Search Results (459)

(1) Background: Hyperuricemia (HUA) has become an increasingly important global public health concern. Workers in non-ferrous metal mining and smelting industries may be exposed to multiple metals through occupational activities. This study aimed to investigate the associations between mixed-metal exposure and the risk of incident HUA in an occupational population using a nested case–control design within the Jinchang Cohort. (2) Methods: Plasma samples from 2000 participants from Jinchuan Group Co., Ltd. were available for metal measurement, of whom 1722 participants with complete data were included in the final analysis, comprising 333 incident HUA cases and 1389 controls. Eleven metals, including lithium (Li), aluminum (Al), titanium (Ti), strontium (Sr), vanadium (V), chromium (Cr), copper (Cu), zinc (Zn), iron (Fe), arsenic (As), and selenium (Se), were analyzed. Multivariable logistic regression and restricted cubic spline (RCS) models were applied to evaluate the associations between individual metal exposures and HUA risk. Bayesian kernel machine regression (BKMR) and weighted quantile sum (WQS) regression models were used to assess the associations between metal mixtures and HUA risk. (3) Results: Individual metal analyses indicated that Al, Sr, and V exposures were positively associated with HUA risk, whereas Se exposure was inversely associated with HUA risk. Both BKMR and WQS models demonstrated a positive association between mixed-metal exposure and HUA risk, with V and Sr identified as major contributors to the mixture effect. (4) Conclusions: Occupational exposure to metal mixtures was associated with an increased risk of HUA, with V and Sr playing important roles in the overall mixture effect. These findings highlight the potential health impacts of mixed-metal exposure on HUA development among occupational populations. Full article

Autonomous trading research pipelines can generate large volumes of machine-produced strategy proposals; however, testing them without leakage, overfitting, or silent selection bias remains an open problem. Existing autonomous trading systems primarily optimize strategy generation, whereas considerably less attention has been devoted to the governance of autonomous research pipelines that continuously generate, validate, and discard machine-generated hypotheses under strict temporal-integrity constraints. We introduce a season-structured validation and governance architecture that addresses this gap by combining leakage-free feature constraints, layered temporal validation, Bayesian risk-parameter optimization, and season-structured human oversight into a single reproducible workflow: within each season a local large language model (LLM) generates candidate strategies, and across season boundaries a human reviewer updates objectives and gates, forming a mixed-initiative governance layer. Across twelve adaptive seasons, 34,333 proposals produced 329 accepted strategies. Two case studies—a BNB/USDT candidate and a BTC/USDT replication—passed the predefined validation gates and showed positive selection-holdout performance under conservative calendar-time risk metrics. A feature-catalog sensitivity analysis on the BTC/USDT case study found that 3 of 50 random feature subsets of the same catalog reached the selected candidate’s recorded holdout Sharpe, with holdout Sharpe increasing with subset size and the comparison not budget-matched. A post hoc code audit conducted during the revision of this paper, disclosed here as an author-identified correction, found that the feature list included in every LLM proposal was never applied to training in any season: every reported model was trained on the full leakage-free catalog. As a result, this analysis compares the full catalog against random subsets, not LLM feature selection against chance. Therefore, the contribution is the validation and governance architecture itself, which turns a large stream of machine-generated proposals into a small set of credibly testable candidates, rather than any superior LLM feature intuition. Because the reviewer’s between-season adjustments consume out-of-sample information, the season sequence is a prolonged model-selection procedure. The case studies are therefore descriptive evidence of candidate rejection, not prospective proof of profitability. Full article

In the context of artificial vegetation restoration in arid and semi-arid regions, competition for shallow soil water between shrubs and herbaceous species often hampers restoration efficacy. In this study, we investigated the water acquisition strategies and interspecific interactions of a typical shrub–herb community comprising Caragana korshinskii and Stipa capillata in a loess hilly region under relatively dry soil conditions, using stable hydrogen and oxygen isotopes combined with Bayesian mixing models. The results revealed that the 20–70 cm soil layer, characterized by a high sand content and sustained root water extraction, exhibited low soil moisture, which drove Stipa capillata into a passive defensive state during water shortage. In contrast, Caragana korshinskii exhibited considerable resource-acquisition plasticity, with deep soil water (>70 cm) accounting for 40.6–50.3% of its total water uptake. This deep-water exploitation sustained its survival and reduced direct hydrological competition with shallow-rooted species in the surface layers. Collectively, our findings elucidate the water-use dynamics of shrubs and herbaceous species, providing scientific support for vegetation restoration and water resource management under specific geological contexts. Full article

