检索结果-南通市图书馆

DDI-Transform:A neural network for predicting drug-drug interaction events

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Quantitative Biology 2024年第2期12卷 155-163页

作者： Jiaming su ying Qian School of Computer Science and Technology Shanghai Frontiers Science Center of Molecule Intelligent SynthesesEast China Normal UniversityShanghaiChina

Drug-drug interaction(DDI)event prediction is a challenging problem,and accurate prediction of DDI events is critical to patient health and new drug development.Recently,many machine learning-based techniques have been proposed for predicting DDI events.However,most of the existing methods do not effectively integrate the multidimensional features of drugs and provide poor mitigation of noise to get effective feature information.To address these limitations,we propose a DDI-Transform neural network framework for DDI event prediction.In DDI-Transform,we design a drug structure information feature extraction module and a drug bind-protein feature extraction module to obtain multidimensional feature information.A stack of DDI-Transform layers in the DDI-Transform network module are then used for adaptive learning,thus adaptively selecting the effective feature information for prediction.The results show that DDI-Transform can accurately predict DDI events and outperform the state-of-the-art models.Results on different scale datasets confirm the robustness of the method.

关键词： adaptive learning graph convolutional networks interaction prediction meta-path

Tungsten oxide/nitrogen-doped carbon nanotubes composite catalysts for enhanced redox kinetics in lithium-sulfur batteries

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Chinese Journal of Chemical Engineering 2024年第3期67卷 58-67页

作者： Deqing He Zihao Xie Qian Yang Wei Wang chao su School of Energy and Power Jiangsu University of Science and TechnologyZhenjiang 212100China State Key Laboratory of Materials-Oriented Chemical Engineering College of Chemical EngineeringNanjing Tech UniversityNanjing 211816China

The sluggish redox kinetics of polysulfides in lithium-sulfur(Li-S)batteries are a significant obstacle to their widespread adoption as energy storage devices.However,recent studies have shown that tungsten oxide(WO_(3))can facilitate the conversion kinetics of polysulfides in Li-S batteries.Herein,we fabricated host materials for sulfur using nitrogen-doped carbon nanotubes(N-CNTs)and WO_(3).We used low-cost components and simple procedures to overcome the poor electrical conductivity that is a disadvantage of metal oxides.The composites of WO_(3) and N-CNTs(WO_(3)/N-CNTs)create a stable framework structure,fast ion diffusion channels,and a 3D electron transport network during electrochemical reaction processes.As a result,the WO_(3)/N-CNT-Li2S6 cathode demonstrates high initial capacity(1162 mA·h·g^(-1) at 0.5℃),excellent rate performance(618 mA·h·g^(-1) at 5.5℃),and a low capacity decay rate(0.093%up to 600 cycles at 2℃).This work presents a novel approach for preparing tungsten oxide/carbon composite catalysts that facilitate the redox kinetics of polysulfide conversion.

关键词： Li-S batteries Composites Ion diffusion channels 3D electron transport network Redox kinetics

维普期刊数据库博看期刊

Small-sized Ni-Co/Mo_(2)C/Co_(6)Mo_(6)C_(2)@C for efficient alkaline and acidic hydrogen evolution reaction by an anchoring calcination strategy

在线全文

学校读者我要写书评

暂无评论

Frontiers of Chemical Science and Engineering 2024年第5期18卷 95-103页

作者： Jianxia Gu ying Zhu Haiyan Zheng Chunyi sun Zhongmin su Department of Chemistry Xinzhou Normal UniversityXinzhou 034000China School of Chemistry Northeast Normal UniversityChangchun 130024China

