近年来,深度学习模型在农药发现和从头分子设计方面取得了显著进展。然而目前用于农药分子设计的深度生成模型中,基于骨架的分子生成模型较少。并且基于骨架的分子生成方法面临着生成分子质量和多样性不足的挑战。为此,该研究提出了一...近年来,深度学习模型在农药发现和从头分子设计方面取得了显著进展。然而目前用于农药分子设计的深度生成模型中,基于骨架的分子生成模型较少。并且基于骨架的分子生成方法面临着生成分子质量和多样性不足的挑战。为此,该研究提出了一种基于骨架结构的循环神经网络模型(multi head attention-recurrent neural network,MHA-RNN),首先生成简化分子线性输入规范(simplified molecular input line entry system,SMILES)格式的分子骨架,然后对骨架进行装饰以生成新的分子。试验结果表明,模型生成的分子在有效性、新颖性和唯一性方面分别达到了97.18%、99.87%和100.00%。此外,生成分子在脂水分配系数(logarithm of partition coefficient,LogP)、拓扑极性表面积(topological polar surface area,TPSA)、相对分子质量(molecular weight,MW)、类药性(quantitative estimate of drug-likeness,QED)、氢键受体(hydrogen bond acceptor,HBA)、氢键供体(hydrogen bond donor,HBD)、旋转键数(rotatable bonds,RotB)等性质上的分布与现有分子高度相似,研究结果为农药新药研发提供了技术支持与参考。展开更多
为满足铁路接触网腕臂智能检修作业中机械臂自动导航需求,提出一种综合解决路径规划和障碍物避让问题的研究方法。该方法将双重目标转化为单一的约束优化问题。在此基础上,对标准快速搜索随机树(Rapidly exploring Random Tree,RRT)算...为满足铁路接触网腕臂智能检修作业中机械臂自动导航需求,提出一种综合解决路径规划和障碍物避让问题的研究方法。该方法将双重目标转化为单一的约束优化问题。在此基础上,对标准快速搜索随机树(Rapidly exploring Random Tree,RRT)算法进行改进,引入地图复杂程度评估策略和高斯混合分布采样策略,以约束随机采样点的生成方向。通过加入角度约束策略和临近障碍物的变步长机制,确保随机树始终向目标点方向生长,从而规划出渐进最优的路径。此外,设计一种基于甲虫嗅觉探测的递归神经网络(Recurrent Neural Network based on Beetle Olfactory Detection,RNNBOD)算法,配置最优关节角度,驱动冗余机械臂末端执行器沿规划的参考路径移动,从而降低其计算成本。仿真结果表明,该方法不仅有效提升了标准RRT算法的搜索效率、节点利用率和路径质量,还成功解决了冗余机械臂在运行过程中的跟踪控制难题。展开更多
为解决传统特高压直流保护对高阻故障检测准确率不高、故障检测时间过长以及故障选极不完善的问题,提出基于长短时记忆(long short term memory,LSTM)循环神经网络(recurrent neural network,RNN)的特高压直流输电线路继电保护故障检测...为解决传统特高压直流保护对高阻故障检测准确率不高、故障检测时间过长以及故障选极不完善的问题,提出基于长短时记忆(long short term memory,LSTM)循环神经网络(recurrent neural network,RNN)的特高压直流输电线路继电保护故障检测方法。首先,基于快速傅里叶变换分析特高压直流输电系统暂态故障特征,使用相模变换和小波变换提取出故障特征量作为输入数据。其次,将输入数据输入到LSTM-RNN中进行前向传播,对系统故障特征进行深度学习,同时使用反向传播方式更新网络参数,将深层的特征量输入到Softmax分类器中进行分类,把故障识别分成区外故障、母线故障和线路故障,故障分类为正极故障、负极故障和双极故障,并输出识别结果。最后,在PSCAD/EMTDC仿真条件下,搭建特高压直流输电模型。验证结果表明:所提的方法在特高压直流输电线路继电保护的故障检测、故障选极上具有更好的效果,相比于人工神经网络、卷积神经网络、支持向量机,故障识别准确率分别提升4.71%、6.57%、9.32%。展开更多
Projective synchronization problems of a drive system and a particular response network were investigated,where the drive system is an arbitrary system with n+1 dimensions;it may be a linear or nonlinear system,and ev...Projective synchronization problems of a drive system and a particular response network were investigated,where the drive system is an arbitrary system with n+1 dimensions;it may be a linear or nonlinear system,and even a chaotic or hyperchaotic system,the response network is complex system coupled by N nodes,and every node is showed by the approximately linear part of the drive system.Only controlling any one node of the response network by designed controller can achieve the projective synchronization.Some numerical examples were employed to verify the effectiveness and correctness of the designed controller.展开更多
The burgeoning development of nanomedicine has provided state-of-the-art technologies and innovative methodologies for contemporary biomedical research,presenting unprecedented opportunities for resolving pivotal biom...The burgeoning development of nanomedicine has provided state-of-the-art technologies and innovative methodologies for contemporary biomedical research,presenting unprecedented opportunities for resolving pivotal biomedical challenges.Nanomaterials possess distinctive structures and properties.Through the exploration of the fabrication of emerging nanomedicines,multiple functions can be integrated to enable more precise diagnosis and treatment,thereby compensating for the limitations of traditional treatment modalities.Among various substances,polyphenols are natural organic compounds classified as plant secondary metabolites and are ubiquitously present in vegetables,teas,and other plants.Polyphenols are rich in active groups,including hydroxyl,carboxyl,amino,and conjugated double bonds.They exhibit robust adhesion,antioxidant,anti-inflammatory,and antibacterial biological activities and are extensively applied in pharmaceutical formulations.Additionally,polyphenols are characterized by their low cost,ready availability,and do not necessitate intricate chemical synthesis processes.Nevertheless,when natural polyphenol-based nanomedicines are utilized in isolation,they encounter several issues.These include poor water solubility,feeble stability,low bioavailability,the requirement for high dosages,and difficulties in precisely reaching the site of