Graph neural network-based scheduling for multi-UAV-enabled communications in D2D networks

In this paper,we jointly design the power control and position dispatch for Multi-Unmanned Aerial Vehicle(UAV)-enabled communication in Device-to-Device(D2D)networks.Our objective is to maximize the total transmission rate of Downlink Users(DUs).Meanwhile,the Quality of Service(QoS)of all D2D users must be satisfied.We comprehensively considered the interference among D2D communications and downlink transmissions.The original problem is strongly non-convex,which requires high computational complexity for traditional optimization methods.And to make matters worse,the results are not necessarily globally optimal.In this paper,we propose a novel Graph Neural Networks(GNN)based approach that can map the considered system into a specific graph structure and achieve the optimal solution in a low complexity manner.Particularly,we first construct a GNN-based model for the proposed network,in which the transmission links and interference links are formulated as vertexes and edges,respectively.Then,by taking the channel state information and the coordinates of ground users as the inputs,as well as the location of UAVs and the transmission power of all transmitters as outputs,we obtain the mapping from inputs to outputs through training the parameters of GNN.Simulation results verified that the way to maximize the total transmission rate of DUs can be extracted effectively via the training on samples.Moreover,it also shows that the performance of proposed GNN-based method is better than that of traditional means.

A hybrid physics-informed data-driven neural network for CO_(2) storage in depleted shale reservoirs

To reduce CO_(2) emissions in response to global climate change,shale reservoirs could be ideal candidates for long-term carbon geo-sequestration involving multi-scale transport processes.However,most current CO_(2) sequestration models do not adequately consider multiple transport mechanisms.Moreover,the evaluation of CO_(2) storage processes usually involves laborious and time-consuming numerical simulations unsuitable for practical prediction and decision-making.In this paper,an integrated model involving gas diffusion,adsorption,dissolution,slip flow,and Darcy flow is proposed to accurately characterize CO_(2) storage in depleted shale reservoirs,supporting the establishment of a training database.On this basis,a hybrid physics-informed data-driven neural network(HPDNN)is developed as a deep learning surrogate for prediction and inversion.By incorporating multiple sources of scientific knowledge,the HPDNN can be configured with limited simulation resources,significantly accelerating the forward and inversion processes.Furthermore,the HPDNN can more intelligently predict injection performance,precisely perform reservoir parameter inversion,and reasonably evaluate the CO_(2) storage capacity under complicated scenarios.The validation and test results demonstrate that the HPDNN can ensure high accuracy and strong robustness across an extensive applicability range when dealing with field data with multiple noise sources.This study has tremendous potential to replace traditional modeling tools for predicting and making decisions about CO_(2) storage projects in depleted shale reservoirs.

Outage Probability Analysis for D2D-Enabled Heterogeneous Cellular Networks with Exclusion Zone:A Stochastic Geometry Approach

Interference management is one of the most important issues in the device-to-device(D2D)-enabled heterogeneous cellular networks(HetCNets)due to the coexistence of massive cellular and D2D devices in which D2D devices reuse the cellular spectrum.To alleviate the interference,an efficient interference management way is to set exclusion zones around the cellular receivers.In this paper,we adopt a stochastic geometry approach to analyze the outage probabilities of cellular and D2D users in the D2D-enabled HetCNets.The main difficulties contain three aspects:1)how to model the location randomness of base stations,cellular and D2D users in practical networks;2)how to capture the randomness and interrelation of cellular and D2D transmissions due to the existence of random exclusion zones;3)how to characterize the different types of interference and their impacts on the outage probabilities of cellular and D2D users.We then run extensive Monte-Carlo simulations which manifest that our theoretical model is very accurate.