A novel,cheap and highly efficient Ni-Co/Mo_(2)C/Co_(6)Mo_(6)C_(2)@C nanocomposite has been successfully constructed through simple one-step carbonization method in a nitrogen atmosphere.Polyethyleneimine in the precursor can effectively anchor molybdenum-based Keggin-type polyoxometallate and NiCo-layered double hydroxide through electrostatic and coordination interactions,which avoids the aggregation of catalyst particles during the pyrolysis process.After optimization,the obtained Ni-Co/Mo_(2)C/Co_(6)Mo_(6)C_(2)@C possesses small size(3-8 nm),large specific surface area and hierarchical pore structure.More importantly,Ni-Co/Mo_(2)C/Co_(6)Mo_(6)C_(2)@C presents remarkable hydrogen evolution reaction activity with low overpotentials in 0.5 mol·L^(-1)H_(2)SO_(4)(102.3 mV)and 1 mol·L^(-1)KOH(95 mV)to afford the current density of 10 mA·cm^(-2),as well as small Tafel slopes of 82.49 and 99.92 mV·dec^(-1),respectively.Simultaneously,this catalyst also shows outstanding stability for 12 h without a significant change in current density.The excellent catalytic performance of Ni-Co/Mo_(2)C/Co_(6)Mo_(6)C_(2)@C can put down to the synergistic effect between multiple components and the small size of the catalyst.This work provides unique insights into the preparation of efficient transition metal-based catalysts for HER.

关键词： polyoxometallates layered double hydroxide transition metal-based electrocatalysts hydrogen evolution reaction

Automated Machine Learning Algorithm Using Recurrent Neural Network to Perform Long-Term Time Series Forecasting

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2024年第3期78卷 3529-3549页

作者： ying su Morgan C.Wang Shuai Liu Department of Statistics and Data Science University of Central FloridaOrlandoFL32816-2370USA School of Mathematics and Statistics Beijing Technology and Business UniversityBeijing100048China

Long-term time series forecasting stands as a crucial research domain within the realm of automated machine learning(AutoML).At present,forecasting,whether rooted in machine learning or statistical learning,typically relies on expert input and necessitates substantial manual involvement.This manual effort spans model development,feature engineering,hyper-parameter tuning,and the intricate construction of time series models.The complexity of these tasks renders complete automation unfeasible,as they inherently demand human intervention at multiple junctures.To surmount these challenges,this article proposes leveraging Long Short-Term Memory,which is the variant of Recurrent Neural Networks,harnessing memory cells and gating mechanisms to facilitate long-term time series prediction.However,forecasting accuracy by particular neural network and traditional models can degrade significantly,when addressing long-term time-series tasks.Therefore,our research demonstrates that this innovative approach outperforms the traditional Autoregressive Integrated Moving Average(ARIMA)method in forecasting long-term univariate time series.ARIMA is a high-quality and competitive model in time series prediction,and yet it requires significant preprocessing efforts.Using multiple accuracy metrics,we have evaluated both ARIMA and proposed method on the simulated time-series data and real data in both short and long term.Furthermore,our findings indicate its superiority over alternative network architectures,including Fully Connected Neural Networks,Convolutional Neural Networks,and Nonpooling Convolutional Neural Networks.Our AutoML approach enables non-professional to attain highly accurate and effective time series forecasting,and can be widely applied to various domains,particularly in business and finance.

关键词： Automated machine learning autoregressive integrated moving average neural networks time series analysis

Unraveling the hydraulic properties of loess for landslide prediction:A study on variations in loess landslides in Lanzhou,Dingxi,and Tianshui,China

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

China Geology 2024年第2期7卷 291-302页

作者： Gao-chao Lin Wei Liu Xing su School of Civil and Environmental Engineering University of New South WalesSydneyNSW 2052Australia Institute of Transportation Inner Mongolia UniversityHohhot 010070China Institute of Geological Hazards Prevention Gansu Academy of SciencesLanzhou 730000China

Loess has distinctive characteristics,leading to frequent landslide disasters and posing serious threats to the lives and properties of local re sidents.The involvement of water repre sents a critical factor in inducing loess landslides.This study focuses on three neighboring cities sequentially situated on the Loess Plateau along the direction of aeolian deposition of loess,namely Lanzhou,Dingxi,and Tianshui,which are densely populated and prone to landslide disasters.The variations in hydraulic properties,including water retention capacity and permeability,are investigated through Soil Water Characteristic Curve(SWCC)test and hydraulic conductivity test.The experimental findings revealed that Tianshui loess exhibited the highest water retention capacity,followed by Dingxi loess,while Lanzhou loess demonstrated the lowest water retention capacity.Contrastingly,the results for the saturated permeability coefficient were found to be the opposite:Tianshui loess showed the lowest permeability,whereas Lanzhou loess displayed the highest permeability.These results are supported and analyzed by scanning electron microscopy(SEM)observation.In addition,the water retention capacity is mathematically expressed using the van Genuchten model and extended to predict unsaturated hydraulic properties of loess.The experimental results exhibit a strong accordance with one another and align with the regional distribution patterns of disasters.

