Network Structure Research Articles

A Generalized Bayesian Stochastic Block Model for Microbiome Community Detection.

Advances in next-generation sequencing technology have enabled the high-throughput profiling of metagenomes and accelerated microbiome studies. Recently, there has been a rise in quantitative studies that aim to decipher the microbiome co-occurrence network and its underlying community structure based on metagenomic sequence data. Uncovering the complex microbiome community structure is essential to understanding the role of the microbiome in disease progression and susceptibility. Taxonomic abundance data generated from metagenomic sequencing technologies are high-dimensional and compositional, suffering from uneven sampling depth, over-dispersion, and zero-inflation. These characteristics often challenge the reliability of the current methods for microbiome community detection. To study the microbiome co-occurrence network and perform community detection, we propose a generalized Bayesian stochastic block model that is tailored for microbiome data analysis where the data are transformed using the recently developed modified centered-log ratio transformation. Our model also allows us to leverage taxonomic tree information using a Markov random field prior. The model parameters are jointly inferred by using Markov chain Monte Carlo sampling techniques. Our simulation study showed that the proposed approach performs better than competing methods even when taxonomic tree information is non-informative. We applied our approach to a real urinary microbiome dataset from postmenopausal women. To the best of our knowledge, this is the first time the urinary microbiome co-occurrence network structure in postmenopausal women has been studied. In summary, this statistical methodology provides a new tool for facilitating advanced microbiome studies.

Novel machine learning-driven comparative analysis of CSP, STFT, and CSP-STFT fusion for EEG data classification across multiple meditation and non-meditation sessions in BCI pipeline

This study focuses on classifying multiple sessions of loving kindness meditation (LKM) and non-meditation electroencephalography (EEG) data. This novel study focuses on using multiple sessions of EEG data from a single individual to train a machine learning pipeline, and then using a new session data from the same individual for the classification. Here, two meditation techniques, LKM-Self and LKM-Others were compared with non-meditation EEG data for 12 participants. Among many tested, three BCI pipelines we built produced promising results, successfully detecting features in meditation/ non-meditation EEG data. While testing different feature extraction algorithms, a common neural network structure was used as the classification algorithm to compare the performance of the feature extraction algorithms. For two of those pipelines, Common Spatial Patterns (CSP) and Short Time Fourier Transform (STFT) were successfully used as feature extraction algorithms where both these algorithms are significantly new for meditation EEG. As a novel concept, the third BCI pipeline used a feature extraction algorithm that fused the features of CSP and STFT, achieving the highest classification accuracies among all tested pipelines. Analyses were conducted using EEG data of 3, 4 or 5 sessions, totaling 3960 tests on the entire dataset. At the end of the study, when considering all the tests, the overall classification accuracy using SCP alone was 67.1%, and it was 67.8% for STFT alone. The algorithm combining the features of CSP and STFT achieved an overall classification accuracy of 72.9% which is more than 5% higher than the other two pipelines. At the same time, the highest mean classification accuracy for the 12 participants was achieved using the pipeline with the combination of CSP STFT algorithm, reaching 75.5% for LKM-Self/ non-meditation for the case of 5 sessions of data. Additionally, the highest individual classification accuracy of 88.9% was obtained by the participant no. 14. Furthermore, the results showed that the classification accuracies for all three pipelines increased with the number of training sessions increased from 2 to 3 and then to 4. The study was successful in classifying a new session of EEG meditation/ non-meditation data after training machine learning algorithms using a different set of session data, and this achievement will be beneficial in the development of algorithms that support meditation.

Novel Polymer Gel Lubricant Functionalized with a Phosphate Anion for Friction Reduction and Film Thickness Enhancement in Multiple Lubrication Conditions.

In this study, a supramolecular polymer gelator functionalized with a phosphate anion, PMUS-P, has been successfully synthesized through radical polymerization, and its physicochemical, rheological properties and tribological performance were carefully evaluated as a gel lubricant formed through non-covalent self-assembly in 500SN base oil. The results showed that the gel has a dense network structure, providing excellent stability and mechanical strength. Additionally, the PMUS-P gel exhibits good shear-thinning behavior and excellent creep recovery, effectively avoiding the volatility of lubricants. Under a steel-steel contact, the PMUS-P gel showed excellent tribological performance in long-term wear tests and a high-load, high-frequency, or high-temperature condition. For instance, in long wear tests, the 15 wt % PMUS-P gel showed a 44.90% reduction in average coefficient of friction (COF) compared to 500SN base oil, along with an 88.05% decrease in wear. The lubrication mechanism study revealed that the chemical reactive film formed by friction played a key role in reducing friction and wear, preventing the friction pairs from direct contact. In terms of film-forming properties, the PMUS-P gel demonstrates superior lubrication performance in comparison to 500SN base oil, achieving higher film thickness. Given these advantages, the PMUS-P gel has significant potential for prolonging machinery service life and reducing operational energy consumption, promising to become a new high-performance lubricant.