action.To address these concerns,researchers have developed nanomedicines by combining metal ions and functional ligands through metal coordination strategies.Nanomaterials,owing to their unique electronic and optical properties,have been successfully introduced into the realm of medical biology.Nano preparations not only enhance the stability of natural products but also endow them with targeting capabilities,thus enabling precise drug delivery.Polyphenols can further synergize with metal ions,anti-cancer drugs,or photosensitizers via supramolecular interactions to achieve multifunctional synergistic therapies,such as targeted drug delivery,efficacy enhancement,and the construction of engineering scaffolds.Metal-Polyphenol Coordination Polymers(MPCPs),composed of metal ions and phenolic ligands,are regarded as ideal nanoplatforms for disease diagnosis and treatment.In recent years,MPCPs have attracted extensive research in the biomedical field on account of their advantages,including facile synthesis,adjustable structure,excellent biocompatibility,and pH responsiveness.In this review,the classification and preparation strategies of MPCPs were systematically presented.Subsequently,their remarkable achievements in biomedical domains,such as bioimaging,biosensing,drug delivery,tumor therapy,and antimicrobial applications were highlighted.Finally,the principal limitations and prospects of MPCPs were comprehensi vely discussed.展开更多
Human disturbance activities is one of the main reasons for inducing geohazards.Ecological impact assessment metrics of roads are inconsistent criteria and multiple.From the perspective of visual observation,the envir...Human disturbance activities is one of the main reasons for inducing geohazards.Ecological impact assessment metrics of roads are inconsistent criteria and multiple.From the perspective of visual observation,the environment damage can be shown through detecting the uncovered area of vegetation in the images along road.To realize this,an end-to-end environment damage detection model based on convolutional neural network is proposed.A 50-layer residual network is used to extract feature map.The initial parameters are optimized by transfer learning.An example is shown by this method.The dataset including cliff and landslide damage are collected by us along road in Shennongjia national forest park.Results show 0.4703 average precision(AP)rating for cliff damage and 0.4809 average precision(AP)rating for landslide damage.Compared with YOLOv3,our model shows a better accuracy in cliff and landslide detection although a certain amount of speed is sacrificed.展开更多
Objective Traditional Chinese medicine(TCM)constitutes a valuable cultural heritage and an important source of antitumor compounds.Poria(Poria cocos(Schw.)Wolf),the dried sclerotium of a polyporaceae fungus,was first ...Objective Traditional Chinese medicine(TCM)constitutes a valuable cultural heritage and an important source of antitumor compounds.Poria(Poria cocos(Schw.)Wolf),the dried sclerotium of a polyporaceae fungus,was first documented in Shennong’s Classic of Materia Medica and has been used therapeutically and dietarily in China for millennia.Traditionally recognized for its diuretic,spleen-tonifying,and sedative properties,modern pharmacological studies confirm that Poria exhibits antioxidant,anti-inflammatory,antibacterial,and antitumor activities.Pachymic acid(PA;a triterpenoid with the chemical structure 3β-acetyloxy-16α-hydroxy-lanosta-8,24(31)-dien-21-oic acid),isolated from Poria,is a principal bioactive constituent.Emerging evidence indicates PA exerts antitumor effects through multiple mechanisms,though these remain incompletely characterized.Neuroblastoma(NB),a highly malignant pediatric extracranial solid tumor accounting for 15%of childhood cancer deaths,urgently requires safer therapeutics due to the limitations of current treatments.Although PA shows multi-mechanistic antitumor potential,its efficacy against NB remains uncharacterized.This study systematically investigated the potential molecular targets and mechanisms underlying the anti-NB effects of PA by integrating network pharmacology-based target prediction with experimental validation of multi-target interactions through molecular docking,dynamic simulations,and in vitro assays,aimed to establish a novel