共找到 20 条结果
Lambda-cyhalothrin is a synthetic pyrethroid insecticide widely used in agriculture, but its potential to contaminate aquatic ecosystems and exert high toxicity on fish at low concentrations raises ecotoxicological concerns. While standard histopathological and biochemical assessments are common, the novel integration of pigment-based biomarkers (hemosiderin and lipofuscin) offers unexplored mechanistic insights into tissue resilience and residual oxidative stress during postexposure recovery. This study evaluated liver changes (degree of tissue changes-DTC), pigment deposition, and biochemical profiles in Oreochromis niloticus exposed to sublethal lambda-cyhalothrin (1.24 μg L-1) for 3, 7, and 14 days, followed by a 15-day recovery phase. Exposure induced a progressive, time-dependent increase in DTC, transitioning from early reversible inflammatory/vascular lesions to irreversible parenchymal and hepatopancreatic necrosis by Day 14. This structural decline correlated with significant elevations in serum ALT and AST activities. Critically, exposed fish exhibited a sharp increase in hepatic hemosiderin and lipofuscin. Following the 15-day recovery period, lipofuscin levels and GGT activity normalized, reflecting immediate metabolic detoxification. However, elevated DTC, serum transaminases, and prominent hemosiderin deposits persisted. By tracking distinct temporal trajectories, our unique multibiomarker approach demonstrates that while immediate oxidative metabolic stress subsides after exposure ceases, iron dysregulation and structural liver damage remain unresolved. These findings emphasize that standard recovery windows in clean water are insufficient to fully restore tissues, highlighting hidden ecological risks for fish farming and wild populations exposed to pulses of agricultural runoff.
In 2024, an outbreak of piglet diarrhea occurred in Guangxi Zhuang Autonomous Region, China. RT-PCR assay of clinical samples confirmed that porcine deltacoronavirus (PDCoV) was the primary pathogen. In this study, the PDCoV GX/2024 strain was successfully isolated in PK-15 cells with optimized trypsin treatment conditions. The cytopathic effect (CPE) of this strain gradually intensified during serial passages. After plaque purification, the strain achieved stable proliferation with high titer, reaching a peak titer of 107·2 TCID50/mL, which indicated its potential as an inactivated vaccine candidate. Phylogenetic analysis demonstrated that the strain belonged to the Chinese lineage, with nucleotide similarity ranging from 97.89% to 99.76% compared with other domestic and international strains. Analysis of the Spike protein (S protein) revealed that the strain harbored critical amino acid substitutions in the S1 domain and the S1/S2 cleavage site, which led to conformational changes in the protein surface and might be associated with viral infection efficiency. Animal experiments verified that PDCoV GX/2024 exhibited strong pathogenicity in newborn piglets. Diarrhea, vomiting and other symptoms appeared within 24 hours post-infection. The virus mainly colonized the epithelial cells at the tip of intestinal villi and caused severe pathological damage. This study provides important theoretical basis for the genetic evolution, pathogenic mechanism and transmission potential of PDCoV, and offers reference value for disease prevention and control as well as vaccine development.
Sickle cell disease (SCD) is caused by pathogenic variants in the β-globin gene (HBB), most commonly the variant responsible for hemoglobin S, and affects an estimated 515,000 newborns each year, with the highest burden occurring in sub-Saharan Africa. Gene editing and hematopoietic stem cell transplantation have changed the therapeutic landscape, but their cost, technical complexity, and procedure-related risks still limit their wider use. For this reason, pharmacological induction of fetal hemoglobin (HbF) remains an important therapeutic strategy. HbF reduces HbS polymerization and is associated with lower disease severity, morbidity, and mortality. Among the mechanisms involved in γ-globin silencing, epigenetic regulation offers several targets that can be explored using small molecules. This review discusses medicinal chemistry approaches primarily targeting HDAC1/2, LSD1, and DNMT1, with emphasis on inhibitor classes, binding mechanisms, structural features, preclinical evidence, and translational limitations. The available data show that each target presents a distinct set of challenges. HDAC-directed strategies require improved isoform and cellular selectivity; LSD1 inhibitors must reconcile strong HbF induction with the risks associated with prolonged target engagement; and DNMT1 modulation is moving from DNA-incorporating nucleoside analogs toward reversible non-nucleoside inhibitors. We also discuss emerging approaches, including multi-target epigenetic modulation and targeted protein degradation. Together, these strategies show how a better understanding of γ-globin repression may guide the development of safer and more accessible HbF-inducing agents.