关键词： Loess Landslide Hydraulic properties Water retention capacity and permeability Soil Water Characteristic Curve(SWCC) Hydraulic conductivity Van Genuchten model Hydrogeological engineering Geological hazards prevention engineering

维普期刊数据库

A highly reduced Mo_(74) polyoxometalate featuring high proton conductivity accessed by building block strategy

在线全文

学校读者我要写书评

暂无评论

Science China Chemistry 2024年第3期67卷 862-868页

作者： Shi-Yi Liu Xue-Xin Li Wei-chao Chen Kui-Zhan Shao Xin-Long Wang chao Qin Zhong-Min su Key Laboratory of Polyoxometalate and Reticular Material Chemistry of Ministry of Education Department of ChemistryNortheast Normal UniversityChangchun 130024China State Key Laboratory of supramolecular Structure and Materials Institute of Theoretical ChemistryCollege of ChemistryJilin UniversityChangchun 130021China

Highly reduced polyoxometalates(POMs) are predicted to be used as rather high energy density materials;however,it still suffers from the limited cluster species and reduction ratio.Here we demonstrate that it is possible to employ the building block strategy to generate a highly reduced polyoxomolybdate(C_(2)H_(8)N)_(14)(NH_(4))_(4)H_(14)[Mo_(48)-ⅤMo_(26)ⅥO_(202)(OH)_(12)(SO_(4))_(6)]·46H_(2)O(Mo_(74)).The fundamental Mo-based{Mo_x}(x=4,5,and 6) building blocks,which are templated by tetra-coordinated anions{MoO_(4)}or{SO_(4)},not only lay foundation for the formation of Mo_(74) featuring an unprecedented reduction ratio of 65%,but also give rise to SBBs-mediated(secondary building blocks) supramolecular dense packing interactions among the isolated Mo_(74) clusters that are favorable for proton conduction.Remarkably,high proton conductivity(2.04×10^(-2)S cm^(-1)) had been realized at 50℃ and 90% relative humidity,revealing one of the well-known POMs-based crystalline proton conducting materials.This result highlights that this building block approach possesses great potential in producing highly reduced POM systems that can achieve controllable reduced ratio and desirable properties.

关键词： polyoxometalate building block molybdenum highly reduced state proton conduction

维普期刊数据库

Understanding adversarial attacks on observations in deep reinforcement learning

在线全文

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2024年第5期67卷 69-83页

作者： You QIAOBEN Chengyang ying Xinning ZHOU Hang su Jun ZHU Bo ZHANG Department of Computer Science and Technology Beijing National Research Center for Information Science and Technology Tsinghua-Bosch Joint Center for Machine Learning Institute for Artificial Intelligence Tsinghua University Peng Cheng Laboratory

Deep reinforcement learning models are vulnerable to adversarial attacks that can decrease the cumulative expected reward of a victim by manipulating its observations. Despite the efficiency of previous optimization-based methods for generating adversarial noise in supervised learning, such methods might not achieve the lowest cumulative reward since they do not generally explore the environmental dynamics.Herein, a framework is provided to better understand the existing methods by reformulating the problem of adversarial attacks on reinforcement learning in the function space. The reformulation approach adopted herein generates an optimal adversary in the function space of targeted attacks, repelling them via a generic two-stage framework. In the first stage, a deceptive policy is trained by hacking the environment and discovering a set of trajectories routing to the lowest reward or the worst-case performance. Next, the adversary misleads the victim to imitate the deceptive policy by perturbing the observations. Compared to existing approaches, it is theoretically shown that our adversary is strong under an appropriate noise level. Extensive experiments demonstrate the superiority of the proposed method in terms of efficiency and effectiveness, achieving state-of-the-art performance in both Atari and MuJoCo environments.

