Efficient federated learning for pediatric pneumonia on chest X-ray classification.

According to the World Health Organization (WHO), pneumonia kills about 2 million children under the age of 5 every year. Traditional machine learning methods can be used to diagnose chest X-rays of pneumonia in children, but there is a privacy and security issue in centralizing the data for training. Federated learning prevents data privacy leakage by sharing only the model and not the data, and it has a wide range of application in the medical field. We use federated learning method for classification, which effectively protects data security. And for the data heterogeneity phenomenon existing in the actual scenario, which will seriously affect the classification effect, we propose a method based on two-end control variables. Specifically, based on the classical federated learning FedAvg algorithm, we modify the loss function on the client side by adding a regular term or a penalty term, and add momentum after the average aggregation on the server side. The federated learning approach prevents the data privacy leakage problem compared to the traditional machine learning approach. In order to solve the problem of low classification accuracy due to data heterogeneity, our proposed method based on two-end control variables achieves an average improvement of 2% and an accuracy of 98% on average, and 99% individually, compared to the previous federated learning algorithms and the latest diffusion model-based method. The classification results and methodology of this study can be utilized by clinicians worldwide to improve the overall detection of pediatric pneumonia.

An integrated strategy for in-depth profiling of N-acylethanolamines in biological samples by UHPLC-HRMS.

BACKGROUND: N-acylethanolamines (NAEs) are a class of naturally occurring bioactive lipids that play crucial roles in various physiological processes, particularly exhibiting neuroprotective and anti-inflammatory properties. However, the comprehensive profiling of endogenous NAEs in complex biological matrices is challenging due to their low abundance, structural similarity and the limited availability of commercial standards. Here, we propose an integrated strategy for comprehensive profiling of NAEs that combines chemical derivatization and a three-dimensional (3D) prediction model based on quantitative structure-retention time relationship (QSRR) using liquid chromatography coupled with high-resolution tandem mass spectrometry (LC-HRMS). RESULTS: After acetyl chloride (ACC) derivatization, the detection sensitivity of NAEs was significantly improved. We developed a QSRR prediction model to construct an in-house database for 141 NAEs, encompassing information on RT, MS1 (m/z), and MS/MS spectra. Propargylamine-labeled fatty acids were synthesized as RT calibrants across various analytical conditions to enhance the robustness of the RT prediction model. NAEs in biological samples were then in-depth profiled using parallel reaction monitoring (PRM) acquisition. This integrated strategy identified and annotated a total of 50 NAEs across serum, hippocampus and cortex tissues from a 5xFAD mouse model of Alzheimer's disease (AD). Notably, the levels of polyunsaturated NAEs, particularly NAE 20:5 and NAE 22:6, were significantly decreased in 5xFAD mice compared to WT mice, as confirmed by accurate quantitation using ACC-d0/d3 derivatization. SIGNIFICANCE: Our integrated strategy exhibits great potential for the in-depth profiling of NAEs in complex biological samples, facilitating the elucidation of NAE functions in diverse physiological and pathological processes.

Formulating Spray-Dried Albumin-Modified Lipid Nanoparticles Encapsulating Acyclovir for Enhanced Pulmonary Drug Delivery.