Vigabatrin (VGB) is an antiseizure medication used as an add-on therapy for refractory focal seizures. Despite its clinical efficacy, VGB use is limited by retinal toxicity that may lead to permanent visual loss. This narrative literature review aims to elucidate the emerging molecular and metabolic mechanisms underlying VGB-induced retinal toxicity, with particular emphasis on the interplay among gamma-aminobutyrate (GABA) accumulation, taurine depletion, ornithine metabolism dysregulation, and vitamin B6-dependent pathways to identify potential therapeutic targets for preventing visual impairment. A comprehensive literature search was conducted using PubMed, ScienceDirect, and Google Scholar. Current evidence indicates that VGB-induced retinal toxicity arises from a complex interplay among GABA accumulation, taurine depletion, taurine transporter modulation, dysregulated ornithine metabolism, and vitamin B6-pyridoxal 5'-phosphate (PLP)-dependent metabolic alterations. Understanding this complex interplay between GABA-vitamin B6 axis and metabolic crosstalk offers opportunities for safer therapeutic approaches and potential protective strategies for patients undergoing long-term VGB therapy. Future research should focus on mechanistic studies, pharmacokinetic profiling, genetic studies, and longitudinal clinical investigations to maximize both seizure control and retinal safety.
Enzymatic biodiesel production using lipases offers a sustainable alternative to chemical catalysis; however, challenges such as thermal instability and methanol-induced deactivation can limit industrial application. This study investigates the catalytic efficiency and structural robustness of a reconstructed ancestral LUCA lipase (last universal common ancestor) derived from family 1.3 lipases and immobilized on Seplite LX120 for efficient biodiesel production from waste cooking oil (WCO). Experimental validation demonstrated that immobilization significantly enhanced thermal durability, achieving a maximum half-life of 61.45 h at 70 °C-approximately 4-fold higher than the free LUCA lipase (15.04 h). The immobilized LUCA lipase maintained high residual activity (58.23%-77.57%) after 180 min exposure to 25% (v/v) methanol across 60 °C-80 °C, whereas the free LUCA exhibited greater solvent sensitivity. Notably, MD simulations (50 °C-90 °C) demonstrate the exceptional rigidity of the LUCA catalytic triad framework in both water and methanol, characterized by a unique, methanol-specific 'thermostable plateau' at 70 °C. This structural resilience enabled rapid and efficient biodiesel production, with maximum biodiesel yield achieved within 2 h by the immobilized LUCA lipase and 3 h by the free LUCA lipase at 70 °C. The immobilized biocatalyst maintained 100% biodiesel yield for nine consecutive cycles, with only a slight decrease to 99.13% at the tenth cycle. Furthermore, the process was successfully scaled up 100-fold in a 1 kg stirred tank reactor, demonstrating the industrial applicability of the immobilized ancestral LUCA lipase. These findings provide a robust technological framework for utilizing ancestral enzymes in sustainable industrial-scale bioenergy production.
