检索结果-南通市图书馆

GPS: graph contrastive learning via multi-scale augmented views from adversarial pooling

在线全文

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2025年第1期68卷 145-158页

作者： wei JU Yiyang GU Zhengyang MAO Ziyue QIAO Yifang QIN Xiao LUO hui XIONG Ming ZHANG School of Computer Science National Key Laboratory for Multimedia Information ProcessingPeking University Artificial Intelligence Thrust The Hong Kong University of Science and Technology Department of Computer Science University of California

Self-supervised graph representation learning has recently shown considerable promise in a range of fields, including bioinformatics and social networks. A large number of graph contrastive learning approaches have shown promising performance for representation learning on graphs, which train models by maximizing agreement between original graphs and their augmented views(i.e., positive views). Unfortunately, these methods usually involve pre-defined augmentation strategies based on the knowledge of human experts. Moreover, these strategies may fail to generate challenging positive views to provide sufficient supervision signals. In this paper, we present a novel approach named graph pooling contrast(GPS) to address these *** by the fact that graph pooling can adaptively coarsen the graph with the removal of redundancy, we rethink graph pooling and leverage it to automatically generate multi-scale positive views with varying emphasis on providing challenging positives and preserving semantics, i.e., strongly-augmented view and weakly-augmented view. Then, we incorporate both views into a joint contrastive learning framework with similarity learning and consistency learning, where our pooling module is adversarially trained with respect to the encoder for adversarial robustness. Experiments on twelve datasets on both graph classification and transfer learning tasks verify the superiority of the proposed method over its counterparts.

关键词： graph representation learning graph neural networks graph contrastive learning graph augmentations graph pooling

Rapid detection of colored and colorless macroand micro-plastics in complex environment via near-infrared spectroscopy and machine learning

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Journal of Environmental Sciences 2025年第1期147卷 512-522页

作者： hui-Huang Zou Pin-Jing He wei Peng Dong-Ying Lan Hao-Yang Xian Fan Lü Hua Zhang Institute of Waste Treatment&Reclamation College of Environmental Science and EngineeringTongji UniversityShanghai 200092China Shanghai Institute of Pollution Control and Ecological Security Shanghai 200092China

To better understand the migration behavior of plastic fragments in the environment,development of rapid non-destructive methods for in-situ identification and characterization of plastic fragments is ***,most of the studies had focused only on colored plastic fragments,ignoring colorless plastic fragments and the effects of different environmental media(backgrounds),thus underestimating their *** address this issue,the present study used near-infrared spectroscopy to compare the identification of colored and colorless plastic fragments based on partial least squares-discriminant analysis(PLS-DA),extreme gradient boost,support vector machine and random forest *** effects of polymer color,type,thickness,and background on the plastic fragments classification were ***-DA presented the best and most stable outcome,with higher robustness and lower misclassification *** models frequently misinterpreted colorless plastic fragments and its background when the fragment thickness was less than 0.1mm.A two-stage modeling method,which first distinguishes the plastic types and then identifies colorless plastic fragments that had been misclassified as background,was *** method presented an accuracy higher than 99%in different *** summary,this study developed a novel method for rapid and synchronous identification of colored and colorless plastic fragments under complex environmental backgrounds.

关键词： Colorless microplastics Near-infrared hyperspectral imaging Plastic identification Partial least squares discriminant analysis Machine learning

High-dimensional large-scale mixed-type data imputation under missing at random

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Science China Mathematics 2025年

作者： wei Liu Guizhen Li ling Zhou Lan Luo School of Mathematics Sichuan University Center of Statistical Research and School of Statistics Southwestern University of Finance and Economics Department of Statistics and Actuarial Science University of Iowa

Missingness in mixed-type variables is commonly encountered in a variety of areas. The requirement of complete observations necessities data imputation when a moderate or large proportion of data is ***, inappropriate imputation would downgrade the performance of machine learning algorithms, leading to bad predictions and unreliable statistical inference. For high-dimensional large-scale mixed-type missing data, we develop a computationally efficient imputation method, missing value imputation via generalized factor models（MIG）, under missing at random. The proposed MIG method allows missing variables to be of different types,including continuous, binary, and count variables, and are scalable to both data size n and variable dimension p while existing imputation methods rely on restrictive assumptions such as the same type of missing variables,the low dimensionality of variables, and a limited sample size. We explicitly show that the imputation error of the proposed MIG method diminishes to zero with the rate Op(max{n-1/2, p-1/2}) as both n and p tend to infinity. Five real datasets demonstrate the superior empirical performance of the proposed MIG method over existing methods that the average normalized absolute imputation error is reduced by 5.3%–34.1%.