BACKGROUND: Viral pneumonia, a pressing global health issue, necessitates innovative therapeutic approaches. Acyclovir, a potent ring-opening antiviral agent with broad-spectrum activity, faces water solubility, oral bioavailability, and drug resistance challenges. The aim of this study was to increase the efficacy of acyclovir through respiratory delivery by encapsulating it within albumin-modified lipid nanoparticles and formulate it as a spray. METHODS: Nanoparticles was synthesized via the reverse evaporation method; its physicochemical characteristics were rigorously evaluated, including particle size, zeta potential, morphology, encapsulation efficiency, drug loading, and release profile. Furthermore, the cytotoxicity of nanoparticles and its therapeutic potential against viral pneumonia were assessed through cellular and animal model experiments. Result s: Nanoparticles exhibited a spherical morphology, with a mean particle size of 97.48 ± 5.36 nm and a zeta potential of 30.28 ± 4.72 mv; they demonstrated high encapsulation efficiency (93.26 ± 3.27%), drug loading (11.36 ± 0.48%), and a sustained release profile of up to 92% under neutral conditions. Notably, nanoparticles showed low cytotoxicity and efficient intracellular delivery of acyclovir. In vitro studies revealed that nanoparticles significantly reduced interleukin-6 levels induced by influenza virus stimulation. In vivo, nanoparticles treatment markedly decreased mortality, attenuated the inflammatory markers interleukin-6 and tumor necrosis factor-α levels, and mitigated inflammatory lung injury in mice with viral pneumonia. CONCLUSIONS: In this study, albumin was modified with polyethylene glycol (PEG) containing cationic lipid nanoparticles (LN) to prepare albumin-modified lipid nanoparticles encapsulating acyclovir (ALN-Acy), which can effectively deliver Acy into tissues and cells, prolong the survival of mice, and reduce lung injury and inflammatory factors. White albumin LN can be used as efficient drug delivery carriers, and the delivery of Acy via albumin LN is expected to be a therapeutic strategy for treating inflammatory diseases.

ROI constrained optimal online allocation in sponsored search.

Sponsored search plays a major role in the revenue contribution of e-commerce platforms. Advertising systems are designed to maximize platform revenue, but other goals also need to be considered, such as user experience, advertiser utility, and how to achieve the long-term revenue goal. A key component of a sponsored search system is online allocation, which makes real-time decisions to match users' search requests with relevant ad campaigns to maximize platform revenue within constraints such as campaign budgets. Although much progress has been made, most of the research work on allocation problem has focused on satisfying guaranteed deals for display ads, and those challenges for allocation problems in sponsored search are not properly addressed. In this paper, we develop a framework to solve the large-scale sponsored search ad allocation problem, consisting of two main parts. One is an optimization problem solved offline by a parameter-server based architecture, and the other is an online strategy to alleviate the conflict with the auction mechanism during online service. Comprehensive offline evaluation on real production data and online A/B testing on real production system have been made. The experimental results demonstrate that through better allocating user queries to appropriate ads, the proposed model can significantly increase the platform's revenue without sacrificing advertisers' ROI.

Development of a risk prediction model for the first occurrence of thrombosis in patients with OAPS.

Objectives: The aim of this study is to assess the risk factors associated with thrombotic events in obstetric antiphospholipid syndrome (OAPS) patients and to develop a predictive model specifically tailored to predict the risk of postpartum thrombosis in OAPS patients without prior thrombotic events. This research seeks to enhance clinician's awareness regarding the postpartum care and monitoring of OAPS patients. Methods: A retrospective study was conducted at the First Affiliated Hospital of the Fourth Military Medical University including 269 consecutive inpatients diagnosed with antiphospholipid syndrome (APS) from July 1, 2008 to July 31, 2022. All participants met the 2006 Sydney APS classification criteria or the "non-criteria OAPS classification". Out of 98 candidate clinical and laboratory parameters considered, 40 potential variables were selected for analysis based on expert opinion. The logistic regression mode with the Least Absolute Shrinkage and Selection Operator (LASSO) were used to identify optimal predictive characteristics. All samples were included in the model building and a nomogram was generated based on these characteristics. The differentiation, calibration, and clinical utility of the predictive model were evaluated using the area under the curve (AUC), calibration curve, and decision curve analysis. The model was also validated by a 1000 bootstrap tests. Results: 126 patients with OAPS were enrolled, and a total of 89 OAPS patients who had never experienced thrombosis were retrospectively analyzed. After 3 years follow-up, 32.58% of the patients (29/89) developed thrombosis. In order to create, LASSO logistic regression identified three optimal variables: the platelet count less than 125×109/L, more than one positive aPLs (antiphospholipid antibody), and the use of low molecular weight heparin (LMWH) or low dose aspirin (LDA) after delivery. A predictive model was conducted using these three predictive indicators for patients with OAPS who experience thrombosis for the first-time. This prediction model has good distinction, good calibration, and fair clinical practicality. Conclusion: Our model has good predictive ability in assessing the risk of thrombosis in patients with OAPS without prior thrombotic events. This model is easy to predict, has good discriminability and calibration, and can be utilized as a routine tool for thrombus screening in OAPS patients.

