammatory cytokines, hepatic steatosis, and gut microbiota dysbiosis, generate circulating mediators that subsequently exacerbate central neuroinflammation. This bidirectional crosstalk establishes a self-perpetuating pathogenic cycle in which peripheral deterioration may further impede intrinsic neural repair. By mapping these multi-organ pathological interactions, this review underscores the limitations of isolated spinal interventions. Ultimately, future therapeutics should extend beyond localized spinal repair to target the restoration of systemic physiological networks, thereby advancing clinical management toward comprehensive, multi-targeted interventions.
暂无摘要（点击查看详情）
暂无摘要（点击查看详情）
Heterogeneity of treatment effect has yielded decades of negative critical care trials. Syndromic diagnoses like sepsis and acute respiratory distress syndrome mask distinct molecular programs that respond differently to the same intervention, and single biomarkers lack the resolution to capture this complexity. Recent evidence now demonstrates that each step of the enrichment pipeline, from real-time bedside endotyping to prospective endotype-matched therapy, is clinically operational. However, current approaches rely on limited biomarker panels that capture only surface-level biology. Pathway biology, examining coordinated dysregulation of molecular networks rather than isolated analytes, offers the deeper resolution needed to match patients to targeted therapies. We propose a translational pipeline integrating multiomics, pathway analysis, machine learning, and point-of-care assays to advance critical care toward pathway-focused predictive enrichment.
暂无摘要（点击查看详情）
暂无摘要（点击查看详情）
Polycystic ovary syndrome (PCOS) is the most common endocrine disorder in reproductive-age women, but its true burden remains underestimated due to diagnostic variability and underreporting. To estimate the morbidities in reproductive-age women identified with a diagnosis of PCOS in the French national health data system. We conducted a retrospective cohort study using the French national health data system (SNDS) to identify reproductive-age women (18-43 years) diagnosed with PCOS (ICD-10 code E28.2) between 2014 and 2022. Morbidity data were assessed in 2022 and compared with an age-matched control group of women without PCOS in the same year. Specific comorbidities were identified using a set of algorithms applied to the system's data and based on both ICD-10 codes and medical acts and drug use. Logistic regression models were used to compare comorbidity rates, with odds ratios adjusted for age and social deprivation. We identified 21,144 women with PCOS in the health data system. Median age was 35 years [IQR: 29-39]. Compared with the general female population, women with PCOS had significantly higher rates of infertility (adjusted odds ratio (aOR)=24.35), gestational diabetes (aOR=1.64), type 2 diabetes (aOR 14.71), pulmonary embolism (aOR=2.04), arterial hypertension (aOR 5.86), acute stroke (aOR= 2.39), cardioneurovascular disease (aOR 2.12) and psychiatric disorders (aOR = 1.66). Given the high morbidity among women identified with PCOS on ICD-10 code E28.2, improved diagnostic algorithms are needed to better identify PCOS cases in health databases and assess their public health implications.
Nonprimary maternal cytomegalovirus (CMV) infections, resulting from reactivation or reinfection in seropositive women, are increasingly recognized as contributors to congenital CMV (cCMV) disease. However, their clinical impact compared with primary infections remains insufficiently characterized. We conducted a retrospective cohort study of symptomatic cCMV identified from a single-center series (2005-2022) and a multicenter registry (2023-2025). Infants were included if they had a positive urine CMV polymerase chain reaction within the first 3 weeks of life and met criteria for symptomatic infection based on central nervous system involvement. Maternal infection type (primary vs. nonprimary) was determined by serologic testing. Clinical characteristics, neuroimaging findings and auditory outcomes were compared between groups. We identified 360 symptomatic infants; maternal infection type was unknown in 55, leaving 305 for analysis [243 (79.7%) primary; 62 (20.3%) nonprimary]. The proportion of nonprimary infections among symptomatic cases increased from 11.4% in 2005-2009 to 37.3% in 2020-2025 (odds ratio: 4.77; 95% confidence interval: 1.63-13.92; P = 0.004). Absolute counts mirrored these trends: nonprimary 5 of 44 in 2005-2009 to 22 of 59 in 2020-2025. Clinical manifestations, including neuroimaging and hearing outcomes, were broadly similar between groups. In the era of maternal screening and interventions focused on primary CMV infection, nonprimary infections represent a rising share of symptomatic cCMV and a relevant clinical burden. These findings highlight the importance of recognizing nonprimary infections in clinical care and public health planning and support the need to reassess current strategies for maternal counseling and neonatal care.