关键词：

The cGAS-STING-interferon regulatory factor 7 pathway regulates neuroinflammation in Parkinson's disease

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Neural Regeneration Research 2025年第8期20卷 2361-2372页

作者： Shengyang Zhou Ting Li wei Zhang Jian Wu hui Hong wei Quan Xinyu Qiao Chun Cui Chenmeng Qiao weijiang Zhao Yanqin Shen Laboratory of Neurodegenerative and Neuroinjury Diseases Wuxi Medicine SchoolJiangnan UniversityWuxiJiangsu ProvinceChina

Interferon regulatory factor 7 plays a crucial role in the innate immune ***,whether interferon regulatory factor 7-mediated signaling contributes to Parkinson's disease remains *** we report that interferon regulatory factor 7 is markedly up-regulated in a 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine-induced mouse model of Parkinson's disease and co-localizes with microglial *** the selective cyclic guanosine monophosphate adenosine monophosphate synthase inhibitor RU.521 and the stimulator of interferon genes inhibitor H151 effectively suppressed interferon regulatory factor 7 activation in BV2 microglia exposed to 1-methyl-4-phenylpyridinium and inhibited transformation of mouse BV2 microglia into the neurotoxic M1 *** addition,si RNA-mediated knockdown of interferon regulatory factor 7 expression in BV2 microglia reduced the expression of inducible nitric oxide synthase,tumor necrosis factorα,CD16,CD32,and CD86 and increased the expression of the anti-inflammatory markers ARG1 and *** together,our findings indicate that the cyclic guanosine monophosphate adenosine monophosphate synthase-stimulator of interferon genes-interferon regulatory factor 7 pathway plays a crucial role in the pathogenesis of Parkinson's disease.

关键词： cyclic guanosine monophosphate adenosine monophosphate synthase H151 interferon regulatory factor 7 M1 phenotype neurodegenerative disease neuroinflammation Parkinson’s disease RU521 STING type I interferon

Chlorogenic acid induces hepatocellular carcinoma cell ferroptosis via PTGS2/AKR1C3/GPX4 axis-mediated reprogramming of arachidonic acid metabolism

在线全文

学校读者我要写书评

暂无评论

World Journal of Gastrointestinal Oncology 2025年第3期17卷 202-220页

作者： ling Wu Hong-Yao Chen Jing-Ting Zhang Ren-Yi Yang Zhi-Bin Wang Pei-Sen Xue wei Peng Ke-Xiong Li Wen-hui Gao Pu-Hua Zeng College of Traditional Chinese Medicine Hunan University of Chinese MedicineChangsha 410208Hunan ProvinceChina Department of Oncology Hunan Provincial Hospital of Integrated Traditional Chinese and Western ChineseChangsha 410006Hunan ProvinceChina Cancer Research Institute of Hunan Academy of Traditional Chinese Medicine Hunan Provincial Hospital of Integrated Traditional Chinese and Western ChineseChangsha 410006Hunan ProvinceChina

BACKGROUND Ferroptosis is an iron-dependent programmed non-apoptotic cell death characterized by the accumulation of free iron ions and lipid *** is associated with the inactivation of glutathione peroxidase(GPX)and the accumulation of lipid peroxides within *** is closely related to the occurrence and development of hepatocellular carcinoma(HCC).Chlorogenic acid(CGA),an important bioactive component found in 61 traditional Chinese medicines such as Eucommia ulmoides,has been extensively studied for its effects on various malignant ***,the specific role and potential mechanism of CGA in HCC remain *** To elucidate the anti-tumor characteristics and potential mechanisms of CGA in inducing ferroptosis in HCC *** The effects of CGA on the proliferation,migration,and invasion of HCC cells were evaluated through in vitro *** analysis combined with network pharmacology was used to study the potential targets and molecular mechanisms of CGA intervention in HCC *** vitro experiments were conducted to verify and explore the anti-HCC effects and mechanisms of CGA through the ferroptosis *** In vitro experiments showed that CGA dose-dependently inhibited the proliferation,invasion,and migration of HCC *** analysis combined with network pharmacology revealed that the pathway of CGA intervention in HCC cell ferroptosis was mainly enriched in the prostaglandin endoperoxide synthase 2(PTGS2)/aldoketo reductase family 1 member C3(AKR1C3)/GPX4 signaling pathway,which was associated with arachidonic *** vitro experiments further confirmed that CGA-induced ferroptosis in HCC cells was related to mitochondrial damage through the reprogramming of arachidonic acid metabolism by regulating the PTGS2/AKR1C3/GPX4 signaling *** This study demonstrates that CGA inhibits HCC cell proliferation,migration,and invasion by inducing ferroptosis through the PTGS2/A