Role of microglia polarization induced by glucose metabolism disorder in the cognitive impairment of mice from PM2.5 exposure.

Studies have found that PM2.5 can damage the brain, accelerate cognitive impairment, and increase the risk of developing a variety of neurodegenerative diseases. However, the potential molecular mechanisms by which PM2.5 causes learning and memory problems are yet to be explored. In this study, we evaluated the neurotoxic effects in mice after 12 weeks of PM2.5 exposure, and found that this exposure resulted in learning and memory disorders, pathological brain damage, and M1 phenotype polarization on microglia, especially in the hippocampus. The severity of this damage increased with increasing PM2.5 concentration. Proteomic analysis, as well as validation results, suggested that PM2.5 exposure led to abnormal glucose metabolism in the mouse brain, which is mainly characterized by significant expression of hexokinase, phosphofructokinase, and lactate dehydrogenase. We therefore administered the glycolysis inhibitor 2-deoxy-d-glucose (2-DG) to the mice exposed to PM2.5, and showed that inhibition of glycolysis by 2-DG significantly alleviated PM2.5-induced hippocampal microglia M1 phenotype polarization, and reduced the release of inflammatory factors, improved synaptic structure and related protein expression, which alleviated the cognitive impairment induced by PM2.5 exposure. In summary, our study found that abnormal glucose metabolism-mediated inflammatory polarization of microglia played a role in learning and memory disorders in mice exposed to PM2.5. This study provides new insights into the neurotoxicity caused by PM2.5 exposure, and provides some theoretical references for the prevention and control of cognitive impairment induced by PM2.5 exposure.

scGAA: a general gated axial-attention model for accurate cell-type annotation of single-cell RNA-seq data.

Single-cell RNA sequencing (scRNA-seq) is a key technology for investigating cell development and analysing cell diversity across various diseases. However, the high dimensionality and extreme sparsity of scRNA-seq data pose great challenges for accurate cell type annotation. To address this, we developed a new cell-type annotation model called scGAA (general gated axial-attention model for accurate cell-type annotation of scRNA-seq). Based on the transformer framework, the model decomposes the traditional self-attention mechanism into horizontal and vertical attention, considerably improving computational efficiency. This axial attention mechanism can process high-dimensional data more efficiently while maintaining reasonable model complexity. Additionally, the gated unit was integrated into the model to enhance the capture of relationships between genes, which is crucial for achieving an accurate cell type annotation. The results revealed that our improved transformer model is a promising tool for practical applications. This theoretical innovation increased the model performance and provided new insights into analytical tools for scRNA-seq data.

Porphyrin-based-MOF nanocomposite hydrogels for synergistic sonodynamic and gas therapy against tumor.

The glioma is one of the most aggressive tumors in humans, which is difficult to eradicate clinically. Therefore, we devised a porphyrin-based metal-organic frameworks (MOFs) crosslinking hyaluronic acid (HA) hydrogel nanocomposite through double-network (Cu-MOF-S-S-HA-Gel, CSSH-Gel), which is tumor responsive for enhanced gas therapy and sonodynamic therapy (SDT). Firstly, the hydrogels show extraordinary injectability and biocompatibility, which enables intratumor administration to circumvent the danger associated with surgery. The Cu-MOF-Cys and HA-Cys are interconnected through ether and disulfide bonds to establish a dual-network gel structure. The overexpressed glutathione (GSH) in tumor microenvironment (TME) reacts with disulfide bonds to release of the nanosensitizer (Cu-MOF). Subsequently, Cu-MOF generates reactive oxygen species (ROS) upon ultrasound irradiation for SDT, and releases L-cysteine(L-Cys) catalyzed by 3-mercapto pyruvate sulfotransferase (3-MST) to generate H2S for gas therapy. The CSSH-Gel obtained excellent synergistic anti-tumor effects (82.34 % inhibition ratio in vivo), which holds tremendous promise for the advancement of minimally invasive glioma therapies.