Nested sampling is widely used in astrophysics and cosmology to reliably infer model parameters and compare models within a Bayesian framework. To address models with many parameters, Markov Chain Monte Carlo (MCMC) random walks are used within nested sampling to advance a live point population. Diagnostics for nested sampling inference are essential to verify that trustworthy astrophysical conclusions can be drawn. We develop a diagnostic to identify poorly mixed random walks that do not meet the requirements of nested sampling. The distance from the start to the end of the random walk, the jump distance, is compared to the typical neighbour distance between live points, to obtain a relative jump distance (RJD). A robust estimate of a typical neighbour Mahalanobis distance is obtained with the MLFriends algorithm. In mock and real-world inference applications such as inferring the distance to GW170817, we observe the relative jump distance. We propose the geometric mean <RJD> and fraction of RJD > 1 as new summary diagnostics. To evaluate the sensitivity of the new diagnostics, problematic runs are identified by comparison with reruns using much longer MCMC chains. Trustworthy inference runs are consistently associated with <RJD> above 1.1, while problematic runs can fall below these thresholds. The RJD is more sensitive than previous tests based on the live point insertion order. The RJD diagnostic is proposed as a widely applicable diagnostic to verify inference with nested sampling. It is implemented in the UltraNest package in version 4.1. Full article

Hemoglobin A1c (HbA1c), an important diagnostic biomarker for type 2 diabetes (T2D), is also associated with aging, cognitive performance, and mortality. To identify epistatic interactions, we assessed 133 known gene variants associated with HbA1c among 3778 non-diabetic subjects of European ancestry in the Long Life Family Study (LLFS). We applied Bayesian imputation based association mapping (BIMBAM) to identify significant pairwise epistatic interactions among 133 genetic variants. To take into account confounding effects, we adjusted age, sex, field centers, body mass index (BMI), and genetic principal components (PCs). This analysis yielded seven pairs with log10(BF) > 10; of those, six pairs were confirmed using a full-term mixed regression model. Specifically, these included significant interactions of HK1-rs17476364 with variants in GCK (rs2971670, rs4607517) or G6PC2 (rs560887), as well as between HK1-rs16926246 and the same variants (p values for each term ≤ 7.14 × 10⁻³). All epistatic interactions between HK1 and GCK, and between HK1 and G6PC2 were replicated at p < 0.05 in at least one of two large independent studies (namely, Framingham Offspring Study; Health and Retirement Study). The present study revealed that HK1 and GCK interact to contribute to regulating levels of HbA1c and are likely to be involved in molecular mechanisms underlying healthy aging processes. Full article

In the Yellow River Delta, changes in the Yellow River’s main stream flow, groundwater exchange, and seasonal precipitation jointly regulate soil water and influence local hydrological processes. Understanding the sources of soil water is crucial for water resource management in this region. A total of 210 soil water samples were collected between August and October 2022 and analyzed for δ²H and δ¹⁸O. Subsequently, the contribution of each water source in the 10–100 cm soil profile was quantified using a five-source Bayesian mixture model. The results showed that the isotopic characteristics of the six types of water bodies were different, with surface water and tidal water having a higher degree of evaporation enrichment than precipitation and Yellow River water. The isotopic changes in soil water during the flood season and the post-flood season are mainly concentrated in the shallow and middle layers of the soil profile, while the changes in the deeper layers are relatively small. The Bayesian mixture model analysis results show that the Yellow River water is the main water source of the entire soil profile, with a contribution rate of 39.7–57.1%, followed by precipitation (23.5–37.2%), while the contribution rates of groundwater, local surface water and tides are relatively low. The probability of the Yellow River as the main water source ranges from 55.9% to 80.8% at different depths. These findings indicate that the Yellow River water source has a continuous impact on soil water at different depths in the Yellow River Delta, deepening understanding of soil water recharge processes and water source connections in the coastal delta. Full article