Although immunotherapy has revolutionized cancer treatment, hepatocellular carcinoma (HCC) continues to demonstrate limited clinical responses, highlighting the urgent need for novel immunomodulatory strategies. Trained immunity, an emerging paradigm wherein innate immune cells develop a memory-like phenotype through epigenetic and metabolic reprogramming, offers a promising avenue to remodel the immunosuppressive tumor microenvironment. This study investigated whether β-glucan-induced trained immunity could potentiate antitumor immunity against HCC. We established orthotopic HCC mouse models to investigate the role of trained immunity induced by whole β-glucan particle (WGP) in the HCC microenvironment, particularly in modulating hepatic apolipoprotein E (APOE)-positive monocytes/macrophages. Transcriptional changes in trained monocytes/macrophages were identified by analyzing single-cell RNA sequencing and bulk RNA-sequencing data from the livers of WGP-treated and control mice. Mechanistic studies were performed using Apoe -/- mice and in situ monocyte/macrophage engineering. Flow cytometry was performed to assess immune cell phenotypes and phagocytosis, while luminescence-based assays were used to evaluate cytotoxic activity. The translational potential was assessed using human monocyte training assays. This study demonstrated that preconditioning with WGP, a trained immunity inducer, increased the accumulation of trained monocytes/macrophages in the liver and suppressed tumor progression in HCC mouse models. Mechanistically, WGP-trained APOE+ monocytes/macrophages exhibited a decrease in lipid accumulation and endoplasmic reticulum stress, thereby enhancing their antitumor function. Genetic deletion of Apoe in monocytes/macrophages abrogated the antitumor effects of WGP, demonstrating that APOE+ monocytes/macrophages are essential mediators of WGP-induced trained immunity. Adoptive transfer of WGP-trained bone marrow-derived macrophages suppressed the growth of HCC in recipient mice. Furthermore, WGP induced trained immunity in human monocytes, leading to enhanced killing of HCC cells. Notably, combination therapy with WGP and anti-programmed death-ligand 1 antibody achieved superior tumor control compared with either monotherapy. These findings identify a critical role for trained APOE+ monocytes/macrophages in WGP-mediated antitumor immunity in the liver. Harnessing WGP-induced peripheral trained immunity represents a novel therapeutic strategy for HCC.
To develop and validate UNetrDose, a Transformer-based deep learning model designed for fast and accurate photon beamlet dose prediction. The study aims to achieve Monte Carlo (MC)-level dosimetric accuracy using only beamlet-specific CT images and beamlet coordinates as input, enabling the efficient reconstruction of complete 3D dose distributions for intensity-modulated radiation therapy (IMRT) plans. 
Approach. For each beamlet, a fixed-size 3D CT patch was extracted along its propagation path, centered on the beamlet trajectory. The geometric information, defined as the beamlet's relative position within the beam field, was used alongside the CT patch as model input. The ground-truth dose distributions were generated using MC simulations. The proposed UNetrDose model combined convolutional layers for local feature extraction with Transformer modules to capture long-range dependencies. A total of 90 fixed-beam IMRT plans (51 esophageal and 39 rectal cases) were used for model training and validation. Model performance was comprehensively assessed, evaluating spatial accuracy with 3D gamma pass rates and clinical acceptability through Dose-Volume Histogram (DVH) comparisons and other dosimetric parameters. 
Main results. UNetrDose demonstrated high fundamental accuracy at the individual beamlet level, where pass rates for the stringent γ(1 mm, 1%) criterion exceeded 96% for both esophageal and rectal cases. This translated to excellent clinical performance on full IMRT plans, where the model achieved mean γ(2 mm, 2%) pass rates ranging from 97.06 ± 2.05% for esophageal cases to 98.75 ± 0.78% for rectal cases. The model was also highly efficient, with an average inference time of approximately 28 ms per beamlet. 
Significance. UNetrDose offers a promising alternative to traditional dose calculation engines by providing a balance between high dosimetric accuracy and fast computation. Its ability to predict dose distributions using only CT images and beamlet positions simplifies the workflow, making it highly applicable for time-sensitive clinical scenarios. 
.
Controlling pollution from waste landfills remains a critical environmental challenge, particularly in developing regions and fragile ecosystems. An effective tool to support this process is the hydrological assessment of disposal sites through simulation models. This study provides a comprehensive systematic review of mathematical models used to simulate landfill water balance over the past 15 years. A total of 32 relevant studies were identified from the Scopus, Web of Science, and SciELO databases, classifying them by model type, application scope (real or experimental), and climatic region. Five main models were recognized: HELP, UNSAT-H, VADOSE/W, MODUELO, and EMWMF. Their structural features, simulation capabilities, and applicability are examined. Processes such as evapotranspiration, evaporation, infiltration, runoff, and leachate generation are represented using distinct approaches, ranging from empirical to physically based methods. The findings show that HELP is the most widely applied model, whereas MODUELO and VADOSE/W enable more detailed simulations and/or account for spatial variability. A knowledge gap persists in tropical and low-income regions, where landfill hydrology remains insufficiently investigated. The study emphasizes the need for models adapted to local conditions and data availability to enhance landfill management and reduce risks that may be intensified by climate change. This review offers a foundation to support the selection and application of models in underexplored or resource-limited contexts.