Target Fishing Reveals a Novel Mechanism of N-Acylamino Saccharin Derivatives Targeting Glyceraldehyde-3-Phosphate Dehydrogenase toward Cyanobacterial Blooms Control.

Based on current challenges of poor targeting and limited choices in chemical control methods of cyanobacterial blooms (CBs), identifying new targets is an urgent and formidable task in the quest for target-based algaecides. This study discovered N-acylamino saccharin derivatives exhibiting potent algicidal activity. Thus, using N-acylamino saccharin as the probes, glyceraldehyde-3-phosphate dehydrogenase from cyanobacterial (CyGAPDH) was identified as a new target of algaecides through the activity-based protein profiling (ABPP) strategy for the first time. Building upon the structure of Probe2, a series of derivatives were designed and synthesized, with compound b6 demonstrating the most potent inhibitory activity against CyGAPDH and Synechocystis sp. PCC6803 (IC50 = 1.67 µM and EC50 = 1.15 µM). Furthermore, the potential covalent binding model of b6 to the cysteine residue C154 was explored through covalent possibility prediction, LC-MS experiments, substrate competitive inhibition experiments, and molecular docking. Especially, the results revealed C154 as a crucial covalent binding site, with residues T184 and R11 forming robust hydrophobic interactions and H181 establishing significant hydrogen-bonding interactions with b6, highlighting their potential as essential pharmacophores. In summary, this study not only identifies a novel target of algaecides for the control of CB but also lays the solid foundation for the development of targeted covalent algaecides.

Comprehensive analysis and experimental verification of the mechanism of anoikis related genes in pancreatic cancer.

Background: Pancreatic cancer (PC), characterized by its aggressive nature and low patient survival rate, remains a challenging malignancy. Anoikis, a process inhibiting the spread of metastatic cancer cells, is closely linked to cancer progression and metastasis through anoikis-related genes. Nonetheless, the precise mechanism of action of these genes in PC remains unclear. Methods: Study data were acquired from the Cancer Genome Atlas (TCGA) database, with validation data accessed at the Gene Expression Omnibus (GEO) database. Differential expression analysis and univariate Cox analysis were performed to determine prognostically relevant differentially expressed genes (DEGs) associated with anoikis. Unsupervised cluster analysis was then employed to categorize cancer samples. Subsequently, a least absolute shrinkage and selection operator (LASSO) Cox regression analysis was conducted on the identified DEGs to establish a clinical prognostic gene signature. Using risk scores derived from this signature, patients with cancer were stratified into high-risk and low-risk groups, with further assessment conducted via survival analysis, immune infiltration analysis, and mutation analysis. External validation data were employed to confirm the findings, and Western blot and immunohistochemistry were utilized to validate risk genes for the clinical prognostic gene signature. Results: A total of 20 prognostic-related DEGs associated with anoikis were obtained. The TCGA dataset revealed two distinct subgroups: cluster 1 and cluster 2. Utilizing the 20 DEGs, a clinical prognostic gene signature comprising two risk genes (CDKN3 and LAMA3) was constructed. Patients with pancreatic adenocarcinoma (PAAD) were classified into high-risk and low-risk groups per their risk scores, with the latter exhibiting a superior survival rate. Statistically significant variation was noted across immune infiltration and mutation levels between the two groups. Validation cohort results were consistent with the initial findings. Additionally, experimental verification confirmed the high expression of CDKN3 and LAMA3 in tumor samples. Conclusion: Our study addresses the gap in understanding the involvement of genes linked to anoikis in PAAD. The clinical prognostic gene signature developed herein accurately stratifies patients with PAAD, contributing to the advancement of precision medicine for these patients.

Polystyrene microplastics and di-2-ethylhexyl phthalate co-exposure: Implications for female reproductive health.