Effect of Aging and Sex on Clot Formation, Structure, and Lysis in Trauma Patients.

The dynamics of blood clotting are influenced by age and sex, potentially impacting treatment outcomes and susceptibility to complications in trauma patients. This study aimed to identify how age and sex impact clot formation and resolution kinetics, and network structure in trauma patients, exploring their potential implications for personalized treatment strategies. Trauma patients meeting the highest level of trauma designation had plasma isolated after emergency department admission prior to any resuscitative fluid administration. Thrombin generation kinetics, rheological clot mechanics, fibrin network properties and optical density were measured in patient plasma. Sixty-three patients were included in this study. The majority (76%) were male, and the age range was 18 to 87 years old, with median age of 38. Older patients had increased mortality at similar ISS as young patients, while exhibiting greater clot stiffness and increased time to thrombin generation. Increasing patient age was associated with increased thrombin lag time, time to peak thrombin and fibrinogen normalized stiffness. Male and female patients had similar mortality rates and Injury Severity Scores but exhibited significant differences in clot stiffness and lysis as they age. This study suggests that aging in trauma patients correlates with prolonged clot formation rate despite the presence of stiffer clots and reduced fibrinolysis in older males. Our findings highlight the importance of age and sex when developing resuscitation protocols and tailoring treatment strategies for acute traumatic coagulopathy in trauma patients. Further research is needed to effectively develop and implement such protocols to optimize clinical outcomes in this patient population.

Is the Rosary Still Relevant? Exploring its Impact on Mental Health and Well-Being: A Multinational Study.

The study of benefits linked to meditation and prayer has provided a considerable amount of empirical evidence revealing their positive influence on health, coping, resilience and flourishing. The practice of praying Rosary has drawn much less attention in academic circles, even if it could probably be enlisted with other forms of meditation and prayer with similar positive effects. The present research based on a broad international sample from three Catholic countries (Italy, Poland and Spain) assesses the extent to which such prayer is linked to subjective well-being, empathy and other expressions of religiosity. The sample size consisted of 361 participants. Sampling was conducted using a chain or network method, initiated by reaching out to individuals involved in Catholic movements and devotional groups. The estimated response rate was approximately 65%. The results point to positive moderate correlations of rosary with those variables, like reducing depression, increasing empathy and lowering religious struggles. The network structure analysis reveals that religiosity and religious struggle have the highest number of associations with other variables, establishing them as central factors. Additionally, the qualitative analysis of open-ended questions highlights the perceived protective effect of this prayer, which serves as a source of inner peace and a coping mechanism during times of distress.

Optimization method for supply chain logistics network structure in intelligent manufacturing system

Optimization method for supply chain logistics network structure in intelligent manufacturing system

Longitudinal Relationships of Phubbing, Depression, and Anxiety in the Middle and High School Students: A Cross-Lagged Panel Network Analysis.

Prior research has documented the associations among phubbing, depression, and anxiety, while the cross-sectional design failed to clarify the temporal directionality of the relationships between these mental disorders and behavioral issues. To bridge this gap, the present study utilizing longitudinal data aimed to articulate the temporal relationships between these mental disorders and behavioral issues. A total of 3296 adolescents from China (54.5% girls; Mage = 15.17) participated in the study. Symptoms of phubbing, depression, and anxiety were assessed 18 months later (May 2023) after the baseline (November, 2021). The cross-sectional network and cross-lagged panel network models were conducted to explore the associations between the network structures of phubbing, depression, and anxiety. The network comparison test (NCT) was then performed to unveil whether the network structures vary based on school grade. In the cross-sectional network, significant differences in the overall structures between middle and high school students were observed. For the longitudinal network, the core symptoms responsible for temporal relationships were mostly between depressive and anxiety symptoms. Phubbing-related symptoms and restlessness (anxiety symptom) were the bridge symptoms of phubbing, depression, and anxiety. Besides, the central bridges associated with phubbing-related symptoms differed significantly across different school stages. Successfully regulating negative emotions can play a pivotal role in tackling the root causes linked to phubbing. Apart from addressing restlessness, future interventions focusing on nomophobia and interpersonal conflict in middle school students, as well as self-isolation in high school students, contributed to mitigating phubbing, depression, and anxiety.