Comparative study of anti-inflammatory effects of different processed products through the COX-2/PGE2 signaling pathway: based on network pharmacology and molecular docking

Background:Radix Aconiti Lateralis Preparata(Fu-zi)is a traditional Chinese medicinal herb,which has been widely used in the clinic and has potent anti-inflammatory activities.we aimed to explore the mechanisms of extract containing alkaloids from different Fu-zi Processed Products(FPP)in treating inflammation,especially rheumatoid arthritis(RA).Methods:Firstly,using network pharmacology technology,the ingredients,and targets of Fu-zi were obtained by searching and screening,the targets involving RA were acquired,the intersection targets were constructed a"component-target-pathway"network.A comprehensive investigation was conducted on the anti-rheumatoid arthritis mechanisms of 5 FPPs in lipopolysaccharide(LPS)induced RAW264.7 cells,which serve as a model for RA.The production of NO and inflammatory cytokines were measured by ELISA kit.Quantitative Real-time PCR(qRT-PCR)was utilized to measure the mRNA levels.COX-2/PGE2 signaling pathway-associated proteins were determined by western blot.Results:According to a network pharmacological study,16 chemical components and 43 common targets were found in Fu-zi and 6 key targets including PTGS2 were closely related to the mechanism of Fu-zi in treating RA.The in vitro study revealed that the levels of NO,TNF-α,and IL-1βwere substantially decreased by the 5 FPPs.The 5 FPPs significantly suppressed the expression of proteins COX-2,iNOS,and NF-κB,with particularly notable effects observed for PFZ and XFZ.Conclusion:Altogether,these results demonstrated that the 5 PPS containing alkaloids have a good anti-RA-related inflammatory effect,and the mechanism may be related to COX-2/PGE2 signaling pathway,particularly,Fu-zi prepared utilizing a traditional Chinese technique.

基于抑菌实验和网络药理学探讨D-柠檬烯、2-茨醇对白色念珠菌的抑制作用

目的:采用抑菌实验研究蛇床子-冰片药对成分中的D-柠檬烯及2-茨醇的体外抗白色念珠菌作用,并运用网络药理学预测D-柠檬烯和2-茨醇治疗念珠菌病的核心靶点和通路。方法:以白色念珠菌为研究对象,K-B纸片扩散法分别测定0.5、1.0、1.5 mg的D-柠檬烯、2-茨醇、制霉菌素的药液抑菌圈直径;采用试管双倍稀释法和棋盘法,测定D-柠檬烯、2-茨醇的最低抑菌浓度(MIC)以及两两联用的MIC,计算出联合抑菌分数(FIC)。通过Pubchem、SwissTargetPrediction数据库预测D-柠檬烯、2-茨醇的有效靶点;通过GeneCards、OMIM数据库检索念珠菌病相关的疾病靶点;运用Venny软件获得两种化学成分和念珠菌病的共同靶点;运用Cytoscape 3.9. 0软件构建“成分-靶点-疾病”网络;利用STRING数据库构建蛋白互作PPI网络;利用R软件进行GO功能及KEGG通路富集分析。结果:D-柠檬烯的MIC为5 mg/mL,2-茨醇的MIC为2.5 mg/mL。D-柠檬烯与2-茨醇联用的FIC指数为0.75,呈相加作用。网络药理学筛选得到两种成分相关作用靶点152个,疾病靶点893个,两者交集靶点为24个;网络拓扑分析得到核心靶点为肿瘤坏死因子(TNF)、过氧化物酶体增殖物激活受体γ(PPARG)、雌激素受体(ESR1)等;KEGG分析得到核心通路为C型凝集素受体信号通路(C-type lectin receptor signaling pathway)、Fc epsilon RI信号通路(Fc epsilon RI signaling pathway)、催乳素信号通路(prolactin signaling pathway)等。结论:D-柠檬烯、2-茨醇对白色念珠菌均有抑制作用,且2种组分药物联合使用具有一定的协同作用。网络药理学预测初步提示D-柠檬烯、2-茨醇可能通过作用于TNF、PPARG、ESR1等核心靶点调控C型凝集素受体信号通路(C-type lectin receptor signaling pathway)、Fc epsilon RI信号通路(Fc epsilon RI signaling pathway)等以治疗念珠菌病。