Microplastics and phthalates are prevalent and emerging pollutants that pose a potential impact on human health. Previous studies suggest that both microplastics and phthalates can adversely affect the reproductive systems of humans and mammals. However, the combined impact of these pollutants on the female reproductive system remains unclear. Here we show the impacts of exposure to polystyrene microplastics (PS-MPs) and di-2-ethylhexyl phthalate (DEHP) on female Sprague-Dawley rats' reproductive systems. We find that co-exposure to PS-MPs and DEHP results in a marked increase in cystic and atretic follicles, oxidative stress, fibrosis, and dysregulation of serum sex hormone homeostasis in the ovaries of the rats. Proteomic analysis identified differentially expressed proteins that were predominantly enriched in signaling pathways related to fatty acid metabolism and tight junctions, regulated by transforming growth factor ß1 (TGF-ß1). We further confirm that co-exposure to DEHP and PS-MPs activates the TGF-ß1/Smad3 signaling pathway, and inhibiting this pathway alleviates oxidative stress, hormonal dysregulation, and ovarian fibrosis. These results indicate that exposure to the combination of microplastics and phthalates leads to a significant increase in atretic follicles and may increase the risk of polycystic ovary syndrome (PCOS). Our study provides new insights into the reproductive toxicity effects of microplastics and DEHP exposure on female mammals, highlighting the potential link between environmental pollutants and the occurrence of PCOS. These findings highlight the need for comprehensive assessments of the reproductive health risks posed by microplastic pollution to women and contribute to the scientific basis for evaluating such risks.

Triple three-dimensional MS/MS spectrum facilitates quantitative ginsenosides-targeted sub-metabolome characterization in notoginseng.

Although serving as the workhorse, MS/MS cannot fully satisfy the analytical requirements of quantitative sub-metabolome characterization. Because more information intrinsically correlates to more structural and concentration clues, here, efforts were devoted to comprehensively tracing and deciphering MS/MS behaviors through constructing triple three-dimensional (3×3D)-MS/MS spectrum. Ginsenosides-targeted metabolomics of notoginseng, one of the most famous edible medicinal plants, was employed as a proof-of-concept. Serial authentic ginsenosides were deployed to build the correlations between 3×3D-MS/MS spectra and structure/concentration features. Through assaying ginsenosides with progressive concentrations using QTOF-MS to configure 1st 3D spectrum, the generations of MS1 spectral signals, particularly multi-charged multimer anions, e.g., [2M-2H]2- and [2M+2HCOO]2- ions, relied on both concentration and the amount of sugar chains. By programming progressive collision energies to the front collision cell of Qtrap-MS device to gain 2nd 3D spectrum, optimal collision energy (OCE) corresponding to the glycosidic bond fission was primarily correlated with the masses of precursor and fragment ions and partially governed by the glycosidation site. The quantitative relationships between OCEs and masses of precursor and fragment ions were utilized to build large-scale quantitative program for ginsenosides. After applying progressive exciting energies to the back collision chamber to build 3rd 3D spectrum, the fragment ion and the decomposition product anion exhibited identical dissociation trajectories when they shared the same molecular geometry. After ginsenosides-focused quantitative metabolomics, significant differences occurred for sub-metabolome amongst different parts of notoginseng. The differential ginsenosides were confirmatively identified by applying the correlations between 3×3D-MS/MS spectra and structures. Together, 3×3D-MS/MS spectrum covers all MS/MS behaviors and dramatically facilitates sub-metabolome characterization from both quantitative program development and structural identification.

Comprehensive Analysis of the Mechanism of Anoikis in Hepatocellular Carcinoma.