The Ankang catchment, the primary water-producing area in the upper reaches of the Danjiangkou Reservoir catchment, experiences immense pressure on its water environmental carrying capacity. However, existing studies remain unable to differentiate between anthropogenic and natural inputs of riverine phosphorus (P), constraining catchment nutrient management. To address the issue of riverine P sources in the Ankang catchment, this study, based on data from three hydrological phases, applied a Bayesian phosphate oxygen isotope mixing model for the first time to quantify the natural and anthropogenic sources of riverine P. Subsequently, statistical methods were employed to reveal the mechanisms controlling riverine P export, thereby providing a scientific basis for riverine P management. Quantitative apportionment results showed that riverine P was primarily derived from natural sources (48.6%) in the normal phase. During the wet phase, the proportion of natural sources increased in some tributaries due to enhanced precipitation. The contribution of agricultural sources increased to 46.5% during the dry phase, driven by a new agricultural production cycle. Correlation analysis indicates that conserving water and increasing water use efficiency lower wastewater loading, and that decreasing fertilizer application or improving fertilizer use efficiency contributes to reducing agricultural inputs. Full article

To investigate the effects of hazard visibility on drivers’ avoidance behaviors and safety risks, this study developed driving scenarios involving overt and covert hazards using a high-fidelity driving simulator and conducted a series of driving experiments. A linear mixed model (LMM) was employed to systematically examine the variations in vehicle kinematic characteristics, operational response behaviors, and collision risks under different hazard visibility conditions. Furthermore, a driver avoidance capability classification model was developed by integrating factor analysis with Bayesian linear discriminant analysis. The results showed that hazard visibility was a critical factor influencing drivers’ responses. Compared with overt hazards, covert hazards significantly reduced time to collision (TTC) and increased the maximum braking pedal force of the driver. The driver avoidance capability classification model achieved an accuracy of 85.8%, enabling effective identification of drivers with low, medium, and high avoidance capabilities. These findings provide insights into driver avoidance behavior and may support driver safety assessment and traffic safety education, with potential implications for reducing crash-related injuries and economic losses, as well as promoting the sustainable development of transportation. Full article

Accelerated global urbanization places multiple pressures on the cement concrete industry, including resource consumption, carbon emissions, and solid waste disposal. Converting solid waste into construction materials offers a viable path towards its large-scale, high-value utilization, driving the green, low-carbon transition of the construction sector. However, solid waste concrete exhibits high-dimensional, multivariable, and strongly nonlinear characteristics owing to coupled factors such as waste type, source, substitution rate, mix proportions, and curing conditions. Consequently, traditional empirical methods fail to meet engineering needs, particularly in efficient mix design, performance prediction, and quality control. In conventional concrete research, predictive modeling has evolved from classical algorithms to ensemble and deep learning frameworks, often optimized via Bayesian optimization or genetic algorithms. Post-hoc interpretability methods such as SHAP and PDP are widely applied to accurately predict compressive strength, identify critical factors, reveal micro-mechanisms, and reconstruct empirical equations. In contrast, research on solid waste concrete has focused heavily on data challenges stemming from varied waste sources, including data heterogeneity, small sample sizes, and noise. In this field, swarm intelligence algorithms (such as particle swarm, grey wolf, and firefly algorithms) are commonly combined with ensemble models to improve prediction accuracy and engineering usability. However, limitations remain in data availability, feature consistency, cross-material transferability, and practical engineering application. Future research should leverage transfer learning, physics-informed neural networks (PINNs), and in-situ monitoring to advance models from standalone prediction tools into transferable, interpretable decision-support systems. Full article