基于m×2正则化交叉验证的神经网络超参数调优方法

超参数调优是神经网络建模的关键问题。针对传统的超参数调优方法存在的问题,该文提出了一种基于m×2正则化交叉验证的超参数调优方法。目的是给出一种适用于复杂模型、大数据集背景下的计算开销较小且稳健的超参数调优方法。该方法的思想是从完整的数据集上选取少部分数据进行调优,避免模型在数据集较大时非常耗时的超参数调优难题;在m×2交叉验证的基础上设置正则化条件均衡训练集与验证集之间的分布差异,从而减少分布不一致带来的性能波动;使用信噪比作为调优的优化目标,从而可以综合考虑模型性能评价指标的均值和方差;并采用正交设计选择相关性较低的超参数组合以提高调优效率。以命名实体任务为例进行实验,在CoNLL 2003数据集上的实验结果显示,提出的调优方法能够选到和网格搜索性能上没有显著差异的超参数组合,且调优时间可显著降低约66%。

非奈利酮与SGLT2抑制剂对2型糖尿病和/或慢性肾脏病患者心血管事件的影响

目的比较非奈利酮与钠-葡萄糖共转运蛋白-2(sodium-glucose cotransporter-2,SGLT2)抑制剂对2型糖尿病和/或慢性肾脏病患者心血管事件的影响。方法检索PubMed、Cochrane Library、Web of Science和Embase数据库关于2型糖尿病和/或慢性肾脏病患者的随机对照试验,时间为建库至2023年7月3日。基于频率模型,使用STATA 17.0软件进行网状荟萃分析(network meta-analysis,NMA)。结果共纳入7项随机对照试验,包括33206例患者。涉及的治疗方式包括非奈利酮和SGLT2抑制剂,其中SGLT2抑制剂包含恩格列净、卡格列净、达格列净和索格列净(双重SGLT抑制剂)。在心血管复合事件方面,根据累计曲线下的概率面积(surface under the cumulative ranking area,SUCRA)排序,索格列净最有效。在心血管死亡方面,根据SUCRA排序,恩格列净最有效。在心力衰竭住院方面,根据SUCRA排序,卡格列净最有效。在全因死亡方面,根据SUCRA排序,达格列净最有效。非奈利酮和SGLT2抑制剂在不良事件、严重不良事件和急性肾损害的安全性方面比较,差异均无统计学意义(均P>0.05)。与采用非奈利酮治疗的患者相比,采用SGLT2抑制剂治疗的患者高钾血症发生率更低(RR=0.41,95%CI 0.32~0.52)。结论与非奈利酮相比,SGLT2抑制剂能更好地降低心血管事件的发生率,可作为2型糖尿病和/或慢性肾脏病患者的基础治疗,帮助预防或减少心血管事件。

基于改进INFO-Bi-LSTM模型的SO_(2)排放质量浓度预测

针对火电机组SO_(2)排放质量浓度的影响因素众多,难以准确预测的问题,提出一种改进向量加权平均(weighted mean of vectors,INFO)算法与双向长短期记忆(bi-directional long short term memory,Bi-LSTM)神经网络相结合的预测模型(改进INFO-Bi-LSTM模型)。采用Circle混沌映射和反向学习产生高质量初始化种群,引入自适应t分布提升INFO算法跳出局部最优解和全局搜索的能力。选取改进INFO-Bi-LSTM模型和多种预测模型对炉内外联合脱硫过程中4种典型工况下的SO_(2)排放质量浓度进行预测,将预测结果进行验证对比。结果表明:改进INFO算法的寻优能力得到提升,并且改进INFO-Bi-LSTM模型精度更高,更加适用于SO_(2)排放质量浓度的预测,可为变工况下的脱硫控制提供控制理论支撑。

Long-lasting,reinforced electrical networking in a high-loading Li_(2)S cathode for high-performance lithium–sulfur batteries