Natural disturbance events are expected to increase with climate change. This is particularly evident in polar regions where reduced winter sea-ice has increased movement of icebergs, and thus seafloor scouring. This predominantly affects shallow near-coast habitats, where iceberg scours can decimate local benthic ecosystems. Various metrics can be employed to measure recovery and resilience of ecosystems affected by disturbance. Here, we build a food web model for a near-shore benthic ecosystem along the West Antarctic Peninsula to evaluate the ecosystem's response to iceberg scouring and predict its response to increasing future impacts. The overall food web structure was consistent with other Antarctic benthic food webs with a low mean trophic level and connectance, a high degree of omnivory and similar average path length compared to more temperate systems. We show that chronically disturbed shallow (5 m) habitats had lower food web complexity than deeper (10-25m) habitats where disturbance intensity and frequency was reduced. Recovery with time since last disturbance showed that recently disturbed food webs had similar levels of complexity to those that had been undisturbed for 8+ years but complexity fell to a minimum after 2 years of disturbance before recovering gradually. This might be due to the influx of mobile scavengers immediately post scour. We identified some highly connected species within the food web that are found across most of the Antarctic coastal shallows, such as the starfish Odontaster validus, and conclude that these may be key to maintaining resilience in these ecosystems with increasing climate change.
Most of what we know about neural mechanisms of incremental learning through feedback comes from descriptive, univariate analyses. Here, we go one step further, seeking brain activity that is not just statistically reliable (potentially small but significant) but can track such learning at the item level, taking a classifier-based approach to narrow in on basic neural encoding processes. Participants ( N = 45 ) learned 48 word-value mappings through trial-and-error. First, we checked whether established EEG markers of feedback processing, the feedback-related negativity (FRN) and frontal midline theta activity (FMT), are in fact predictive of trial-to-trial learning of the current item-and they were (above chance, but not by much), validating the behavioural relevance of those features. Next, we asked whether there might be considerably more information about encoding on single trials beyond these statistically robust, regular signals. Indeed, multivariate classifiers (LDA and SVM), incorporating signal-features beyond the FRN and FMT, predicted learning more substantially and exceeded previous performance on episodic recognition using the same basic approach (Chakravarty et al., Journal of Neurophysiology, 124(6), 2060-2075, 2020). Time-frequency spectral features produced better classifications (AUC ∼ 0.7) than time-domain features. Finally, a possible shortcut due to accuracy varying systematically with trial number could not explain away classification success. In sum, FRN and FMT are not just descriptive of feedback-driven learning but also a bit predictive-but are the tip of the iceberg (subject-specific, spatiotemporal features) uncovered by the multivariate classifiers. This extends current classifier-based approaches to brain activity from episodic memory to incremental, feedback-driven learning.
Heinrich Stadials provide insights into a potential future weakening of the Atlantic Meridional Overturning Circulation (AMOC), yet the mechanisms driving AMOC weakening during these intervals remain elusive. While Heinrich Stadials are associated with iceberg-discharge events over the North Atlantic (i.e., Heinrich events), AMOC weakening is known to precede Heinrich events. Here we run freshwater hosing experiments to show that contemporaneous Northeast Pacific iceberg-discharge events (i.e., Siku events), which consistently occur at the onsets of Heinrich Stadials, could trigger AMOC weakening by transporting freshwater to the North Atlantic deepwater formation regions. This initial weakening may subsequently be amplified by further meltwater release from the British and Laurentide Ice Sheets, induced by stepwise subsurface warming in the northeastern and northern North Atlantic. Our findings suggest that Siku events may have played a critical role in preconditioning the North Atlantic for Heinrich events, underscoring the interconnected nature of the global ocean-cryosphere system.
The mechanisms by which the gut microbiota influence host disease outcomes remain poorly understood. As an integral yet overlooked component of this microbial community, the role of gut commensal protists in host physiology and pathology is even more ambiguous. Here, we show that the protozoan Tritrichomonas musculis (T. mu) remotely increases host sensitivity to drug-induced liver injury via the production of free sphingosine. Inhibiting sphingosine kinases (SPHKs) with PF543 or K145, or antagonizing sphingosine 1-phosphate receptor (S1PR) with FTY720, abolished the effect of T. mu-mediated exacerbation of acetaminophen (APAP)-induced liver injury (AILI), pointing to a sphingosine-SPHK1/2-S1P-S1PR axis underlying T. mu's remote modulation of hepatic drug sensitivity. Moreover, using U-¹³C‑palmitic acid and U-¹³C‑glucose metabolic flux tracing, we propose a novel model for sphingosine synthesis in T. mu. In this model, T. mu synthesizes sphingosine by using oxaloacetate, likely generated via the glyoxylate cycle, as a two‑carbon unit donor, rather than directly employing intact palmitic acid to supply the long‑chain carbon skeleton. Collectively, these findings not only deepen our understanding of how the gut microbiota, particularly the underappreciated gut protists, influence extra-intestinal disease outcomes via the gut-distal organ axis, but also reveal the tip of the iceberg regarding the unique metabolic pathways of gut commensal protists.