In the hyper-arid Taklamakan Desert, where water availability is extremely limited, key sand-fixing species such as Tamarix ramosissima, Calligonum arborescens, and Haloxylon ammodendron are largely maintained by saline groundwater irrigation. Understanding their water-use strategies is therefore essential for optimizing irrigation practices and species selection in shelterbelt systems. However, the water sources, long-term water-use efficiency (WUE), and interspecific water competition of these species across different establishment years remain insufficiently understood. The primary objective of this study is therefore to elucidate the water-use strategies of these three sand-fixing species and to assess how they vary with stand age and mediate interspecific coexistence. To achieve this objective, stable hydrogen, oxygen, and carbon isotope analysis (δ²H, δ¹⁸O, and, δ¹³C) was integrated with MixSIAR modeling and water niche assessment for stands established in 1999, 2004, and 2009. The results showed that: (1) water sources differed significantly among species and changed dynamically with stand age, with C. arborescens exhibiting the most pronounced variation in water-uptake depth and the greatest water-use plasticity; (2) long-term WUE varied significantly across species, and its relationship with water-niche breadth showed a negative trend, suggesting that broader water-niche breadth may be associated with lower WUE; and (3) interspecific water competition diminished as increasing plantation age increased. These findings demonstrate that in the extreme arid environment of the Taklamakan Desert, sand-fixing species achieve coexistence through divergent water-use strategies, thereby mitigating competition under severe water scarcity. This study provides a theoretical foundation for the sustainable management of shelterbelt ecosystems in arid regions. Full article

China’s carbon mitigation challenge is intricately linked to energy consumption, income growth, innovation performance, and sectoral structure. Using annual aggregate data for 1995–2021, this study investigates these dynamics using Fourier ARDL and Bayesian Fourier ARDL specifications, while evaluating sectoral shares through predictive-content tests. Crucially, the empirical evidence demonstrates that emissions are not cointegrated with the explanatory series. Unresolved integration-order diagnostics formally preclude bounds-test interpretation, establishing that the model yields no verified equilibrium relationship. In first differences, energy consumption displays a strong contemporaneous short-run association with emissions. While the implied lagged-level coefficient ratio for energy is near unity, it captures dynamic parameter scaling within the short-sample ARDL specification. The implied ratios for GDP and aggregate innovation are small, statistically insignificant, and encompass zero. Evidence for an Environmental Kuznets Curve (EKC) is highly specification-dependent, and Granger-causality tests reveal frequency-sensitive short-run predictive content in differenced data, without evidence of permanent structural links. In the absence of automatic error correction, mitigation policy cannot rely on assumed steady-state adjustments, underscoring the necessity of direct, structural decarbonization of the energy mix. The findings demonstrate that long-term environmental sustainability cannot be achieved through passive economic maturation, requiring deliberate structural reforms to align industrial growth with carbon neutrality commitments. Full article

Honey bee colony health is increasingly threatened by multiple stressors, resulting in unsustainable annual mortality rates. Proper nutrition supports colony resilience, but most supplementation strategies are hard to apply at commercial scale. This study is the first commercial-scale evaluation of Reishi (Ganoderma lucidum) extract supplementation, following 326 managed honey bee colonies (108 control, 218 supplemented) across 44 apiary sites over the period 2022–2026. Within each site, colonies were randomly assigned to receive Reishi extract by a dripping method or a syrup-only control. Colony and queen longevity, population dynamics, and Varroa destructor management was modelled using Kaplan–Meier survival curves, Bayesian Cox proportional hazards models, and a discrete Bayesian network. Supplemented colonies survived a mean of 367 days, compared with 247 days for controls. The 95% credible intervals for the Bayesian CoxPH did not overlap between groups for either colony or queen survival. Querying the trained network showed an average mortality rate of 8% in supplemented colonies versus 14% in controls. That gap widened under Varroa pressure: a 13-percentage-point reduction (7% vs. 20%) when mite loads were rising, and a 23-point reduction (2% vs. 25%) in colonies with stable population size and increasing Varroa drop rates. Linear mixed-effects models on three-month colony-attribute percent changes matched these patterns, with supplementation linked to significant increases in frames of bees (p = 0.004) and frames of brood (p < 0.001). Mutual-information analysis of the network showed colony survival and frames of brood as the attributes most strongly tied to supplementation. These findings suggest that Reishi supplementation is a scalable approach to supporting colony robustness in commercial apiaries. Full article