Realizing a lithium sulfide(Li_(2)S)cathode with both high energy density and a long lifespan requires an innovative cathode design that maximizes electrochemical performance and resists electrode deterioration.Herein,a high-loading Li_(2)S-based cathode with micrometric Li_(2)S particles composed of two-dimensional graphene(Gr)and one-dimensional carbon nanotubes(CNTs)in a compact geometry is developed,and the role of CNTs in stable cycling of high-capacity Li–S batteries is emphasized.In a dimensionally combined carbon matrix,CNTs embedded within the Gr sheets create robust and sustainable electron diffusion pathways while suppressing the passivation of the active carbon surface.As a unique point,during the first charging process,the proposed cathode is fully activated through the direct conversion of Li_(2)S into S_(8) without inducing lithium polysulfide formation.The direct conversion of Li_(2)S into S_(8) in the composite cathode is ubiquitously investigated using the combined study of in situ Raman spectroscopy,in situ optical microscopy,and cryogenic transmission electron microscopy.The composite cathode demonstrates unprecedented electrochemical properties even with a high Li_(2)S loading of 10 mg cm^(–2);in particular,the practical and safe Li–S full cell coupled with a graphite anode shows ultra-long-term cycling stability over 800 cycles.

基于SE-Res2Net网络的宫颈癌超声肿瘤特征提取技术

为了有效提高宫颈癌的诊断准确率,提出一种基于SE-Res2Net网络的宫颈癌超声肿瘤特征提取技术。在YOLOv3算法模型的基础上,将SE模块嵌入Res2Net网络中,创建一种能够替换原特征提取网络的SE-Res2Net网络,使模型的特征提取能力得到提升。利用重新构建的下采样模块,保证了下采样操作后信息的完整性。将密集连接网络与残差连接网络相结合,组建Res-DenseNet网络以改进YOLOv3模型的原有残差连接方式。实验结果表明,该方法的性能明显优于传统YOLOv3算法,适于在临床诊断中普及应用。

基于SE-R(2+1)D网络的自然环境下的奶牛行为识别

智能行为识别对于奶牛健康的自动诊断和精准养殖具有重要意义。由于接触式传感器会损害动物福利,对奶牛产生应激反应。因此,本文设计了R(2+1)D模型对奶牛进行行为识别。3D网络作为1种时空卷积网络,可以有效识别奶牛的基本时序行为,但该模型针对奶牛的进食行为与反刍行为不易区分,因此对残差网络中的残差映射部分进行改进,在残差网络中添加注意力机制,将SE模块加入到残差映射部分。首先,利用Kinect相机采集奶牛的行为视频;其次,将采集到的奶牛视频分解成连续帧输入到改进后的模型中,连续帧经过二维空间特征和一维时间特征提取,经过残差网络的注意力模块,忽略一些无关的特征;最后,经过模型的Softmax层进行行为分类。实验证明,和原模型比较,准确率提高了2.36%。本文方法实现了精准的奶牛行为识别,可为智慧畜牧业的发展提供技术支持。

响应面法和人工神经网络对亚临界CO_(2)萃取红花籽油的建模与优化

本文旨在寻找有效建模方法以预测亚临界CO_(2)萃取红花籽油的萃取率,优化其萃取工艺条件。以单因素实验为基础,采用Box-Behnken试验设计,研究了萃取压力、分离温度、萃取时间对红花籽油萃取率的影响,并采用响应面法(RSM)和人工神经网络(ANN)两种方法分别对同一实验进行建模分析,通过RSM数值优化、人工神经网络和遗传算法结合(ANN-GA)两种方法优化其工艺条件。结果表明,RSM与ANN两种模型均能较为精准预测,但通过两种模型的决定系数(R^(2))、平均绝对误差(MAE)、平均绝对百分比误差(MAPE)、均方根误差(RMSE)值比较,得出ANN模型(R^(2)=0.9966)的预测效果较优于RSM模型(R^(2)=0.9950)。ANN-GA确定的最佳萃取条件及萃取率分别为:萃取压力19.04 MPa、分离温度55.50℃、萃取时间134.98 min、萃取率23.52%。综上,RSM和ANN两种方法均可用于亚临界CO_(2)萃取带壳红花籽油的建模与优化,但ANN的预测准确度及拟合能力更为优秀。

