Network Reconstruction Methods Research Articles

Research on deep unfolding network reconstruction method based on scalable sampling of transient signals.

In order to solve the problems of long reconstruction time and low reconstruction accuracy of compressed sensing reconstruction algorithm in the measurement of transient signals, a deep unfolding network reconstruction method based on scalable sampling is proposed to achieve fast and high-quality reconstruction of transient signal under low number of measurements. Firstly, the measurement process of compressed sensing is embedded into the neural network to realize automatic design and optimization of the observation matrix, which can reduce the number of measurements. Secondly, scalable sampling is introduced into the measurement process of compressed sensing, which can realize the training of data with different sampling ratios in the same model. Finally, a deep unfolding network model is designed to reconstruct the transient signal, which not only realizes the interpretability of the reconstructed network, but also achieves fast and high-quality reconstruction of the transient signal under the low number of measurements. Experimental results show that compared with the traditional compressed sensing reconstruction algorithms, the proposed method can obtain high-quality reconstruction accuracy with lower measurement times, and the reconstruction time is greatly reduced. The algorithm in this paper also obtains good reconstruction results under different sampling ratios, which shows that the method in this paper has good adaptability and effectiveness.

Mapping Health Pathways: A Network Analysis for Improved Illness Prediction

ABSTRACTComplex networks and network reconstruction have now become powerful tools for exploring relationships and interactions within systems across various fields to identify patterns and increase predictive accuracy. It provides a structured framework to investigate the relations or connections among individual components or entities within a system. Recently, the importance of the healthcare prediction model in life‐saving efforts has increased. A model is proposed for health prediction that uses network reconstruction methods to build a network from available information, representing features as nodes and the relationships between them as edges. Subsequently, a method is introduced to calculate the value of the decision parameter () for predicting an individual's health status. The proposed model shows substantial improvement over the current state of the prediction model. The first aim of the proposed model is to classify an individual into their appropriate class properly. Another contribution of the proposed model is to measure the factors that are responsible for classifying an individual into a class that shows its significance and impact over the existing state‐of‐the‐art. It provides a new dimension to the prediction mode, emphasizing the importance of identifying critical features and their interdependencies for personalized health diagnostics.

Research on the Recognition and Characterization of Fractures in Coal Rock Based on CT Scanning

Abstract Coal rock fractures play a vital role as key channels for gas migration and storage in the exploration and development of coalbed methane. The characteristics of these fractures, such as porosity distribution, specific surface area, and connectivity, directly impact the adsorption, diffusion, and permeability behaviors of coal reservoirs. Therefore, a detailed analysis of coal rock fracture characteristics is vital in assessing the extractability of coalbed methane and during its exploration and development process. Currently, techniques like imaging logging, seismic detection, and micro-CT are employed for characterizing and studying fractures at various scales. Among these, high-resolution non-destructive CT scanning technology, by constructing three-dimensional data models of core samples, enables a quantitative analysis of fracture distribution features and size parameters. Traditional digital image processing methods, such as threshold segmentation and binary segmentation, though advantageous in computational speed, have limitations in the accuracy of fracture identification. With the advancement of artificial intelligence technology, deep learning-based image processing techniques have increasingly become significant tools for fracture detection due to their powerful feature extraction capabilities. This study conducted micrometer-level CT scanning experiments on coal rock samples, using CT image processing technology to construct a digitalized coal rock core model. Deep learning methods were applied for precise fracture identification, establishing an appropriate deep learning architecture and optimized parameters for coal rock fracture recognition. Furthermore, this paper provides a detailed quantitative characterization of the precisely identified fractures, offering robust technical support for the exploration and development of coalbed methane. The high-resolution non-destructive CT scanning technology and three-dimensional fracture network reconstruction method used in this study have been proven to be effective tools for investigating the micro-features of rocks, enabling the three-dimensional visualization and detailed characterization of rock microstructures.

Inference of human contact networks based on epidemic data

A key challenge in epidemic modeling is the lack of adequate data on population interactions (such as traffic flow or mobility patterns) which result in the spread of infectious diseases. Knowledge of social contact patterns is crucial for public health professionals to devise effective non-pharmaceutical interventions to control epidemics. This paper focuses on inferring social contact rates from reported infection counts during the spread of an infectious disease, addressing the increased difficulty that arises when dealing with “sparse” contact networks where only a small subset of the edges have non-zero weights. Specifically, a new geographically constrained lasso approach for network reconstruction for non-homogeneous mixing Susceptible-Infected-Removed (SIR) disease spread models is presented. The new network reconstruction method can explicitly account for the spatial proximity of network nodes in estimating the disease transmission rates and predicting the future evolution of the epidemic dynamics. Extensive numerical experiments are presented to show the proposed method outperforms existing approaches in terms of accuracy of contact identification under various graph topologies. A case study based on real data from the COVID-19 pandemic is presented to demonstrate the application of the approach for inferring contact structures and a counterfactual scenario analysis to assess effectiveness of containment strategies.