Spatial planning instruments at the regional level in selected Central and Eastern European countries. Their legal and institutional dimensions

ABSTRACT The aim of this article is to identify the common features and differences of spatial planning instruments at the regional level in Central and Eastern Europe. The article compares institutional aspects concerning the regional level of spatial planning in Bulgaria, Estonia, Hungary, Lithuania, Latvia, Poland, Romania, Ukraine and Slovakia. Key regional documents in each country, linked to the spatial planning sphere, were identified. Two types of such documents were classified: development strategies with a spatial dimension and regional spatial plans. In countries where the spatial dimension is included in the strategies, the key issues are the definition of the functional and hierarchical structure of the land, the identification of functional areas and areas of particular importance from a development perspective. Most often, spatial plans distinguish the structure of the settlement network, the layout of the infrastructure network, areas suitable for development and areas requiring special protection.

Bilateral Network with Text Guided Aggregation Architecture for Lung Infection Image Segmentation.

Lung image segmentation is a crucial problem for autonomous understanding of the potential illness. However, existing approaches lead to a considerable decrease in accuracy for lung infection areas with varied shapes and sizes. Recently, researchers aimed to improve segmentation accuracy by combining diagnostic reports based on text prompts and image vision information. However, limited by the network structure, these methods are inefficient and ineffective. To address this issue, this paper proposes a Bilateral Network with Text Guided Aggregation Architecture (BNTGAA) to fully fuse local and global information for text and image vision. This proposed architecture involves (i) a global fusion branch with a Hadamard product to align text and vision feature representation and (ii) a multi-scale cross-fusion branch with positional coding and skip connection, performing text-guided segmentation in different resolutions. (iii) The global fusion and multi-scale cross-fusion branches are combined to feed a mamba module for efficient segmentation. Extensive quantitative and qualitative evaluations demonstrate that the proposed architecture performs better both in accuracy and efficiency. Our architecture outperforms the current best methods on the QaTa-COVID19 dataset, improving mIoU and Dice scores by 3.08\% and 2.35\%, respectively. Meanwhile, our architecture surpasses the computational speed of existing multimodal networks. Finally, the architecture has a quick convergence and generality. It can exceed the performance of the current best methods even if it is trained with only 50\% of the dataset.

An integrated experimental-computational investigation of the mechanical behavior of random nanofiber networks.

An integrated experimental-computational methodology was developed to study the mechanical behavior of random polymer nanofiber networks with controlled network structural parameters. Random nanofiber networks, comprised of continuous polyethylene oxide (PEO) nanofibers with ∼250 nm diameter and controlled mean fiber segment length, were designed with a computer algorithm and printed via near-field electrospinning. The structure of the same networks served as input to a computational model to obtain predictions of the macroscopic mechanical response. This methodology provides consistency in fabricating, testing and simulating nominally identical random fiber networks. Specimens with 500 to 5000 nanofibers were subjected to uniaxial tension and compared to modeling predictions for the network mechanical behavior. The predictions by the computational model, with inputs from the experimental network structure, the measured single PEO nanofiber properties, and the fiber crimp parameter, agreed with the experimental results both quantitatively and with respect to the dependence of the measured quantities on the network parameters. The network stiffness and strength followed a power-law scaling with the network density, with exponents 2.78 ± 0.15 and 1.59 ± 0.04, respectively, while the network stretch at failure gradually decreased with increasing network fiber density. Finally, the experimentally determined network toughness demonstrated a rather weak power-law dependence on the network fiber density (exponent of 1.18 ± 0.12).