perspective on PA’s antitumor activity and explore its potential clinical implications for NB treatment by integrating computational predictions with biological assays.Methods This study employed network pharmacology to identify potential targets of PA in NB,followed by validation using molecular docking,molecular dynamics(MD)simulations,MM/PBSA free energy analysis,RT-qPCR and Western blot experiments.Network pharmacology analysis included target screening via TCMSP,GeneCards,DisGeNET,SwissTargetPrediction,SuperPred,and PharmMapper.Subsequently,potential targets were predicted by intersecting the results from these databases via Venn analysis.Following target prediction,topological analysis was performed to identify key targets using Cytoscape software.Molecular docking was conducted using AutoDock Vina,with the binding pocket defined based on crystal structures.MD simulations were performed for 100 ns using GROMACS,and RMSD,RMSF,SASA,and hydrogen bonding dynamics were analyzed.MM/PBSA calculations were carried out to estimate the binding free energy of each protein-ligand complex.In vitro validation included RT-qPCR and Western blot,with GAPDH used as an internal control.Results The CCK-8 assay demonstrated a concentration-dependent inhibitory effect of PA on NB cell viability.GO analysis suggested that the anti-NB activity of PA might involve cellular response to chemical stress,vesicle lumen,and protein tyrosine kinase activity.KEGG pathway enrichment analysis suggested that the anti-NB activity of PA might involve the PI3K/AKT,MAPK,and Ras signaling pathways.Molecular docking and MD simulations revealed stable binding interactions between PA and the core target proteins AKT1,EGFR,SRC,and HSP90AA1.RT-qPCR and Western blot analyses further confirmed that PA treatment significantly decreased the mRNA and protein expression of AKT1,EGFR,and SRC while increasing the HSP90AA1 mRNA and protein levels.Conclusion It was suggested that PA may exert its anti-NB effects by inhibiting AKT1,EGFR,and SRC expression,potentially modulating the PI3K/AKT signaling pathway.These findings provide crucial evidence supporting PA’s development as a therapeutic candidate for NB.展开更多
Social interaction with peer pressure is widely studied in social network analysis.Game theory can be utilized to model dynamic social interaction,and one class of game network models assumes that people’s decision p...Social interaction with peer pressure is widely studied in social network analysis.Game theory can be utilized to model dynamic social interaction,and one class of game network models assumes that people’s decision payoff functions hinge on individual covariates and the choices of their friends.However,peer pressure would be misidentified and induce a non-negligible bias when incomplete covariates are involved in the game model.For this reason,we develop a generalized constant peer effects model based on homogeneity structure in dynamic social networks.The new model can effectively avoid bias through homogeneity pursuit and can be applied to a wider range of scenarios.To estimate peer pressure in the model,we first present two algorithms based on the initialize expand merge method and the polynomial-time twostage method to estimate homogeneity parameters.Then we apply the nested pseudo-likelihood method and obtain consistent estimators of peer pressure.Simulation evaluations show that our proposed methodology can achieve desirable and effective results in terms of the community misclassification rate and parameter estimation error.We also illustrate the advantages of our model in the empirical analysis when compared with a benchmark model.展开更多
In this paper,we propose a neural network approach to learn the parameters of a class of stochastic Lotka-Volterra systems.Approximations of the mean and covariance matrix of the observational variables are obtained f...In this paper,we propose a neural network approach to learn the parameters of a class of stochastic Lotka-Volterra systems.Approximations of the mean and covariance matrix of the observational variables are obtained from the Euler-Maruyama discretization of the underlying stochastic differential equations(SDEs),based on which the loss function is built.The stochastic gradient descent method is applied in the neural network training.Numerical experiments demonstrate the effectiveness of our method.展开更多