Background: Hepatocellular carcinoma (HCC), ranking as the second-leading cause of global mortality among malignancies, poses a substantial burden on public health worldwide. Anoikis, a type of programmed cell death, serves as a barrier against the dissemination of cancer cells to distant organs, thereby constraining the progression of cancer. Nevertheless, the mechanism of genes related to anoikis in HCC is yet to be elucidated. Methods: This paper's data (TCGA-HCC) were retrieved from the database of the Cancer Genome Atlas (TCGA). Differential gene expression with prognostic implications for anoikis was identified by performing both the univariate Cox and differential expression analyses. Through unsupervised cluster analysis, we clustered the samples according to these DEGs. By employing the least absolute shrinkage and selection operator Cox regression analysis (CRA), a clinical predictive gene signature was generated from the DEGs. The Cell-Type Identification by Estimating Relative Subsets of RNA Transcripts (CIBERSORT) algorithm was used to determine the proportions of immune cell types. The external validation data (GSE76427) were procured from Gene Expression Omnibus (GEO) to verify the performance of the clinical prognosis gene signature. Western blotting and immunohistochemistry (IHC) analysis confirmed the expression of risk genes. Results: In total, 23 prognostic DEGs were identified. Based on these 23 DEGs, the samples were categorized into four distinct subgroups (clusters 1, 2, 3, and 4). In addition, a clinical predictive gene signature was constructed utilizing ETV4, PBK, and SLC2A1. The gene signature efficiently distinguished individuals into two risk groups, specifically low and high, demonstrating markedly higher survival rates in the former group. Significant correlations were observed between the expression of these risk genes and a variety of immune cells. Moreover, the outcomes from the validation cohort analysis aligned consistently with those obtained from the training cohort analysis. The results of Western blotting and IHC showed that ETV4, PBK, and SLC2A1 were upregulated in HCC samples. Conclusion: The outcomes of this paper underscore the effectiveness of the clinical prognostic gene signature, established utilizing anoikis-related genes, in accurately stratifying patients. This signature holds promise in advancing the development of personalized therapy for HCC.

An approach for full space inverse materials design by combining universal machine learning potential, universal property model, and optimization algorithm.

We present a full space inverse materials design (FSIMD) approach that fully automates the materials design for target physical properties without the need to provide the atomic composition, chemical stoichiometry, and crystal structure in advance. Here, we used density functional theory reference data to train a universal machine learning potential (UPot) and transfer learning to train a universal bulk modulus model (UBmod). Both UPot and UBmod were able to cover materials systems composed of any element among 42 elements. Interfaced with optimization algorithm and enhanced sampling, the FSIMD approach is applied to find the materials with the largest cohesive energy and the largest bulk modulus, respectively. NaCl-type ZrC was found to be the material with the largest cohesive energy. For bulk modulus, diamond was identified to have the largest value. The FSIMD approach is also applied to design materials with other multi-objective properties with accuracy limited principally by the amount, reliability, and diversity of the training data. The FSIMD approach provides a new way for inverse materials design with other functional properties for practical applications.

OGG1: An emerging multifunctional therapeutic target for the treatment of diseases caused by oxidative DNA damage.

Oxidative DNA damage-related diseases, such as incurable inflammation, malignant tumors, and age-related disorders, present significant challenges in modern medicine due to their complex molecular mechanisms and limitations in identifying effective treatment targets. Recently, 8-oxoguanine DNA glycosylase 1 (OGG1) has emerged as a promising multifunctional therapeutic target for the treatment of these challenging diseases. In this review, we systematically summarize the multiple functions and mechanisms of OGG1, including pro-inflammatory, tumorigenic, and aging regulatory mechanisms. We also highlight the potential of OGG1 inhibitors and activators as potent therapeutic agents for the aforementioned life-limiting diseases. We conclude that OGG1 serves as a multifunctional hub; the inhibition of OGG1 may provide a novel approach for preventing and treating inflammation and cancer, and the activation of OGG1 could be a strategy for preventing age-related disorders. Furthermore, we provide an extensive overview of successful applications of OGG1 regulation in treating inflammatory, cancerous, and aging-related diseases. Finally, we discuss the current challenges and future directions of OGG1 as an emerging multifunctional therapeutic marker for the aforementioned challenging diseases. The aim of this review is to provide a robust reference for scientific researchers and clinical drug developers in the development of novel clinical targeted drugs for life-limiting diseases, especially for incurable inflammation, malignant tumors, and age-related disorders.