A UAV-Assisted V2X Network Architecture with Separated Data Transmission and Network Control

With the explosive increasing number of connecting devices such as smart phones, vehicles,drones, and satellites in the wireless networks, how to manage and control such a huge number of networking nodes has become a great challenge. In this paper, we combine the advantages of centralized networks and distributed networks approaches for vehicular networks with the aid of Unmanned Aerial Vehicle(UAV), and propose a Center-controlled Multihop Wireless(CMW) networking scheme consisting of data transmission plane performed by vehicles and the network control plane implemented by the UAV.Besides, we jointly explore the advantages of Medium Access Control(MAC) protocols in the link layer and routing schemes in the network layer to facilitate the multi-hop data transmission for the ground vehicles.Particularly, the network control plane in the UAV can manage the whole network effectively via fully exploiting the acquired network topology information and traffic requests from each vehicle, and implements various kinds of control based on different traffic demands, which can enhance the networking flexibility and scalability significantly in vehicular networks.Simulation results validate the advantages of the proposed scheme compared with existing methods.

A Transmission Design in Dynamic Heterogeneous V2V Networks Through Multi-Agent Deep Reinforcement Learning

In highly dynamic and heterogeneous vehicular communication networks,it is challenging to efficiently utilize network resources and ensure demanding performance requirements of safetyrelated applications.This paper investigates machinelearning-assisted transmission design in a typical multi-user vehicle-to-vehicle(V2V)communication scenario.The transmission process proceeds sequentially along the discrete time steps,where several source nodes intend to deliver multiple different types of messages to their respective destinations within the same spectrum.Due to rapid movement of vehicles,real-time acquirement of channel knowledge and central coordination of all transmission actions are in general hard to realize.We consider applying multi-agent deep reinforcement learning(MADRL)to handle this issue.By transforming the transmission design problem into a stochastic game,a multi-agent proximal policy optimization(MAPPO)algorithm under a centralized training and decentralized execution framework is proposed such that each source decides its own transmission message type,power level,and data rate,based on local observations of the environment and feedback,to maximize its energy efficiency.Via simulations we show that our method achieves better performance over conventional methods.

Type 2 Diabetes Risk Prediction Using Deep Convolutional Neural Network Based-Bayesian Optimization

Diabetes mellitus is a long-term condition characterized by hyperglycemia.It could lead to plenty of difficulties.According to rising morbidity in recent years,the world’s diabetic patients will exceed 642 million by 2040,implying that one out of every ten persons will be diabetic.There is no doubt that this startling figure requires immediate attention from industry and academia to promote innovation and growth in diabetes risk prediction to save individuals’lives.Due to its rapid development,deep learning(DL)was used to predict numerous diseases.However,DLmethods still suffer from their limited prediction performance due to the hyperparameters selection and parameters optimization.Therefore,the selection of hyper-parameters is critical in improving classification performance.This study presents Convolutional Neural Network(CNN)that has achieved remarkable results in many medical domains where the Bayesian optimization algorithm(BOA)has been employed for hyperparameters selection and parameters optimization.Two issues have been investigated and solved during the experiment to enhance the results.The first is the dataset class imbalance,which is solved using Synthetic Minority Oversampling Technique(SMOTE)technique.The second issue is the model’s poor performance,which has been solved using the Bayesian optimization algorithm.The findings indicate that the Bayesian based-CNN model superbases all the state-of-the-art models in the literature with an accuracy of 89.36%,F1-score of 0.88.6,andMatthews Correlation Coefficient(MCC)of 0.88.6.