In the field of image processing,the analysis of Synthetic Aperture Radar(SAR)images is crucial due to its broad range of applications.However,SAR images are often affected by coherent speckle noise,which significantl...In the field of image processing,the analysis of Synthetic Aperture Radar(SAR)images is crucial due to its broad range of applications.However,SAR images are often affected by coherent speckle noise,which significantly degrades image quality.Traditional denoising methods,typically based on filter techniques,often face challenges related to inefficiency and limited adaptability.To address these limitations,this study proposes a novel SAR image denoising algorithm based on an enhanced residual network architecture,with the objective of enhancing the utility of SAR imagery in complex electromagnetic environments.The proposed algorithm integrates residual network modules,which directly process the noisy input images to generate denoised outputs.This approach not only reduces computational complexity but also mitigates the difficulties associated with model training.By combining the Transformer module with the residual block,the algorithm enhances the network's ability to extract global features,offering superior feature extraction capabilities compared to CNN-based residual modules.Additionally,the algorithm employs the adaptive activation function Meta-ACON,which dynamically adjusts the activation patterns of neurons,thereby improving the network's feature extraction efficiency.The effectiveness of the proposed denoising method is empirically validated using real SAR images from the RSOD dataset.The proposed algorithm exhibits remarkable performance in terms of EPI,SSIM,and ENL,while achieving a substantial enhancement in PSNR when compared to traditional and deep learning-based algorithms.The PSNR performance is enhanced by over twofold.Moreover,the evaluation of the MSTAR SAR dataset substantiates the algorithm's robustness and applicability in SAR denoising tasks,with a PSNR of 25.2021 being attained.These findings underscore the efficacy of the proposed algorithm in mitigating speckle noise while preserving critical features in SAR imagery,thereby enhancing its quality and usability in practical scenarios.展开更多
Low Earth orbit(LEO)satellite networks exhibit distinct characteristics,e.g.,limited resources of individual satellite nodes and dynamic network topology,which have brought many challenges for routing algorithms.To sa...Low Earth orbit(LEO)satellite networks exhibit distinct characteristics,e.g.,limited resources of individual satellite nodes and dynamic network topology,which have brought many challenges for routing algorithms.To satisfy quality of service(QoS)requirements of various users,it is critical to research efficient routing strategies to fully utilize satellite resources.This paper proposes a multi-QoS information optimized routing algorithm based on reinforcement learning for LEO satellite networks,which guarantees high level assurance demand services to be prioritized under limited satellite resources while considering the load balancing performance of the satellite networks for low level assurance demand services to ensure the full and effective utilization of satellite resources.An auxiliary path search algorithm is proposed to accelerate the convergence of satellite routing algorithm.Simulation results show that the generated routing strategy can timely process and fully meet the QoS demands of high assurance services while effectively improving the load balancing performance of the link.展开更多
文摘近年来,深度学习模型在农药发现和从头分子设计方面取得了显著进展。然而目前用于农药分子设计的深度生成模型中,基于骨架的分子生成模型较少。并且基于骨架的分子生成方法面临着生成分子质量和多样性不足的挑战。为此,该研究提出了一种基于骨架结构的循环神经网络模型(multi head attention-recurrent neural network,MHA-RNN),首先生成简化分子线性输入规范(simplified molecular input line entry system,SMILES)格式的分子骨架,然后对骨架进行装饰以生成新的分子。试验结果表明,模型生成的分子在有效性、新颖性和唯一性方面分别达到了97.18%、99.87%和100.00%。此外,生成分子在脂水分配系数(logarithm of partition coefficient,LogP)、拓扑极性表面积(topological polar surface area,TPSA)、相对分子质量(molecular weight,MW)、类药性(quantitative estimate of drug-likeness,QED)、氢键受体(hydrogen bond acceptor,HBA)、氢键供体(hydrogen bond donor,HBD)、旋转键数(rotatable bonds,RotB)等性质上的分布与现有分子高度相似,研究结果为农药新药研发提供了技术支持与参考。