GBTM: Community detection and network reconstruction for noisy and time-evolving data

Community detection and network reconstruction are two major concerns in network analysis. However, these two tasks are extremely challenging since most of the existing methods are not suitable for noisy and red time-evolving data, which are common in real world situations. To cope with this, we propose a novel method called the group-based binary time-evolving mixture (GBTM) model to detect communities and recover network structures jointly. This is the first to study address the challenges of community detection and network reconstruction in scenarios where data are dynamic and cannot be directly observed. In this work, the hidden Markov method is employed to capture the temporal evolution of node connections. In addition, we develop the grouped Baum-Welch algorithm for parameter estimation using a forward-backward procedure. Our GBTM model shows that conducting community detection and network reconstruction simultaneously can yield synergistic benefits. Furthermore, we introduce an innovative Bayesian information criterion (BIC) for determining the number of communities. The results of various simulations under different settings and two real-world networks show that the proposed GBTM model outperforms the existing community detection or network reconstruction methods and has great potential for solving time-evolving and noisy network problems.

A guided network estimation approach using multi-omic information

IntoductionIn systems biology, an organism is viewed as a system of interconnected molecular entities. To understand the functioning of organisms it is essential to integrate information about the variations in the concentrations of those molecular entities. This information can be structured as a set of networks with interconnections and with some hierarchical relations between them. Few methods exist for the reconstruction of integrative networks.ObjectiveIn this work, we propose an integrative network reconstruction method in which the network organization for a particular type of omics data is guided by the network structure of a related type of omics data upstream in the omic cascade. The structure of these guiding data can be either already known or be estimated from the guiding data themselves.MethodsThe method consists of three steps. First a network structure for the guiding data should be provided. Next, responses in the target set are regressed on the full set of predictors in the guiding data with a Lasso penalty to reduce the number of predictors and an L2 penalty on the differences between coefficients for predictors that share edges in the network for the guiding data. Finally, a network is reconstructed on the fitted target responses as functions of the predictors in the guiding data. This way we condition the target network on the network of the guiding data.ConclusionsWe illustrate our approach on two examples in Arabidopsis. The method detects groups of metabolites that have a similar genetic or transcriptomic basis.

Exploring the Molecular Terrain: A Survey of Analytical Methods for Biological Network Analysis

Biological systems, characterized by their complex interplay of symmetry and asymmetry, operate through intricate networks of interacting molecules, weaving the elaborate tapestry of life. The exploration of these networks, aptly termed the “molecular terrain”, is pivotal for unlocking the mysteries of biological processes and spearheading the development of innovative therapeutic strategies. This review embarks on a comprehensive survey of the analytical methods employed in biological network analysis, focusing on elucidating the roles of symmetry and asymmetry within these networks. By highlighting their strengths, limitations, and potential applications, we delve into methods for network reconstruction, topological analysis with an emphasis on symmetry detection, and the examination of network dynamics, which together reveal the nuanced balance between stable, symmetrical configurations and the dynamic, asymmetrical shifts that underpin biological functionality. This review equips researchers with a multifaceted toolbox designed to navigate and decipher biological networks’ intricate, balanced landscape, thereby advancing our understanding and manipulation of complex biological systems. Through this detailed exploration, we aim to foster significant advancements in biological network analysis, paving the way for novel therapeutic interventions and a deeper comprehension of the molecular underpinnings of life.

Gaussian mixture model based reconstruction of undirected networks

The reconstruction of network structure from data represents a significant scientific challenge in the field of complex networks, which has attracted considerable attention from the research community. The most of existing network reconstruction methods transform the problem into a series of linear equation systems, to solve the equations. Subsequently, truncation methods are used to determine the local structure of each node by truncating the solution of each equation system. However, truncation methods frequently exhibit inadequate accuracy, and lack methods of evaluating the truncatability of solutions to each system of equations, that is to say, the reconstructability of nodes. In order to address these issues, in this work an undirected network reconstruction method is proposed based on a Gaussian mixture model. In this method, a Gaussian mixture model is first used to cluster the solution results obtainedby solving a series of linear equations, and then the probabilities of the clustering results are utilized to depict the likelihood of connections between nodes. Subsequently, an index of reconstructibility is defined based on information entropy, thus the probability of connections between each node and other nodes can be used to measure the reconstructibility of each node. The proposed method is ultimately applied to undirected networks. Nodes identified with high reconstructibility are used as a training set to guide the structural inference of nodes with lower reconstrucibility, thus enhancing the reconstruction of the undirected network. The symmetrical properties of the undirected network are then employed to infer the connection probabilities of the remaining nodes with other nodes. The experiments on both synthetic and real data are conducted and a variety of methods are used for constructing linear equations and diverse dynamical models. Compared with the results from a previous truncated reconstruction method, the reconstruction outcomes are evaluated. The experimental results show that the method proposed in this work outperforms existing truncation reconstruction methods in terms of reconstruction performance, thus confirming the universality and effectiveness of the proposed method.