Quantitative analysis of volatile organic compounds (VOCs) in biological matrices remains challenging due to matrix-dependent headspace partitioning, the presence of endogenous analytes, and the intrinsically non-linear response of certain detectors. In this context, static headspace gas chromatography-ion mobility spectrometry (HS-GC-IMS) offers high sensitivity for VOC profiling, but these combined effects complicate calibration and quantitative interpretation. In this work, a practical framework for quantitative VOC analysis in urine using HS-GC-IMS is presented, in which calibration is performed in a synthetic urine surrogate matrix and subsequently adapted to real urine samples. The approach preserves the inherent non-linear detector response and accounts for matrix-dependent partitioning effects through an affine adjustment of the concentration axis. The resulting matrix-adapted calibration model can be applied to individual urine samples without requiring patient-specific recalibration. The methodology was evaluated using three colorectal cancer-related VOCs (anisole, 2-heptanone, and 2-pentanone) over a 0-30 ppb concentration range. Matrix adaptation substantially improved quantitative accuracy within the transferable dynamic range, particularly for compounds strongly affected by matrix-dependent partitioning, while highlighting fundamental limitations when endogenous concentrations place the instrument response near saturation. By explicitly addressing the interplay between headspace partitioning and non-linear detector behavior, the proposed strategy allows reliable surrogate-matrix calibration in complex biological samples and is applicable to other headspace-based analytical platforms affected by matrix effects.
Dental hygiene practice is rapidly evolving in response to emerging evidence on the oral-systemic health connection, advances in preventive technologies, and increased emphasis on individualized, minimally invasive care. A commitment to lifelong learning and continuing professional development is essential for maintaining clinical competence and integrating new scientific knowledge into practice. Biological dental hygiene has emerged as a complementary framework that emphasizes whole-body health, biocompatibility, reduction of toxic exposures, and prevention-centered care. This approach expands traditional dental hygiene practice by incorporating enhanced risk assessment, salivary diagnostics, targeted preventive strategies, as well as patient education focused on nutrition, inflammation reduction, and systemic health influences. This short report describes the principles of biological dental hygiene and its clinical application, including minimally invasive periodontal therapy, risk-based preventive protocols, and the use of biocompatible materials and adjunctive technologies. It also reviews professional development opportunities through the International Academy of Oral Medicine and Toxicology and the International Academy of Biological Dentistry and Medicine, which provide structured educational programs, ongoing professional development opportunities, and advanced credentials in biologically oriented oral health care. While these certifications do not alter licensure scope of practice, they support professional development and interdisciplinary collaboration. Biological dental hygiene offers an expanded framework for integrating oral and systemic health considerations into preventive care. Continued engagement in evidence-informed education and professional certification may enhance patient education, clinical decision-making, and overall care delivery within dental hygiene practice.
Wheat gluten index is an important indicator of food quality inspection. Although hyperspectral technology offers non-destructive and rapid detection potential, its application faces challenges including high-dimensional data redundancy, multicollinearity among adjacent wavelengths, and model overfitting risks in small-sample scenarios. This study proposes a rapid prediction method for wheat gluten index integrating adaptive spectral preprocessing with machine learning. Based on 89 wheat flour samples, adaptive wavelet denoising (average SNR improvement: 11.92 dB), successive projections algorithm (SPA) feature selection, and variance inflation factor (VIF) collinearity diagnosis reduced 2001 spectral features to 5 core variables (587 nm, 436nm_square, 1241 nm, 1080nm_square, 940nm_square). Ten algorithms including PLSR, SVR, XGBoost, and CatBoost were systematically compared under 70% training/30% test split with 10-fold cross-validation. CatBoost achieved optimal performance (R2 = 0.8552, RMSE = 7.8602), surpassing PLSR (R2 = 0.7090) and XGBoost (R2 = 0.7923) by 20.6% and 8.0% respectively. Notably, the well-tuned single CatBoost model outperformed stacking ensemble methods (R2 = 0.7803), providing empirical evidence for model selection in small-sample spectral analysis. SHAP interpretability analysis identified critical spectral bands corresponding to protein and starch absorption characteristics, offering guidance for portable equipment optimization and mechanistic understanding of spectral-quality relationships.