文摘为满足铁路接触网腕臂智能检修作业中机械臂自动导航需求,提出一种综合解决路径规划和障碍物避让问题的研究方法。该方法将双重目标转化为单一的约束优化问题。在此基础上,对标准快速搜索随机树(Rapidly exploring Random Tree,RRT)算法进行改进,引入地图复杂程度评估策略和高斯混合分布采样策略,以约束随机采样点的生成方向。通过加入角度约束策略和临近障碍物的变步长机制,确保随机树始终向目标点方向生长,从而规划出渐进最优的路径。此外,设计一种基于甲虫嗅觉探测的递归神经网络(Recurrent Neural Network based on Beetle Olfactory Detection,RNNBOD)算法,配置最优关节角度,驱动冗余机械臂末端执行器沿规划的参考路径移动,从而降低其计算成本。仿真结果表明,该方法不仅有效提升了标准RRT算法的搜索效率、节点利用率和路径质量,还成功解决了冗余机械臂在运行过程中的跟踪控制难题。
文摘为解决传统特高压直流保护对高阻故障检测准确率不高、故障检测时间过长以及故障选极不完善的问题,提出基于长短时记忆(long short term memory,LSTM)循环神经网络(recurrent neural network,RNN)的特高压直流输电线路继电保护故障检测方法。首先,基于快速傅里叶变换分析特高压直流输电系统暂态故障特征,使用相模变换和小波变换提取出故障特征量作为输入数据。其次,将输入数据输入到LSTM-RNN中进行前向传播,对系统故障特征进行深度学习,同时使用反向传播方式更新网络参数,将深层的特征量输入到Softmax分类器中进行分类,把故障识别分成区外故障、母线故障和线路故障,故障分类为正极故障、负极故障和双极故障,并输出识别结果。最后,在PSCAD/EMTDC仿真条件下,搭建特高压直流输电模型。验证结果表明:所提的方法在特高压直流输电线路继电保护的故障检测、故障选极上具有更好的效果,相比于人工神经网络、卷积神经网络、支持向量机,故障识别准确率分别提升4.71%、6.57%、9.32%。
基金Supported by the National Natural Science Foundation of China (11161027)。
文摘Projective synchronization problems of a drive system and a particular response network were investigated,where the drive system is an arbitrary system with n+1 dimensions;it may be a linear or nonlinear system,and even a chaotic or hyperchaotic system,the response network is complex system coupled by N nodes,and every node is showed by the approximately linear part of the drive system.Only controlling any one node of the response network by designed controller can achieve the projective synchronization.Some numerical examples were employed to verify the effectiveness and correctness of the designed controller.
文摘The burgeoning development of nanomedicine has provided state-of-the-art technologies and innovative methodologies for contemporary biomedical research,presenting unprecedented opportunities for resolving pivotal biomedical challenges.Nanomaterials possess distinctive structures and properties.Through the exploration of the fabrication of emerging nanomedicines,multiple functions can be integrated to enable more precise diagnosis and treatment,thereby compensating for the limitations of traditional treatment modalities.Among various substances,polyphenols are natural organic compounds classified as plant secondary metabolites and are ubiquitously present in vegetables,teas,and other plants.Polyphenols are rich in active groups,including hydroxyl,carboxyl,amino,and conjugated double bonds.They exhibit robust adhesion,antioxidant,anti-inflammatory,and antibacterial biological activities and are extensively applied in pharmaceutical formulations.Additionally,polyphenols are characterized by their low cost,ready availability,and do not necessitate intricate chemical synthesis processes.Nevertheless,when natural polyphenol-based nanomedicines are utilized in isolation,they encounter several issues.These include poor water solubility,feeble stability,low bioavailability,the requirement for high dosages,and difficulties in precisely reaching the site of action.To address these concerns,researchers have developed nanomedicines by combining metal ions and functional ligands through metal coordination strategies.Nanomaterials,owing to their unique electronic and optical properties,have been successfully introduced into the realm of medical biology.Nano preparations not only enhance the stability of natural products but also endow them with targeting capabilities,thus enabling precise drug delivery.Polyphenols can further synergize with metal ions,anti-cancer drugs,or photosensitizers via supramolecular interactions to achieve multifunctional synergistic therapies,such as targeted drug delivery,efficacy enhancement,and the construction of engineering scaffolds.Metal-Polyphenol Coordination Polymers(MPCPs),composed of metal ions and phenolic ligands,are regarded as ideal nanoplatforms for disease diagnosis and treatment.In recent years,MPCPs have attracted extensive research in the biomedical field on account of their advantages,including facile synthesis,adjustable structure,excellent biocompatibility,and pH responsiveness.In this review,the classification and preparation strategies of MPCPs were systematically presented.Subsequently,their remarkable achievements in biomedical domains,such as bioimaging,biosensing,drug delivery,tumor therapy,and antimicrobial applications were highlighted.Finally,the principal limitations and prospects of MPCPs were comprehensi vely discussed.