SKPNet: snake KAN perceive bridge cracks through semantic segmentation

As the demands for ensuring bridge safety continue to rise, crack detection technology has become more crucial than ever. In this context, deep learning methods have been widely applied in the field of intelligent crack detection for bridges. However, existing methods are often constrained by complex backgrounds and computational limitations, struggling with issues such as weak crack continuity and insufficient detail representation. Inspired by biological mechanisms, a dynamic snake convolution (DSC) with tubular offsets is incorporated to tackle these challenges effectively. Additionally, a channel-wise self-attention (CWSA) mechanism is introduced to efficiently fuse multi-scale features in U-Net, significantly enhancing the ability of the model to capture fine details. In the classification head, the traditional linear layer is replaced with a Kolmogorov-Arnold network (KAN) structure, which strengthens the robustness and generalization capacity of the model. Experimental results demonstrate that the proposed model improves detection accuracy, achieving a mean intersection over union (mIoU) of 0.877, while maintaining almost the same number of parameters, showcasing exceptional performance and practical applicability. Our project is released at https://github.com/ruanyudi/KanSeg-Bi .

Rare Earth Element Adsorption from Water Using Alkali-Activated Waste Fly Ash

As new technologies are developed, the demand for rare earth elements (REEs) has increased, despite limited awareness of their significant impact on people and the environment. In this study, waste fly ash was used as a precursor to synthesize inorganic aluminosilicate polymers by adding an activator to the alumina and silica compounds of the ash. Due to their structure and adsorption potential, their application for the removal of selected REEs (Gd3+, Y3+, and Sc3+) from water has been investigated. A decrease in the intensity of the quartz peak at 2θ of 26.6° in the XRD spectrum and the disappearance of the albite and mullite peaks due to dissolution during alkaline activation in both modified samples were observed. The appearance of a peaks at 2θ of 29.3° and 39.3° corresponding to calcite in the modified sample indicates the presence of wood ash. A shifting of the band in the DRIFT spectrum to 1030 cm−1 on the spectra of modified samples corresponds to the vibrations of Al-O and Si-O bonds and the formation of a polymeric network structure (Si-O-Si or Si-O-Al). According to pHPZC values, thermodynamic and kinetic parameters, and chemical composition, the presumed mechanism of REE adsorption is chemisorption and ion exchange. The highest adsorption efficiencies (up to 95%) for all examined REEs in both single and mixed REE solutions were obtained from an alkali-activated mixture of fly ash and wood ash. The results of this research are significant for expanding knowledge about the removal of REEs from the environment, the reduction of waste ash by their modification, and their potential subsequent use in construction as additives.

Isolation and Characterisation of Acid Soluble Collagens and Pepsin Soluble Collagens from Eel (Anguilla japonica Temminck et Schlegel) Skin and Bone

Eel (Anguilla japonica) is an important and valuable food fish in East Asia and its by-products have been reported to include bioactive and profitable components. This study aimed to extract, characterise, and compare the structure and properties of acid-soluble collagens (ASCs) and pepsin-soluble collagens (PSCs) from the skin and bone of eel (Anguilla japonica), providing insights into their composition, structure, and properties for various applications. The yields of ASC-S (from skin), PSC-S (from skin), ASC-B (from bone), and PSC-B (from bone) were 12.16%, 15.54%, 0.79%, and 1.34% on a dry weight basis, respectively. Glycine, the dominant amino acid, accounted for 16.66% to 22.67% of total amino acids in all samples. SDS-PAGE and FTIR analyses showed the typical triple-helical structure of type I collagen with slight variations in molecular order in extract and intermolecular cross-linking between skin and bone collagens. The denaturation temperature (Tmax1) measured by differential scanning calorimetry (DSC) is 81.39 °C and 74.34 °C, respectively, for ASC-B and ASC-S. Bone collagen has higher thermal resistance than skin collagen. Surface morphology imaged using a scanning electron microscope (SEM) showed that the bone collagen had a denser network structure, whilst the skin collagen was more fibrous and porous. The findings suggest that eel-derived collagens from skin and bone can serve as potential alternatives in the food, cosmetic, and healthcare industries.

Synthesis of Porphyrin Derivatives Bearing Six Carboxylic Acids for the Formation of Three-Dimensional Hydrogen-Bonded Network Structures.

we synthesized a series of porphyrin derivatives with six carboxylic acid groups to explore their potential in forming hydrogen-bonded networks (YSHs). By systematically varying the position and structure of these carboxylic acid groups, we observed distinct types of hydrogen-bonded frameworks, including two- and three-dimensional networks. Using single-crystal X-ray crystallography, we confirmed that these derivatives form YSHs with unique structural properties influenced by carboxylic acid positioning and π-π interactions. The 1Zn derivative forms a robust 3D hydrogen-bonded network stabilized by π-π stacking interactions, while the 2Ni derivative, with no such stacking, exhibits reduced stability and collapses upon solvent removal. Structural variations like the terphenylene and biphenyl linkers in 3Zn and 4Zn lead to flexible frameworks, while the 5Zn dimer forms two distinct structures depending on the solvent environment. Our findings reveal that careful control over carboxylic acid orientation and linker structure enables the design of diverse hydrogen-bonded networks with tunable stability and dimensionality. These insights advance our understanding of supramolecular assembly principles in porphyrin-based materials and offer new pathways for developing high-performance frameworks for applications in catalysis, sensing, and materials science.