Reconstructing early transmission networks of SARS-CoV-2 using a genomic mutation model

The coronavirus disease 2019(COVID-19)pandemic has greatly damaged human society,but the origins and early transmission patterns of the severe acute respiratory syndrome coronavirus 2(SARS-CoV-2)pathogen remain unclear.Here,we reconstructed the transmission networks of SARS-CoV-2 during the first three and six months since its first report based on ancestor-offspring relationships using BANAL-52-referenced mutations.We explored the position(i.e.,root,middle,or tip)of early detected samples in the evolutionary tree of SARS-CoV-2.In total,6799 transmission chains and 1766 transmission networks were reconstructed,with chain lengths ranging from 1-9 nodes.The root node samples of the 1766 transmission networks were from 58 countries or regions and showed no common ancestor,indicating the occurrence of many independent or parallel transmissions of SARS-CoV-2 when first detected(i.e.,all samples were located at the tip position of the evolutionary tree).No root node sample was found in any sample(n=31,all from the Chinese mainland)collected in the first 15 days from 24 December 2019.Results using six-month data or RaTG13-referenced mutation data were similar.The reconstruction method was verified using a simulation approach.Our results suggest that SARS-CoV-2 may have already been spreading independently worldwide before the outbreak of COVID-19 in Wuhan,China.Thus,a comprehensive global survey of human and animal samples is essential to explore the origins of SARS-CoV-2 and its natural reservoirs and hosts.

Network pharmacology and molecular docking identify mechanisms of medicinal plant-derived 1,2,3,4,6-penta-O-galloyl-beta-D-glucose treating gastric cancer

Background:1,2,3,4,6-penta-O-galloyl-beta-D-glucose(PGG)is a natural polyphenolic compound derived from multiple medicinal plants with favorable anticancer activity.Methods:In this study,the mechanisms of PGG against gastric cancer were explored through network pharmacology and molecular docking.First,the targets of PGG were searched in the Herbal Ingredients’Targets(HIT),Similarity Ensemble Approach(SEA),and Super-PRED databases.The potential targets related to gastric cancer were predicted from the Human Gene Database(GeneCards)and DisGeNET databases.The intersecting targets of PGG and gastric cancer were obtained by Venn diagram and then subjected to protein-protein interaction analysis to screen hub targets.Functional and pathway enrichment of hub targets were analyzed through Gene Ontology and Kyoto Encyclopedia of Genes and Genomes pathway databases.The differential expression and survival analysis of hub targets in gastric cancer were performed based on The Cancer Genome Atlas database.Finally,the affinity of PGG with hub targets was visualized by molecular docking.Results:Three hub targets were screened,including mitogen-activated protein kinase 14(MAPK14),BCL2 like 1(BCL2L1),and vascular endothelial growth factor A(VEGFA).MAPK14 had a higher expression,while BCL2L1 and VEGFA had lower expression in gastric cancer than in normal conditions.Enrichment analysis indicated enrichment of these hub targets in MAPK,neurotrophin,programmed death-ligand 1(PD-L1)checkpoint,phosphatidylinositol 3-kinases/protein kinase B(PI3K-Akt),Ras,and hypoxia-inducible factor-1(HIF-1)signaling pathways.Conclusion:Therefore,network pharmacology and molecular docking analyses revealed that PGG exerts a therapeutic efficacy on gastric cancer by multiple targets(MAPK14,BCL2L1,and VEGFA)and pathways(MAPK,PD-L1 checkpoint,PI3K-Akt,Ras,and HIF-1 pathways).

Analysis on D2D Heterogeneous Networks with State-Dependent Priority Traffic

In this work,we consider the performance analysis of state dependent priority traffic and scheduling in device to device(D2D)heterogeneous networks.There are two priority transmission types of data in wireless communication,such as video or telephone,which always meet the requirements of high priority(HP)data transmission first.If there is a large amount of low priority(LP)data,there will be a large amount of LP data that cannot be sent.This situation will cause excessive delay of LP data and packet dropping probability.In order to solve this problem,the data transmission process of high priority queue and low priority queue is studied.Considering the priority jump strategy to the priority queuing model,the queuing process with two priority data is modeled as a two-dimensionalMarkov chain.A state dependent priority jump queuing strategy is proposed,which can improve the discarding performance of low priority data.The quasi birth and death process method(QBD)and fixed point iterationmethod are used to solve the causality,and the steady-state probability distribution is further obtained.Then,performance parameters such as average queue length,average throughput,average delay and packet dropping probability for both high and low priority data can be expressed.The simulation results verify the correctness of the theoretical derivation.Meanwhile,the proposed priority jump queuing strategy can significantly improve the drop performance of low-priority data.