文摘Human disturbance activities is one of the main reasons for inducing geohazards.Ecological impact assessment metrics of roads are inconsistent criteria and multiple.From the perspective of visual observation,the environment damage can be shown through detecting the uncovered area of vegetation in the images along road.To realize this,an end-to-end environment damage detection model based on convolutional neural network is proposed.A 50-layer residual network is used to extract feature map.The initial parameters are optimized by transfer learning.An example is shown by this method.The dataset including cliff and landslide damage are collected by us along road in Shennongjia national forest park.Results show 0.4703 average precision(AP)rating for cliff damage and 0.4809 average precision(AP)rating for landslide damage.Compared with YOLOv3,our model shows a better accuracy in cliff and landslide detection although a certain amount of speed is sacrificed.
文摘Objective Traditional Chinese medicine(TCM)constitutes a valuable cultural heritage and an important source of antitumor compounds.Poria(Poria cocos(Schw.)Wolf),the dried sclerotium of a polyporaceae fungus,was first documented in Shennong’s Classic of Materia Medica and has been used therapeutically and dietarily in China for millennia.Traditionally recognized for its diuretic,spleen-tonifying,and sedative properties,modern pharmacological studies confirm that Poria exhibits antioxidant,anti-inflammatory,antibacterial,and antitumor activities.Pachymic acid(PA;a triterpenoid with the chemical structure 3β-acetyloxy-16α-hydroxy-lanosta-8,24(31)-dien-21-oic acid),isolated from Poria,is a principal bioactive constituent.Emerging evidence indicates PA exerts antitumor effects through multiple mechanisms,though these remain incompletely characterized.Neuroblastoma(NB),a highly malignant pediatric extracranial solid tumor accounting for 15%of childhood cancer deaths,urgently requires safer therapeutics due to the limitations of current treatments.Although PA shows multi-mechanistic antitumor potential,its efficacy against NB remains uncharacterized.This study systematically investigated the potential molecular targets and mechanisms underlying the anti-NB effects of PA by integrating network pharmacology-based target prediction with experimental validation of multi-target interactions through molecular docking,dynamic simulations,and in vitro assays,aimed to establish a novel perspective on PA’s antitumor activity and explore its potential clinical implications for NB treatment by integrating computational predictions with biological assays.Methods This study employed network pharmacology to identify potential targets of PA in NB,followed by validation using molecular docking,molecular dynamics(MD)simulations,MM/PBSA free energy analysis,RT-qPCR and Western blot experiments.Network pharmacology analysis included target screening via TCMSP,GeneCards,DisGeNET,SwissTargetPrediction,SuperPred,and PharmMapper.Subsequently,potential targets were predicted by intersecting the results from these databases via Venn analysis.Following target prediction,topological analysis was performed to identify key targets using Cytoscape software.Molecular docking was conducted using AutoDock Vina,with the binding pocket defined based on crystal structures.MD simulations were performed for 100 ns using GROMACS,and RMSD,RMSF,SASA,and hydrogen bonding dynamics were analyzed.MM/PBSA calculations were carried out to estimate the binding free energy of each protein-ligand complex.In vitro validation included RT-qPCR and Western blot,with GAPDH used as an internal control.Results The CCK-8 assay demonstrated a concentration-dependent inhibitory effect of PA on NB cell viability.GO analysis suggested that the anti-NB activity of PA might involve cellular response to chemical stress,vesicle lumen,and protein tyrosine kinase activity.KEGG pathway enrichment analysis suggested that the anti-NB activity of PA might involve the PI3K/AKT,MAPK,and Ras signaling pathways.Molecular docking and MD simulations revealed stable binding interactions between PA and the core target proteins AKT1,EGFR,SRC,and HSP90AA1.RT-qPCR and Western blot analyses further confirmed that PA treatment significantly decreased the mRNA and protein expression of AKT1,EGFR,and SRC while increasing the HSP90AA1 mRNA and protein levels.Conclusion It was suggested that PA may exert its anti-NB effects by inhibiting AKT1,EGFR,and SRC expression,potentially modulating the PI3K/AKT signaling pathway.These findings provide crucial evidence supporting PA’s development as a therapeutic candidate for NB.