关键词： deep learning reinforcement learning adversarial robustness adversarial attack

First total synthesis,antitumor evaluation and target identification of mornaphthoate E:A new tubulin inhibitor template acting on PI3K/Akt signaling pathway

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Acta Pharmaceutica Sinica B 2024年第5期14卷 2177-2193页

作者： Peipei Shan Tao Ye ying-De Tang Hui Song chao Wang Kongkai Zhu Feifei Yang Shi-Lei Zhang Pei-Wen su Shuanhu Gao Hua Zhang Institute of Translational Medicine the Affiliated Hospital of Qingdao UniversityCollege of MedicineQingdao UniversityQingdao 266021China School of Biological Science and Technology University of JinanJinan 250022China Advanced Medical Research Institute Cheeloo College of MedicineShandong UniversityJinan 250012China College of Pharmaceutical Sciences Soochow UniversitySuzhou 215127China School of Chemistry and Molecular Engineering East China Normal UniversityShanghai 200062China

Mornaphthoate E(MPE)is a prenylated naphthoic acid methyl ester isolated from the roots of a famous Chinese medicinal plant Morinda officinalis and shows remarkable cytotoxicity against several human tumor cell lines.In the current project,the first total synthesis of(±)-MPE was achieved in seven steps and 5.6%overall yield.Then the in vitro anti-tumor activity of MPE was first assessed for both enantiomers in two breast cancer cells,with the levoisomer exerting slightly better potency.The in vivo anti-tumor effect was further verified by applying the racemate in an orthotopic autograft mouse model.Notably,MPE exerted promising anti-metastasis activity both in vitro and in vivo and showed no obvious toxicity on mice at the therapeutic dosage.Mechanistic investigations demonstrated that MPE acted as a tubulin polymerization stabilizer and disturbed the dynamic equilibrium of microtubules via regulating PI3K/Akt signaling.In conclusion,our work has provided a new chemical template for the future design and development of next-generation tubulin-targeting chemotherapies.

关键词： Mornaphthoate E Tubulin inhibitor Antitumor Anti-metastasis Breast cancer PI3K/Akt

Metabolic disorders in prediabetes:From mechanisms to therapeutic management

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

World Journal of Diabetes 2024年第3期15卷 361-377页

作者： Wen-Xin Ping Shan Hu Jing-Qian su Song-ying Ouyang Biomedical Research Center of South China College of Life SciencesFujian Normal UniversityFuzhou 350117Fujian ProvinceChina

Diabetes,one of the world's top ten diseases,is known for its high mortality and complication rates and low cure rate.Prediabetes precedes the onset of diabetes,during which effective treatment can reduce diabetes risk.Prediabetes risk factors include high-calorie and high-fat diets,sedentary lifestyles,and stress.Consequences may include considerable damage to vital organs,including the retina,liver,and kidneys.Interventions for treating prediabetes include a healthy lifestyle diet and pharmacological treatments.However,while these options are effective in the short term,they may fail due to the difficulty of long-term implementation.Medications may also be used to treat prediabetes.This review examines prediabetic treatments,particularly metformin,glucagon-like peptide-1 receptor agonists,sodium glucose cotransporter 2 inhibitors,vitamin D,and herbal medicines.Given the remarkable impact of prediabetes on the progression of diabetes mellitus,it is crucial to intervene promptly and effectively to regulate prediabetes.However,the current body of research on prediabetes is limited,and there is considerable confusion surrounding clinically relevant medications.This paper aims to provide a comprehensive summary of the pathogenesis of prediabetes mellitus and its associated therapeutic drugs.The ultimate goal is to facilitate the clinical utilization of medications and achieve efficient and timely control of diabetes mellitus.

关键词： Prediabetes Glucagon-like peptide agonists Sodium–glucose cotransporter 2 inhibitors Vitamin D Chinese herbal medicines