A cooperative jamming decision-making method based on multi-agent reinforcement learning

Electromagnetic jamming is a critical countermeasure in defense interception scenarios. This paper addresses the complex electromagnetic game involving multiple active jammers and radar systems by proposing a multi-agent reinforcement learning-based cooperative jamming decision-making method (MA-CJD). The proposed approach achieves high-quality and efficient target allocation, jamming mode selection, and power control. Mathematical models for radar systems and active jamming are developed to represent a multi-jammer and multi-radar electromagnetic confrontation scenario. The cooperative jamming decision-making process is then modeled as a Markov game, where the QMix multi-agent reinforcement learning algorithm is innovatively applied to handle inter-jammer cooperation. To tackle the challenges of a parameterized action space, the MP-DQN network structure is adopted, forming the basis of the MA-CJD algorithm. Simulation experiments validate the effectiveness of the proposed MA-CJD algorithm. Results show that MA-CJD significantly reduces the time defense units are detected while minimizing jamming resource consumption. Compared with existing algorithms, MA-CJD achieves better solutions, demonstrating its superiority in cooperative jamming scenarios.

Perceptual QP optimization for VVC with dual hybrid neural networks

This paper introduces a dual hybrid neural network model combining convolutional neural networks (CNNs) and artificial neural networks (ANNs) to optimize the quantization parameter (QP) for both 64×64 and 32×32 blocks in the versatile video coding (VVC) standard, enhancing video quality and compression efficiency. The model employs CNNs for spatial feature extraction and ANNs for structured data handling, addressing the limitations of current heuristic and just noticeable distortion (JND)-based methods. A dataset of luminance channel image blocks, encoded with various QP values, is generated and preprocessed, and the dual hybrid network structure is designed with convolutional and dense layers. The QP optimization is applied at two levels: the 64×64 model provides a global QP offset, while the 32×32 model refines the QP for further partitioned blocks. Performance evaluations using model error metrics like mean squared error (MSE), root mean squared error (RMSE), mean absolute error (MAE), as well as perceptual metrics like weighted PSNR (WPSNR), MS-SSIM, PSNR-HVS-M, and VMAF, demonstrate the model’s effectiveness. While our approach performs competitively with state-of-the-art algorithms, it significantly outperforms in VMAF, the most advanced and widely adopted perceptual quality metric. Furthermore, the dual-model approach yields better results at lower resolutions, whereas the single-model approach is more effective at higher resolutions. These results highlight the adaptability of the proposed models, offering improvements in both compression efficiency and perceptual quality, making them highly suitable for practical applications in modern video coding.

Alloying Confined Regulation of Nanoparticles in a Hierarchically Directed Porous Carbon for Zinc-Air Batteries.

The rational design of non-noble metal-based electrocatalysts with efficient bifunctional catalytic activity is critical for the widespread application of zinc-air batteries (ZABs). In this study, an FeNi alloy encapsulated three-dimensional honeycomb-like network structure of carbon aerogels (FeNi/CAs) electrocatalyst was constructed using directional freeze-drying technology. The innovative architecture, combined with the synergistic effect between Fe and Ni, endows the FeNi/CAs catalyst with outstanding bifunctional catalytic activity compared with the introduction of a single metal in carbon aerogels. Specifically, the catalyst achieves a high half-wave potential (E1/2) of 0.90 V for the oxygen reduction reaction (ORR) and excellent stability with a negligible shift of E1/2 (9 mV) after 2000 cycles. Moreover, the FeNi/CAs catalyst exhibits a smaller potential difference (ΔE = 0.68 V) between the ORR and oxygen evolution reaction (OER), highlighting its superior bifunctional activity. Furthermore, the rechargeable ZABs with FeNi/CAs catalysts show remarkable power density (226 mW cm-2) and energy density (985 mWh kg-1), as well as over 1200 h of cycling stability. Additionally, the discharge rate performance of the assembled flexible all-solid-state battery based on this catalyst remains stable under different bending angles, suggesting its robustness and potential for use in wearable electronic devices. This work provides a compelling strategy for the construction of advanced electrocatalysts by leveraging hierarchical structural features and metal synergy, paving the way for high-performance and durable ZABs in next-generation energy storage applications.