基金supported by the National Nature Science Foundation of China(71771201,72531009,71973001)the USTC Research Funds of the Double First-Class Initiative(FSSF-A-240202).
文摘Social interaction with peer pressure is widely studied in social network analysis.Game theory can be utilized to model dynamic social interaction,and one class of game network models assumes that people’s decision payoff functions hinge on individual covariates and the choices of their friends.However,peer pressure would be misidentified and induce a non-negligible bias when incomplete covariates are involved in the game model.For this reason,we develop a generalized constant peer effects model based on homogeneity structure in dynamic social networks.The new model can effectively avoid bias through homogeneity pursuit and can be applied to a wider range of scenarios.To estimate peer pressure in the model,we first present two algorithms based on the initialize expand merge method and the polynomial-time twostage method to estimate homogeneity parameters.Then we apply the nested pseudo-likelihood method and obtain consistent estimators of peer pressure.Simulation evaluations show that our proposed methodology can achieve desirable and effective results in terms of the community misclassification rate and parameter estimation error.We also illustrate the advantages of our model in the empirical analysis when compared with a benchmark model.
基金Supported by the National Natural Science Foundation of China(11971458,11471310)。
文摘In this paper,we propose a neural network approach to learn the parameters of a class of stochastic Lotka-Volterra systems.Approximations of the mean and covariance matrix of the observational variables are obtained from the Euler-Maruyama discretization of the underlying stochastic differential equations(SDEs),based on which the loss function is built.The stochastic gradient descent method is applied in the neural network training.Numerical experiments demonstrate the effectiveness of our method.
文摘In the field of image processing,the analysis of Synthetic Aperture Radar(SAR)images is crucial due to its broad range of applications.However,SAR images are often affected by coherent speckle noise,which significantly degrades image quality.Traditional denoising methods,typically based on filter techniques,often face challenges related to inefficiency and limited adaptability.To address these limitations,this study proposes a novel SAR image denoising algorithm based on an enhanced residual network architecture,with the objective of enhancing the utility of SAR imagery in complex electromagnetic environments.The proposed algorithm integrates residual network modules,which directly process the noisy input images to generate denoised outputs.This approach not only reduces computational complexity but also mitigates the difficulties associated with model training.By combining the Transformer module with the residual block,the algorithm enhances the network's ability to extract global features,offering superior feature extraction capabilities compared to CNN-based residual modules.Additionally,the algorithm employs the adaptive activation function Meta-ACON,which dynamically adjusts the activation patterns of neurons,thereby improving the network's feature extraction efficiency.The effectiveness of the proposed denoising method is empirically validated using real SAR images from the RSOD dataset.The proposed algorithm exhibits remarkable performance in terms of EPI,SSIM,and ENL,while achieving a substantial enhancement in PSNR when compared to traditional and deep learning-based algorithms.The PSNR performance is enhanced by over twofold.Moreover,the evaluation of the MSTAR SAR dataset substantiates the algorithm's robustness and applicability in SAR denoising tasks,with a PSNR of 25.2021 being attained.These findings underscore the efficacy of the proposed algorithm in mitigating speckle noise while preserving critical features in SAR imagery,thereby enhancing its quality and usability in practical scenarios.
基金National Key Research and Development Program(2021YFB2900604)。
文摘Low Earth orbit(LEO)satellite networks exhibit distinct characteristics,e.g.,limited resources of individual satellite nodes and dynamic network topology,which have brought many challenges for routing algorithms.To satisfy quality of service(QoS)requirements of various users,it is critical to research efficient routing strategies to fully utilize satellite resources.This paper proposes a multi-QoS information optimized routing algorithm based on reinforcement learning for LEO satellite networks,which guarantees high level assurance demand services to be prioritized under limited satellite resources while considering the load balancing performance of the satellite networks for low level assurance demand services to ensure the full and effective utilization of satellite resources.An auxiliary path search algorithm is proposed to accelerate the convergence of satellite routing algorithm.Simulation results show that the generated routing strategy can timely process and fully meet the QoS demands of high assurance services while effectively improving the load balancing performance of the link.