Physics constrained unsupervised deep learning for rapid, high resolution scanning coherent diffraction reconstruction

By circumventing the resolution limitations of optics, coherent diffractive imaging (CDI) and ptychography are making their way into scientific fields ranging from X-ray imaging to astronomy. Yet, the need for time consuming iterative phase recovery hampers real-time imaging. While supervised deep learning strategies have increased reconstruction speed, they sacrifice image quality. Furthermore, these methods’ demand for extensive labeled training data is experimentally burdensome. Here, we propose an unsupervised physics-informed neural network reconstruction method, PtychoPINN, that retains the factor of 100-to-1000 speedup of deep learning-based reconstruction while improving reconstruction quality by combining the diffraction forward map with real-space constraints from overlapping measurements. In particular, PtychoPINN gains a factor of 4 in linear resolution and an 8 dB improvement in PSNR while also accruing improvements in generalizability and robustness. This blend of performance and computational efficiency offers exciting prospects for high-resolution real-time imaging in high-throughput environments such as X-ray free electron lasers (XFELs) and diffraction-limited light sources.

OR20-03 Leptin-mediated Regulation Of Gene Regulatory Networks In Gonadotropes

Abstract Disclosure: J. Banik: None. A. Lagasse: None. A. Haney: None. S. Byrum: None. U. Boehm: None. G.V. Childs: None. M.C. MacNicol: None. A.M. MacNicol: None. A.K. Odle: None. The female reproductive system is energetically expensive and requires tight control of the hypothalamic-pituitary-gonadal (HPG) axis to promote gonadal priming and ovulation. Thus, a crosstalk between the metabolic and reproductive systems exists to regulate the rate-limiting steps of reproduction. Leptin is an adipokine hormone that is produced and secreted proportionally from white fat cells, and leptin receptors (Lepr) are present in the hypothalamus, pituitary, and gonads. Leptin binding to hypothalamic LepR functions as a sensor of nutritional status and, thus, orchestrates energy expenditure during reproductive events. We hypothesized that leptin is crucial to influence fertility in females via differential gene regulation during the estrus cycle in pituitary gonadotropes. To test this hypothesis, we utilized Cre-LoxP technology to selectively ablate leptin receptors within female gonadotropes (CreGnRHR+/- Leprfl/fl TdTomato+/+, referred to as “Gon-Lepr-null”) and performed single-cell RNA-sequencing (scRNA-seq) to identify leptin-dependent targets within the gonadotrope transcriptome. We observed a significant decrease in cell percentage within the gonadotrope cluster, and differential analysis reported a significant decrease in Fshb expression in the Lepr-null mutant gonadotropes. Moreover, the lack of leptin signaling in gonadotropes also caused a shift in the lactotrope-specific transcript signature in the Gon-Lepr-null mutants. We employed SCENIC, a computational method for simultaneous gene regulatory network reconstruction and cell-state identification from single-cell RNA-seq data, to delineate altered upstream regulators contributing to the reduction of Fshb expression in gonadotropes and the shift in the lactotrope transcriptome signature in Gon-Lepr-null mutants versus control animals. Regulons are co-regulated gene sets that share binding sites for a common transcription factor driver, and we identified 381 and 399 regulons specific to controls and mutants, respectively. Examination of the gonadotrope regulons revealed Gata2, Nhlh2, NeuroD1, and Sox11 to be drivers of Fshb expression in control animals only. Differential expression analysis showed significant downregulation of Gata2 and Sox11 in Lepr-null mutant gonadotropes, suggesting Fshb is a leptin-dependent target of Gata2 and Sox11. Moreover, Gata2, Nhlh2, NeuroD1, and Sox11 target 117 genes in controls and 263 genes in mutants, implying a more restricted ability of Gata2, Nhlh2, NeuroD1, and/or Sox11 to target gene expression in the presence of leptin signaling in controls, as an abrogation of leptin signaling leads to an expansion of the regulatory network in Lepr-null mutant gonadotropes. Overall, our analysis reveals how leptin signaling modulates the gonadotrope transcriptome to optimize pituitary reproductive functions at the molecular level. Presentation: Saturday, June 17, 2023

Reconstructing networks via discrete state dynamical data: A mini-review

The inference of network structure from dynamic data is one of the most challenging scientific problems in network science. To address this issue, researchers have proposed various approaches regarding different types of dynamical data. Since many real evolution processes or social phenomena can be described by discrete state dynamical systems, such as the spreading of epidemic, the evolution of opinions, and the cooperation behaviors, network reconstruction methods driven by discrete state dynamical data were also widely studied. In this letter, we provide a mini-review of recent progresses for reconstructing networks based on discrete state dynamical data. These studies encompass network reconstruction problems where the dynamical processes are known, as well as those where the dynamics are unknown, and extend to the reconstruction of higher-order networks. Finally, we discuss the remaining challenges in this field.