Using Network Pharmacology to Explore the Mechanism of Capsicum in Treating Type 2 Diabetes Mellitus

[Objectives] The paper was to explore the mechanism of capsicum ( Capsicum annuum L.) in treating type 2 diabetes mellitus (T2DM) and search for new targets. [Methods] The active ingredients of capsicum were queried from TCMSP database to obtain the corresponding target proteins. The related targets of T2DM were screened from GeneCards database, and the target intersection of active ingredients of capsicum and diabetes mellitus was obtained via Venny software. The protein-protein interaction (PPI) network of the compounds was constructed using STRING database, and the GO bio-function and KEGG pathway enrichment were further analyzed using Metascape database. [Results] Through TCMSP database query and conditional screening, 14 candidate active molecules, 93 potential targets and 225 related pathways were obtained. [Conclusions] The results of GO and KEGG enrichment analysis show that the main active ingredients of capsicum play a role in the treatment of T2DM by regulating cancer pathways, chemical carcinogenesis—receptor activation, proteoglycans in cancer, and prostate cancer pathways, which will provide an important theoretical basis for subsequent research.

Fang-Xia-Dihuang decoction inhibits breast cancer progression induced by psychological stress via down-regulation of PI3K/AKT and JAK2/STAT3 pathways:An in vivo and a network pharmacology assessment

Background:The development and prognosis of breast cancer are intricately linked to psychological stress.In addition,depression is the most common psychological comorbidity among breast cancer survivors,and reportedly,Fang-Xia-Dihuang decoction(FXDH)can effectively manage depression in such patients.However,its pharmacological and molecular mechanisms remain obscure.Methods:Public databases were used for obtaining active components and related targets.Main active components were further verified by ultra-high-performance liquid chromatography-high-resolution mass spectrometry(UPLC-HRMS).Protein–protein interaction and enrichment analyses were taken to predict potential hub targets and related pathways.Molecule docking was used to understand the interactions between main compounds and hub targets.In addition,an animal model of breast cancer combined with depression was established to evaluate the intervention effect of FXDH and verify the pathways screened by network pharmacology.Results:174 active components of FXDH and 163 intersection targets of FXDH,breast cancer,and depression were identified.Quercetin,methyl ferulate,luteolin,ferulaldehyde,wogonin,and diincarvilone were identified as the principal active components of FXDH.Protein–protein interaction and KEGG enrichment analyses revealed that the phosphoinositide-3-kinase–protein kinase B(PI3K/AKT)and Janus kinase/signal transducer and activator of transcription(JAK2/STAT3)signaling pathways played a crucial role in mediating the efficacy of FXDH for inhibiting breast cancer progression induced by depression.In addition,in vivo experiments revealed that FXDH ameliorated depression-like behavior in mice and inhibited excessive tumor growth in mice with breast cancer and depression.FXDH treatment downregulated the expression of epinephrine,PI3K,AKT,STAT3,and JAK2 compared with the control treatment(p<0.05).Molecular docking verified the relationship between the six primary components of FXDH and the three most important targets,including phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha(PIK3CA),AKT,and STAT3.Conclusion:This study provides a scientific basis to support the clinical application of FXDH for improving depression-like behavior and inhibiting breast cancer progression promoted by chronic stress.The therapeutic effects FXDH may be closely related to the PI3K/AKT and JAK2/STAT3 pathways.This finding helps better understand the regulatory mechanisms underlying the efficacy of FXDH.