Chemical Isotope Labeling and Dual-Filtering Strategy for Comprehensive Profiling of Urinary Glucuronide Conjugates.

Glucuronidation, a crucial process in phase II metabolism, plays a vital role in the detoxification and elimination of endogenous substances and xenobiotics. A comprehensive and confident profiling of glucuronate-conjugated metabolites is imperative to understanding their roles in physiological and pathological processes. In this study, a chemical isotope labeling and dual-filtering strategy was developed for global profiling of glucuronide metabolites in biological samples. N,N-Dimethyl ethylenediamine (DMED-d0) and its deuterated counterpart DMED-d6 were used to label carboxylic acids through an amidation reaction. First, carboxyl-containing compounds were extracted based on a characteristic mass difference (Δm/z, 6.037 Da) observed in MS between light- and heavy-labeled metabolites (filter I). Subsequently, within the pool of carboxyl-containing compounds, glucuronides were identified using two pairs of diagnostic ions (m/z 247.1294/253.1665 and 229.1188/235.1559 for DMED-d0/DMED-d6-labeled glucuronides) originating from the fragmentation of the derivatized glucuronic acid group in MS/MS (filter II). Compared with non-derivatization, DEMD labeling significantly enhanced the detection sensitivity of glucuronides, as evidenced by a 3- to 55-fold decrease in limits of detection for representative standards. The strategy was applied to profiling glucuronide metabolites in urine samples from colorectal cancer (CRC) patients. A total of 685 features were screened as potential glucuronides, among which 181 were annotated, mainly including glucuronides derived from lipids, organic oxygen, and phenylpropanoids. Enzymatic biosynthesis was employed to accurately identify unknown glucuronides without standards, demonstrating the reliability of the dual-filtering strategy. Our strategy exhibits great potential for profiling the glucuronide metabolome with high coverage and confidence to reveal changes in CRC and other diseases.

Natural course of ulcerative colitis in China: Differences from the West?

BACKGROUND AND AIMS: Whether the natural course of ulcerative colitis (UC) in mainland China is similar or different from that in Western countries is unknown, and data on it is limited. We aimed to provide a comprehensive description of the natural course of UC in China and compare it with Western UC patients. METHODS: Based on a prospective Chinese nationwide registry of consecutive patients with inflammatory bowel diseases, the medical treatments and natural history of UC were described in detail, including disease extension, surgery, and neoplasia. The Cox regression model was used to identify factors associated with poor outcomes. RESULTS: A total of 1081 UC patients were included with a median follow-up duration of 5.3 years. The overall cumulative exposure was 99.1% to 5-aminosalicylic acids, 52.1% to corticosteroids, 25.6% to immunomodulators, and 15.4% to biologics. Disease extent at diagnosis was proctitis in 26.9%, left-sided colitis in 34.8%, and extensive colitis in 38.3%. Of 667 patients with proctitis and left-sided colitis, 380 (57.0%) experienced disease extent progression. A total of 58 (5.4%) UC patients underwent colectomy, demonstrating cumulative proportions of surgery at 1, 5, and 10 years after diagnosis of 0.6%, 3.4%, and 8.2%, respectively. In addition, 23 (2.1%) UC patients were diagnosed with neoplasia, demonstrating cumulative proportions of neoplasia at 1, 5, and 10 years after diagnosis of 0.5%, 1.0%, and 3.5%, respectively. CONCLUSIONS: Chinese UC patients had similar cumulative proportions of exposure to IBD-specific treatments but a lower surgical rate than patients in Western countries, indicating a different natural course, and close monitoring needs for UC in China. However, these results must be confirmed in population-based studies because the hospital-based cohort in our study might lead to selection bias.

MSE-VGG: A Novel Deep Learning Approach Based on EEG for Rapid Ischemic Stroke Detection.