Background/Objectives: Optimizing nitrogen fertilizer efficiency is crucial for sustainable agriculture, but the genetic basis of low-nitrogen (LN) tolerance in oat (Avena sativa L.), an important forage and grain crop, remains poorly characterized. Methods: This study investigated eleven LN tolerance phenotypes across 103 oat accessions using high-density resequencing markers, mixed-model association mapping (MLM), a FarmCPU sensitivity analysis, and Bayesian genetic architecture inference (BSLMM). Results: No locus achieved genome-wide Bonferroni significance, and Bayesian estimates indicated a low-to-moderate model-based PVE (0.10–0.23) with no clearly identified major-effect locus, suggesting that genomic selection represents a plausible future direction for LN tolerance in oat. At a suggestive threshold (p < 1 × 10⁻⁵), a cross-model intersection nominated a hypothesis-generating locus at 1D:349,799,042 associated with root dry weight. This locus showed haplotype-level phenotypic segregation (Wilcoxon p = 1.36 × 10⁻⁵) on the discovery panel and concordant signals between the two models (r = 0.92). The nearest nitrogen-metabolism-related gene, P5CDH (22.7 kb away), encodes Δ¹-pyrroline-5-carboxylate dehydrogenase, which catalyses the terminal step of proline catabolism, and its wheat ortholog was significantly down-regulated under nitrogen deficiency. Conclusions: These findings provide a testable hypothesis linking proline catabolism to oat root plasticity under low nitrogen and a framework for validation in larger panels. Full article

Agricultural nitrogen loss accounts for more than half of China’s water pollutant discharge, yet flood-season nitrate sources and transformations remain poorly quantified in cold-region basins under combined agricultural and industrial pollution. We investigated the Woken River, an agricultural–industrial tributary of the Songhua River, during the 2023 flood season (June–August), collecting water, soil, and precipitation samples at 31 sites in three monthly campaigns. Hydrochemical parameters and nitrate dual isotopes (δ¹⁵N-${\mathrm{NO}}_3^{-}$, δ¹⁸O-${\mathrm{NO}}_3^{-}$; denitrifier method) were analyzed, and the contributions of five potential sources were quantified using the MixSIAR model constrained by correlation and principal component analyses. δ¹⁵N-${\mathrm{NO}}_3^{-}$ ranged from +6.033‰ to +13.313‰ and δ¹⁸O-${\mathrm{NO}}_3^{-}$ from −2.195‰ to +6.060‰. The NO₃⁻ dual-isotope study indicates that the primary sources of NO₃⁻ in the Woken River basin are soil nitrogen (S), paper mill wastewater (PMW), and domestic sewage and livestock manure (DS&LM). Model estimation results indicate that PMW and S were the dominant sources, each contributing approximately 43% of riverine nitrate. These results support the targeted control of industrial discharge and agricultural soil nitrogen loss in cold-region basins. Full article

Background/Objectives: Voriconazole dosing in children is complicated by developmental and nonlinear pharmacokinetics, cytochrome P450 2C19 (CYP2C19) variation, variable oral bioavailability, inflammation, organ dysfunction, and drug interactions. We evaluated the characteristics, reporting completeness, and clinical applicability of pediatric and pediatric-inclusive voriconazole population pharmacokinetic (PopPK) models. Methods: PubMed, EMBASE, the Cochrane Library, Web of Science, and supplementary sources were searched from inception to 12 March 2026. Eligible studies developed a voriconazole PopPK model in patients aged 0–18 years or in a broader population with extractable or interpretable pediatric information. Reporting completeness was assessed using the 32-item Clinical Pharmacokinetic (ClinPK) checklist. Results: Eighteen studies published from 2004 to 2025 were included. Model-development cohorts contained 21–173 participants, and 15 studies reported 108–3336 concentrations. The median ClinPK reporting-completeness score was 82.6% (range, 74.2–90.0%). The included studies comprised one infant-specific, nine pediatric-only, six pediatric–young adult, and two mixed pediatric–adult models. Fifteen studies used parametric nonlinear mixed-effects methods and three used nonparametric methods; 15 selected two-compartment models and three selected one-compartment models. Body size was incorporated in 17 models and CYP2C19 phenotype was the most frequent retained covariate (10 models). Elimination structures, exposure targets, and dosing proposals were heterogeneous, and only one model underwent independent external validation. Conclusions: A population-matched model may support initial dosing, but early therapeutic drug monitoring and iterative Bayesian updating remain necessary. Current evidence does not support a universal pediatric model. Independent external validation and prospective implementation studies are required before routine clinical use. Full article