Postural and locomotor dysfunction represent axial symptoms of Parkinson's disease (PD), which remain poorly treated by medication and deep brain stimulation. Whilst non-invasive neuromodulation of the vestibular system via the vestibular nucleus complex (VNC) offers a novel therapeutic avenue, the underlying circuits are still poorly characterized. Here we show that the mouse VNC feeds extensive Vglut2-defined projections into striato-thalamo-subthalamic and caudal medulla motor hubs and receives substantial input from the sensorimotor cortex. Optogenetic activation of excitatory VNC neurons at sub-symptomatic intensities increased cFos-based activity in basal ganglia-associated and brainstem motor targets. Unbiased pose dynamics and motion analysis respectively showed enhancement of behavioural modularity and locomotion with threshold-level stimulation. In a mouse model of PD, the latter further favored naturalistic gait patterns through improved motor coordination. Our data identify excitatory VNC processes as candidates for therapeutic targeting of axial motor dysfunction in the context of PD.
Keratoconus is a common blinding corneal disease among young adults. A large number of atypical or early-stage patients are highly susceptible to missed diagnosis or misdiagnosis due to unremarkable clinical manifestations, leading to the loss of opportunities for early effective intervention and management, and ultimately resulting in irreversible blindness. In recent years, with the expansion of the population undergoing corneal refractive surgery, establishing an accurate screening and diagnostic system for keratoconus has important clinical significance for preventing postoperative complications. In view of this, the Refractive Surgery Group of the Ophthalmology Branch of the Chinese Medical Doctor Association, in collaboration with the Corneal Disease Group of the Ophthalmology Branch of the Chinese Medical Association, has developed consensus opinions on confusing concepts and terms related to early keratoconus, as well as the strategies, criteria, methods and systems for the early screening and diagnosis of keratoconus, based on relevant domestic and international consensus, evidence-based medicine evidence and expert clinical experience, using scientific and standardized methods. This consensus aims to continuously establish and improve the early screening and diagnostic pathway for keratoconus, which not only provides a scientific basis for the preoperative evaluation of corneal refractive surgery and enhances the safety of corneal refractive surgery, but also offers a reference for the early clinical diagnosis and treatment of keratoconus. 圆锥角膜为青年人常见的致盲性角膜疾病，大量非典型或早期患者因临床表现不明显，极易被漏诊或误诊，丧失早期有效干预和处理时机，导致不可逆盲。近年随着角膜屈光手术人群扩大，建立精准的圆锥角膜筛查诊断体系，对于预防术后并发症具有重要临床意义。鉴于此，中国医师协会眼科医师分会屈光手术学组联合中华医学会眼科学分会角膜病学组，基于国内外相关共识、循证医学证据及专家临床经验，采用规范的指南和共识制订方法，针对与早期圆锥角膜相关的易混淆概念和词语，以及圆锥角膜早期筛查诊断的策略、标准、方法、体系等，形成共识性意见，以期建立并不断完善圆锥角膜早期筛查诊断路径，不仅为角膜屈光手术的术前评估提供科学依据，提升角膜屈光手术的安全性，也为临床圆锥角膜的早期诊疗提供指导意见。.
This study examines the psychological mechanisms through which belief in state war propaganda relates to combat motivation among Russian soldiers during the war in Ukraine. Using a cross-sectional survey of 1,060 Russian prisoners of war, we examined the relationship between belief in official propaganda narratives and two key outcomes: voluntary surrender and self-reported intentions to re-enlist. Results revealed that stronger belief in propaganda was associated with a lower likelihood of surrendering voluntarily and higher intentions to re-enlist upon release. Mediation analyses identified perceived legitimacy of the "Special Military Operation" (SMO) as the primary psychological pathway, such that only perceived legitimacy statistically accounted for the association between propaganda beliefs and both outcomes in our models. Contrary to expectations, identity fusion with the "Russian World" ideology and dehumanization of Ukrainians did not exhibit significant relationships in the mediation models once covariates were accounted for. These findings suggest that propaganda's association with combat motivation may primarily be explained by soldiers' perceptions of the war's legitimacy rather than by identity-based or dehumanizing beliefs. The study offers empirical insights on propaganda's link to combat motivation via legitimacy, informing narrative interventions in deradicalization and conflict psychology. This study reveals how belief in state propaganda relates to the willingness of Russian soldiers to fight in the war in Ukraine, showing that soldiers who believe official narratives are less likely to surrender and more likely to intend to re-enlist. Surprisingly, the key driver is not hatred or blind loyalty, but the belief that the war is justified and legitimate. These findings underscore the power of narratives in sustaining conflict and suggest that challenging the perceived legitimacy of war could be crucial for deradicalization and peacebuilding efforts, both in military contexts and in addressing broader societal polarization.