Ischemic stroke is a type of brain dysfunction caused by pathological changes in the blood vessels of the brain which leads to brain tissue ischemia and hypoxia and ultimately results in cell necrosis. Without timely and effective treatment in the early time window, ischemic stroke can lead to long-term disability and even death. Therefore, rapid detection is crucial in patients with ischemic stroke. In this study, we developed a deep learning model based on fusion features extracted from electroencephalography (EEG) signals for the fast detection of ischemic stroke. Specifically, we recruited 20 ischemic stroke patients who underwent EEG examination during the acute phase of stroke and collected EEG signals from 19 adults with no history of stroke as a control group. Afterwards, we constructed correlation-weighted Phase Lag Index (cwPLI), a novel feature, to explore the synchronization information and functional connectivity between EEG channels. Moreover, the spatio-temporal information from functional connectivity and the nonlinear information from complexity were fused by combining the cwPLI matrix and Sample Entropy (SaEn) together to further improve the discriminative ability of the model. Finally, the novel MSE-VGG network was employed as a classifier to distinguish ischemic stroke from non-ischemic stroke data. Five-fold cross-validation experiments demonstrated that the proposed model possesses excellent performance, with accuracy, sensitivity, and specificity reaching 90.17%, 89.86%, and 90.44%, respectively. Experiments on time consumption verified that the proposed method is superior to other state-of-the-art examinations. This study contributes to the advancement of the rapid detection of ischemic stroke, shedding light on the untapped potential of EEG and demonstrating the efficacy of deep learning in ischemic stroke identification.

A DNA-Modularized STING Agonist with Macrophage-Selectivity and Programmability for Enhanced Anti-Tumor Immunotherapy.

The activation of cyclic GMP-AMP (cGAMP) synthase (cGAS) and its adaptor, stimulator of interferon genes (STING), is known to reprogram the immunosuppressive tumor microenvironment for promoting antitumor immunity. To enhance the efficiency of cGAS-STING pathway activation, macrophage-selective uptake, and programmable cytosolic release are crucial for the delivery of STING agonists. However, existing polymer- or lipid-based delivery systems encounter difficulty in integrating multiple functions meanwhile maintaining precise control and simple procedures. Herein, inspired by cGAS being a natural DNA sensor, a modularized DNA nanodevice agonist (DNDA) is designed that enable macrophage-selective uptake and programmable activation of the cGAS-STING pathway through precise self-assembly. The resulting DNA nanodevice acts as both a nanocarrier and agonist. Upon local administration, it demonstrates the ability of macrophage-selective uptake, endosomal escape, and cytosolic release of the cGAS-recognizing DNA segment, leading to robust activation of the cGAS-STING pathway and enhanced antitumor efficacy. Moreover, DNDA elicits a synergistic therapeutic effect when combined with immune checkpoint blockade. The study broadens the application of DNA nanotechnology as an immune stimulator for cGAS-STING activation.

Endothelial Cells Promote Pseudo-islet Function Through BTC-EGFR-JAK/STAT Signaling Pathways.

Interactions between cells are of fundamental importance in affecting cell function. In vivo, endothelial cells and islet cells are close to each other, which makes endothelial cells essential for islet cell development and maintenance of islet cell function. We used endothelial cells to construct 3D pseudo-islets, which demonstrated better glucose regulation and greater insulin secretion compared to conventional pseudo-islets in both in vivo and in vitro trials. However, the underlying mechanism of how endothelial cells promote beta cell function localized within islets is still unknown. We performed transcriptomic sequencing, differential gene analysis, and enrichment analysis on two types of pseudo-islets to show that endothelial cells can promote the function of internal beta cells in pseudo-islets through the BTC-EGFR-JAK/STAT signaling pathway. Min6 cells secreted additional BTC after co-culture of endothelial cells with MIN6 cells outside the body. After BTC knockout in vitro, we found that beta cells functioned differently: insulin secretion levels decreased significantly, while the expression of key proteins in the EGFR-mediated JAK/STAT signaling pathway simultaneously decreased, further confirming our results. Through our experiments, we elucidate the molecular mechanisms by which endothelial cells maintain islet function in vitro, which provides a theoretical basis for the construction of pseudo-islets and islet cell transplants for the treatment of diabetes mellitus.