In vitro bioassays indicative of activation of the aryl hydrocarbon receptor (AhR) are commonly applied to water, sediment and biota. Many equate AhR activity in water to dioxin-like toxicity, but hydrophobic dioxin-like chemicals will not be present in water extracts. Aryl hydrocarbon receptor activity has been observed in wastewater, surface water, drinking water and recycled water, with a diverse range of environmental contaminants, including pesticides, pharmaceuticals and industrial chemicals, contributing to the observed AhR activity. Therefore, this critical perspective advocates for a change in mindset from dioxin-like toxicity to sensitive indicator of chemicals that activate xenobiotic metabolism when AhR assays are applied to water samples. Firstly, we proposed a new environmentally relevant reference compound to express AhR activity in water to replace widely used hydrophobic reference compounds such as 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and benzo(a)pyrene. The fluorescent dye 7-diethylamino-4-methylcoumarin (DEAMC) was identified as a suitable water-soluble reference compound after considering occurrence in wastewater, common mixture effect drivers and specificity. Iceberg modeling, where the predicted mixture effect is compared to the measured effect, demonstrated that very different chemicals activate AhR in water and sediment, supporting the need for a water-soluble reference compound. Effect-based trigger values (EBT) differentiate between acceptable and poor chemical water quality based on the bioassay response, with EBTs for AhR assays primarily derived for surface water to protect ecological health. However, the detection of AhR activity in drinking water and recycled water supports the need for robust EBTs for drinking water to protect human health. We derived a new human EBT (humanEBT) using established read-across methods and applying a mixture factor based on the fraction of AhR activity explained in water. Overall, this perspective clearly demonstrates why we need to look beyond dioxin-like toxicity when applying activation of AhR assays to water samples.
Accounts of menopause across media, policy and workplaces have begun to highlight the relationship between menopause and the economy. However, embedded and emerging socioeconomic dynamics also underpin experiences of menopause in ways that are often overlooked. This commentary introduces three intersecting socioeconomic concepts shaping the contemporary experience of menopause. First, the 'economic iceberg of menopause' highlights how gendered patterns, including cumulative disadvantage, insecure employment, and financial vulnerability, converge during midlife to shape work participation and social and economic outcomes for those experiencing menopause. Second, the increasing reach of 'menomarkets' shows how market-driven ecosystems shape expectations and influence health-seeking behaviour. Finally, the concept of 'menocynicism' is introduced to describe cycles of scepticism surrounding menopausal knowledge and support. This scepticism is intensified by partisanship and patterns of prosumption, operating against the backdrop of the identity threat associated with menopause and its intersection with gendered ageism. Collectively, these socioeconomic concepts have profound implications for evidence-informed behaviours and inclusive menopausal practices. The article proposes enhanced education and awareness of contemporary socioeconomic dynamics and outlines possible interventions to overcome the ways they may impede effective menopause support.
Standardized residency training in emergency medicine is not well-defined in China. Establishing a core competency indicator system for emergency residency training is essential for standardizing and enhancing the quality of training and emergency care. This study aimed to develop a core competency indicator system for emergency residency training in China, based on the competency iceberg model. This study employed a multi-step approach, including a literature review, semi-structured expert interviews, to formulate a list of preliminary indicators. A two-round modified Delphi method was utilized to achieve consensus, with 19 emergency experts recruited from 11 tertiary hospitals across four regions of China. Consensus was determined using a 5-point Likert scale, with a mean importance score > 3.5, a coefficient of variation < 0.20, and a full-score proportion > 20% serving as criteria for indicator retention. All 19 invited experts completed both rounds of consultation, achieving a 100% valid response rate. The authority coefficient of the consultation was 0.895, and Kendall's W values ranged from 0.566 to 0.634 (p < 0.005), indicating strong expert consensus. The final indicator system includes 34 third-level indicators within 14 s-level indicators, categorized into 5 first-level dimensions: motivation, trait, self-concept, knowledge, and skill. This study represents an initial step towards establishing a contemporary competency-based training program for emergency residency training in China, providing a scientific reference for the standardization of emergency residency training.