Alzheimer's disease (AD) is a progressive, irreversible, and multifaceted neurodegenerative disorder characterized by cognitive decline, memory loss, and behavioral impairment, posing a major global health challenge. Its multifactorial pathology includes cholinergic dysfunction, amyloid-β deposition, tau hyperphosphorylation, oxidative stress, and neuroinflammation. Among these, impairment of the cholinergic system, characterized by reduced acetylcholine levels, plays a crucial role in cognitive deficits. The enzymes acetylcholinesterase (AChE) and butyrylcholinesterase (BChE), which hydrolyze acetylcholine, are closely involved in disease progression and serve as important therapeutic and diagnostic targets in AD. This book chapter provides a comprehensive overview of therapeutic and diagnostic agents targeting AChE and BChE in AD, and discusses small-molecule inhibitors, multifunctional ligands, and emerging strategies to modulate cholinesterase activity and restore cholinergic neurotransmission, alleviating disease symptoms. In addition, the chapter highlights advances in diagnostic approaches using fluorescent probes, particularly near-infrared (NIR) probes, for selective detection and imaging of AChE and BChE, including their molecular design, photophysical properties, enzyme selectivity, and mechanisms of action, all of which are critically examined. Targeting AChE and BChE offers a dual advantage in AD by enabling both symptomatic treatment and early-stage diagnosis. This chapter aims to present a clear and comprehensive overview of recent advances in therapeutic and diagnostic approaches, offering meaningful insights for researchers in developing effective strategies for the treatment and monitoring of AD.
In recent years, large language models (LLMs) have achieved remarkable advances in code generation. However, their massive parameter scales hinder deployment in resource-constrained environments. Knowledge distillation has emerged as an effective compression technique that transfers knowledge from a large teacher model to a smaller student model, thereby reducing computational cost while retaining strong generative capability. However, traditional distillation methods usually depend on forward and reverse Kullback-Leibler (KL) divergence, aligning the probability distribution over the entire vocabulary. This process makes them susceptible to long-tail noise and often leads to weaker performance than supervised fine-tuning with labeled data. To address this issue, we propose a distillation approach based on ranking supervision. At each step, the method selects the candidate tokens with the highest probabilities from the teacher's output and applies a ListNet-based loss. This loss encourages the student to learn the teacher's ranking preferences. Unlike conventional KL distillation, ranking distillation avoids exhaustive alignment of low-confidence tokens, achieving comparable training time while significantly reducing GPU memory consumption. We conduct systematic evaluations on four public benchmarks (HumanEval, MBPP, DS-1000, and MultiPL-E). Experimental results demonstrate that the proposed method consistently outperforms supervised fine-tuning as well as FKL and RKL baselines in Python code generation, multilingual generation, and data-science scenarios. Moreover, it maintains stable performance gains across different model scales, including both the Qwen2.5-Coder and DeepSeek-Coder families. Our method provides an effective solution for distilling large language models in code generation and offers guidance for future research in model compression.
Purpose Dental hygienists are uniquely positioned to improve oral and overall health outcomes with evidence-based treatment plans. One recommendation that has yet to see its full potential is the use of stannous fluoride (SnF2) toothpaste. Stannous fluoride is more than just an anticaries agent. Its formulation within a toothpaste offers superior performance in reducing plaque, gingivitis, sensitivity, halitosis, and tooth erosion, while maintaining the sulcular barrier integrity. However, most patients are unaware of these easily available benefits for oral health. This short report highlights the universal efficacy of SnF2 toothpaste, including significant reductions in gingival bleeding as compared with regular fluoride; reviews its proposed mechanisms, including antibacterial and antivirulence activity around teeth and implants as well as substructure remineralization; and explores its potential systemic health impact. The specific formulation of the toothpaste product also affects these factors, emphasizing the value of formulation-specific evidence to inform chairside recommendations. Including a well-formulated SnF2 toothpaste as part of daily oral hygiene leverages an existing behavior and reinforces proper habits for better oral health outcomes.