Application of Reinforcement Learning Methods Combining Graph Neural Networks and Self-Attention Mechanisms in Supply Chain Route Optimization

Optimizing transportation routes to improve delivery efficiency and resource utilization in dynamic supply chain scenarios is a challenging task. Traditional route optimization methods often struggle with complex supply chain network structures and dynamic changes, which require a more efficient and flexible solution. This study proposes a method that integrates Graph Neural Networks (GNNs), self-attention mechanisms, and meta-reinforcement learning (Meta-RL) in order to address route optimization in supply chains. The goal is to develop a path planning method that excels in both static and dynamic environments. First, GNNs model the supply chain network, converting node and edge features into high-dimensional graph representations in order to capture local and global network information. Next, a Transformer-based strategy network captures global dependencies, optimizing path planning. Finally, Meta-RL enables rapid strategy adaptation to dynamic changes (e.g., new demand points or route disruptions) with minimal sample support. Experiments on multiple supply chain datasets show that our method improves path planning quality by about 7%, compared to traditional methods, achieving a path coverage of 92.29%. Ablation studies reveal that the on-time delivery rate improves by nearly 30% over the baseline model. These results demonstrate that the proposed method not only optimizes routes but also significantly enhances the overall efficiency and robustness of supply chain networks. This research provides an efficient route optimization framework applicable to complex supply chain management and other scheduling fields, offering new insights and technical solutions for future research and applications.

Landscape Heterogeneity and Island Identity as Drivers of Mesoscale Structure of Pollination Networks

Indirect interactions enforce ecological and evolutionary dynamics within pollination networks. An effective way to study overall indirect interactions in a network is through motif profiles, which represent the network’s mesoscale structure, as well as species’ structural roles, reflecting their participation in motifs. Here, we surveyed 37 pollination networks across eight Aegean islands, a region with a complex biogeographic history, to examine (a) whether species’ structural roles in pollination networks are determined by species, landscape, or island identity; (b) the impact of landscape heterogeneity and island identity on the mesoscale structure of pollination networks; and (c) the variation explained by landscape drivers and island identity in motif profiles compared to link composition. Using PERMANOVA, we found that all three factors significantly grouped species’ structural roles, indicating the combined influence of niche-based and neutral processes. Interestingly, using two dbRDA models to evaluate the combined effects of landscape context and island identity on motif profiles and network link compositions, we found that the first model explained 57% of the variance, whereas the second model accounted for only 16%. This finding emphasizes the potential of motif profiles in revealing interaction dynamics that might otherwise be overlooked. Furthermore, island identity significantly influenced all three responses, suggesting that regional island features play a key role in shaping local interactions.

Network analysis of interactions of rumination and anxiety on smartphone dependence symptoms

ObjectiveRumination and anxiety have been posited as correlates of smartphone dependence (SPD). However, little is known regarding how the components of both affect SPD symptoms at subtle levels. Therefore, we used the network analysis approach to identify the connections at a micro level to provide possible interventions for reducing SPD symptoms.MethodsUsing symptom-level network analysis, we used the ruminative response scale-10, the generalized anxiety disorder scale-7, and the mobile phone addiction index scale-17 to investigate Chinese preservice teachers (Mage = 21.1, N = 1160). Subsequently, we estimated a graphical lasso correlation network for these teachers, which encompassed rumination components, anxiety components, and SPD symptoms. Specifically, the central and bridge centralities within the network structure were examined for the impacts of rumination and anxiety on SPD symptoms.ResultsThe three intracluster connections of rumination, anxiety, and SPD were tighter than the intercluster, with structural connections in rumination and anxiety networks closer than the triggered SPD symptoms cluster. Importantly, reflection reactions towards “write down what you are thinking and analyze it” (a component of rumination) were identified as a central and bridging node that might be a target for intervention for SPD symptoms.ConclusionWe identify potential edge-bridging rumination and anxiety on SPD and locate highly central components within each cluster via network analysis.