关键词： Chlorogenic acid Hepatocellular carcinoma Ferroptosis Prostaglandin endoperoxide synthase 2/aldo-keto reductase family 1 member C3/glutathione peroxidase 4 axis Bioinformatics In vitro experiment

兰州北山不同坡向刺槐叶脉密度与气孔性状的关联性分析

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

植物生态学报 2016年第12期40卷 1289-1297页

作者：段贝贝赵成章徐婷郑慧玲冯威韩玲西北师范大学地理与环境科学学院甘肃省湿地资源保护与产业发展工程研究中心兰州730070

植物叶脉和气孔性状的关系反映了叶片的水力特性,对认识它们与植物水分利用有关的生理功能间的关系及其调控作用具有重要意义。该文利用GIS(geographic information system)与实验生态学相结合的方法,采用标准化主轴估计方法,研究了兰... 详细信息

植物叶脉和气孔性状的关系反映了叶片的水力特性,对认识它们与植物水分利用有关的生理功能间的关系及其调控作用具有重要意义。该文利用GIS(geographic information system)与实验生态学相结合的方法,采用标准化主轴估计方法,研究了兰州市北山不同坡向人工林刺槐(Robinia pseudoacacia)叶脉密度与气孔密度、气孔大小的关系。结果表明:随着坡向由南坡向东坡、西坡和北坡转变,植被群落的郁闭度、高度和土壤含水量呈逐渐增加的趋势,刺槐的净光合速率(Pn)、蒸腾速率(Tr)、光合有效辐射(PAR)、叶脉密度和气孔密度呈逐渐减小的趋势,气孔与叶面积呈逐渐增大的趋势;各个坡向的刺槐叶脉密度与气孔密度呈显著正相关关系,与气孔大小呈显著负相关关系,且在南坡达到极显著相关关系。生长在南坡的刺槐具有高的叶脉密度和密而小的气孔,生长在北坡的刺槐具有低的叶脉密度和疏而大的气孔。不同坡向刺槐叶脉密度与气孔特征间的资源分配模式,反映了植物在异质性生境中根据其功能需求在自身性状之间进行投资权衡机制的优化。

关键词：刺槐叶脉密度气孔密度气孔大小坡向

Coagulation indices and fibrinogen degradation products as predictive biomarkers for tumor-node-metastasis staging and metastasis in gastric cancer

在线全文

学校读者我要写书评

暂无评论

World Journal of Gastrointestinal Oncology 2025年第1期17卷 110-120页

作者： Yi-Qing Shen Qiu-Wan wei Yi-Ren Tian Yun-Zhi ling Min Zhang Clinical Laboratory Civil Aviation Shanghai HospitalGubei Branch of Ruijin Hospital Affiliated to Shanghai Jiao Tong University School of MedicineShanghai 200000China Clinical Laboratory Tongji Hospital of Tongji UniversityShanghai 200000China

BACKGROUND Gastric cancer(GC)is a prevalent malignancy with a substantial health burden and high mortality rate,despite advances in prevention,early detection,and *** with the global average,Asia,notably China,reports disproportionately high GC *** disease often progresses asymptoma-tically in the early stages,leading to delayed diagnosis and compromised ***,it is crucial to identify early diagnostic biomarkers and enhance treatment strategies to improve patient outcomes and reduce *** Retrospectively analyzed the clinical data of 148 patients with GC treated at the Civil Aviation Shanghai Hospital between December 2022 and December *** associations of coagulation indices-partial thromboplastin time(APTT),prothrombin time(PT),thrombin time(TT),fibrinogen,fibrinogen degradation products(FDP),fasting blood glucose,and D-dimer(D-D)with TNM stage and distant metastasis were *** Prolongation of APTT,PT,and TT was significantly correlated with the GC TNM ***,abnormal coagulation system activation was closely related to disease *** FDP and D-D were significantly associated with distant metastasis in GC(P<0.05),suggesting that increased fibrinolytic activity contributes to increased metastatic *** Our Results reveal coagulation indices,FDPs as GC biomarkers,reflecting abnormal coagulation/fibrinolysis,aiding disease progression,metastasis prediction,and helping clinicians assess thrombotic risk for early intervention and personalized treatment plans.