Per- and polyfluoroalkyl substances (PFAS) pose long-term risks to ecosystems and human health because of their persistence, bioaccumulation, long-range transport, and toxicity. Focusing on the Northeast Black Soil Region of China, a major grain production base with intensive agricultural activities and complex industrial inputs, this study developed a comprehensive analytical framework integrating multi-source data (2005-2025), including literature, environmental monitoring, soil properties, and toxicity databases. The framework combined meta-analysis, principal component analysis (PCA), machine learning, and entropy-weight-based risk assessment to systematically characterize PFAS pollution patterns, spatial heterogeneity, and priority risks. Random-effects meta-analysis revealed moderate-to-strong positive correlations among PFAS congeners (pooled effect size = 0.336) with significant heterogeneity (I² = 96.80%). PCA showed that the first two principal components explained over 48% of the total variance, whereas six components accounted for more than 85%. Among eleven machine-learning models, Logistic Regression achieved the best performance and stable generalization when the sample size exceeded 60 samples. SHapley Additive exPlanations (SHAP) identified latitude, longitude, and soil organic carbon as the dominant predictors. Risk assessment indicated that PFOS and PFOA exhibited the highest priority levels (ToxPi: 0.872 and 0.499; EHPi: 0.839 and 0.679), whereas approximately 60% of PFAS were classified as low risk and 20% as medium-to-high risk. This framework provides quantitative support for targeted PFAS management and offers a transferable strategy for regional-scale assessment of emerging contaminants.
Glioblastoma (GBM) remains one of the most lethal human malignancies, yet extraneural metastases are exceptionally rare and poorly understood. Standard treatment-surgical resection followed by chemoradiotherapy-offers only modest survival benefits, and therapeutic options for metastatic GBM are limited. Here, we describe a rare clinical case of metastatic GBM and provide comprehensive molecular characterization alongside a potential targeted treatment strategy. The tumor exhibited several molecular features associated with aggressive behavior, including alterations in the tumor suppressor genes NF1, TP53, PTEN, and RB1, a mesenchymal DNA‑methylation subclass, and activation of the MAPK signaling pathway. To functionally evaluate therapeutic vulnerabilities, we developed tumor models using patient‑derived GBM cells and their assembloids with cerebral organoids that recapitulate patient‑specific tumor biology and the human brain microenvironment. We identified the MEK inhibitor trametinib as a promising candidate capable of selectively reducing viability and invasion of highly aggressive, stem‑like GBM cells within our personalized patient-derived tumor avatars. This work highlights the proof-of-concept evidence that MEK pathway inhibition may represent a potential therapeutic vulnerability to counteract rapid tumor spread and improve responsiveness to temzolomide in this individual metastatic GBM case. Study underscores the need for continued research to advance targeted, multi‑modal therapeutic approaches for GBM.
White star apple (Gambeya albida) is native to the lowland rainforests of Central, East, and West Africa. This species is highly valued for its nutritious fruits and offers medicinal, socio-cultural, and economic benefits. In West Africa, it contributes to food security for rural and urban communities alike. However, no genomic resources are available to untap the agronomic and medicinal traits of the white star apple. Here, we present its first chromosome-scale genome, generated using PacBio HiFi and Omni-C sequencing. The white star apple genome is highly homozygous, and we assembled 98.9% of the estimated haploid genome size (822 Mbp) into 13 pseudochromosomes. It has a base-level accuracy (QV) of 58.14, an N50 of 57 Mbp, and 97.5% BUSCO completeness, representing a reference-quality assembly. About 58.5% of the genome constitutes repetitive sequences, and ab initio gene prediction identified 33,602 gene models. This reference-quality genome of the white star apple will serve as a valuable resource to enhance our understanding of its nutritional and pharmacological traits and facilitate improvement research.