关键词： Coagulation indexes Fibrinogen degradation products Gastric cancer Tumor-node-metastasis staging Distant metastasis

The interaction between KIF21A and KANK1 regulates dendritic morphology and synapse plasticity in neurons

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Neural Regeneration Research 2025年第1期20卷 209-223页

作者： Shi-Yan Sun lingyun Nie Jing Zhang Xue Fang Hongmei Luo Chuanhai Fu Zhiyi wei Ai-hui Tang Hefei National Research Center for Physical Sciences at the Microscale CAS Key Laboratory of Brain Function and DiseaseMinistry of Education Key Laboratory for Membrane-less Organelles and Cellular DynamicsDivision of Life Sciences and MedicineUniversity of Science and Technology of ChinaHefeiAnhui ProvinceChina Institute of Artificial Intelligence Hefei Comprehensive National Science CenterHefeiAnhui ProvinceChina CAS Center for Excellence in Molecular Cell Sciences Division of Life Sciences and MedicineUniversity of Science and Technology of ChinaHefeiAnhui ProvinceChina Department of Neurobiology School of Life SciencesSouthern University of Science and TechnologyShenzhenGuangdong ProvinceChina Brain Research Center Southern University of Science and TechnologyShenzhenGuangdong ProvinceChina

Morphological alterations in dendritic spines have been linked to changes in functional communication between neurons that affect learning and ***-4 KIF21A helps organize the microtubule-actin network at the cell cortex by interacting with KANK1;however,whether KIF21A modulates dendritic structure and function in neurons remains *** this study,we found that KIF21A was distributed in a subset of dendritic spines,and that these KIF21A-positive spines were larger and more structurally plastic than KIF21A-negative ***,the interaction between KIF21A and KANK1 was found to be critical for dendritic spine morphogenesis and synaptic *** of either KIF21A or KANK1 inhibited dendritic spine morphogenesis and dendritic branching,and these deficits were fully rescued by coexpressing full-length KIF21A or KANK1,but not by proteins with mutations disrupting direct binding between KIF21A and KANK1 or binding between KANK1 and *** down KIF21A in the hippocampus of rats inhibited the amplitudes of long-term potentiation induced by high-frequency stimulation and negatively impacted the animals’cognitive *** together,our findings demonstrate the function of KIF21A in modulating spine morphology and provide insight into its role in synaptic function.

关键词： actin cytoskeleton dendrite KANK1 KIF21A microtubule spine morphology spine synaptic plasticity talin1

Dynamic event-triggered data-driven iterative learning bipartite tracking control for nonlinear MASs with prescribed performance

在线全文

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2025年第1期68卷 292-305页

作者： Tao SHI wei-wei CHE School of Mechatronic Engineering and Automation Shanghai University State Key Laboratory of Synthetical Automation for Process Industries Northeastern University College of Information Science and Engineering Northeastern University

This article proposes a distributed dynamic event-triggered data-driven iterative learning control(DET-DDILC)scheme under a predefined performance to tackle the bipartite tracking control problem for multiagent systems(MASs). An improved dynamic linearization technique is utilized to convert the nonlinear MASs into an iterative linear data model. First,a peer-to-peer mapping function is introduced to map the constrained distributed system output homeomorphism to an unconstrained one. In addition, a DET mechanism based on a time-iteration-varying function is devised to conserve network communication resources. Based on the unconstrained transformation and the designed DET mechanism, the DET-DDILC algorithm is devised to ensure that the bipartite tracking performance of MASs can be within the preset range. Finally, the effectiveness and feasibility of the designed control scheme are demonstrated via a simulation case by a comparison.

关键词： bipartite tracking control prescribed performance event-triggered iterative learning control multiagent systems