Assessing and mitigating fire sales risk under partial information

We consider the problem of assessing and mitigating fire sales risk for banks under partial information. Using data from the European Banking Authority's stress tests, we consider the matrix of asset holdings of different banks. We first analyse fire sales risk under both full and partial information using different matrix reconstruction methods. We then investigate how well some policy interventions aimed at mitigating fire sales risk perform if they are applied based on only partial information. We find that even under partial information, using suitable network reconstruction methods to decide on policy interventions can significantly mitigate risk from fire sales. Furthermore, we show that some interventions based on reconstructed networks significantly outperform ad hoc methods that decide on interventions only based on the size of an institution and do not account for overlapping portfolios.

Research on Road Network Partitioning Considering the Coupling of Network Connectivity and Traffic Attributes

The urban road network is a large and complex system characterized by significant heterogeneity arising from different spatial structures and traffic demands. To facilitate effective management and control, it is necessary to partition the road network into homogeneous sub-areas. In this regard, we aim to propose a hybrid method for partitioning sub-areas with intra-area homogeneity, inter-area heterogeneity, and similar sizes, called CSDRA. It is specifically designed for bidirectional road networks with segment weights that encompass traffic flow, speed, or roadside facility evaluation. Based on community detection and spectral clustering, this proposed method comprises four main modules: initial partition, partitioning of large sub-areas, reassignment of small sub-areas, and boundary adjustment. In the preliminary partitioning work, we also design a road network reconstruction method which further helps to enhance the intra-area homogeneity and inter-area heterogeneity of partitioning results. Furthermore, to align with the requirement for comparable work units in practical traffic management and control, we control the similarity in the size of sub-areas by enforcing upper and lower bound constraints on the size of the sub-areas. We verify the outperformance of the proposed method by an experiment on the partitioning of an urban road network in Guangzhou, China, where we employ sidewalk barrier-free score data as segment weights. The results demonstrate the effectiveness of both the road network reconstruction method and the CSDRA proposed in this paper, as they significantly improve the partitioning outcomes compared with other methods using different evaluation indicators corresponding to the partitioning objectives. Finally, we investigate the influence of constraint parameters on the evaluation indicator. Our findings indicate that appropriately configuring these constraint parameters can effectively minimize sub-region size variations while having minimal impact on other aspects.

Urban waterlogging control: A novel method to urban drainage pipes reconstruction, systematic and automated

Reconstruction of the drainage system is an important measure for urban waterlogging control. For the drainage system of the community, factors such as upstream pipeline water capacity and downstream bottleneck pipeline are often neglected in the process of network reconstruction. Only the network in the waterlogged area is reformed, which is lack of systematicness, and may lead to the failure of waterlogging control. To solve this problem, the Storm Water Management Model (SWMM 5.1) is used to analyze the rationality of the current pipe network reconstruction methods using three indicators: the maximum overflow depth, the maximum overflow volume, and the conduit surcharge. And a new reconstruction method for drainage pipe networks is developed by R programming. The results show that the current pipe network reconstruction method is still insufficient to achieve the goal of eliminating node overflow. For different rainfall conditions, there are still existing low-risk areas, with maximum overflow depth can reach up to 0.15 m and maximum overflow volume can reach up to 7.12 m3. The novel developed drainage pipe network reconstruction method (optimizing pipe network) can effectively identify the bottleneck pipes, and by repeatedly expanding the bottleneck pipes, the goal of node overflow can be achieved under a given rainfall event.

MCPNet: a parallel maximum capacity-based genome-scale gene network construction framework.

Gene network reconstruction from gene expression profiles is a compute- and data-intensive problem. Numerous methods based on diverse approaches including mutual information, random forests, Bayesian networks, correlation measures, as well as their transforms and filters such as data processing inequality, have been proposed. However, an effective gene network reconstruction method that performs well in all three aspects of computational efficiency, data size scalability, and output quality remains elusive. Simple techniques such as Pearson correlation are fast to compute but ignore indirect interactions, while more robust methods such as Bayesian networks are prohibitively time consuming to apply to tens of thousands of genes. We developed maximum capacity path (MCP) score, a novel maximum-capacity-path-based metric to quantify the relative strengths of direct and indirect gene-gene interactions. We further present MCPNet, an efficient, parallelized gene network reconstruction software based on MCP score, to reverse engineer networks in unsupervised and ensemble manners. Using synthetic and real Saccharomyces cervisiae datasets as well as real Arabidopsis thaliana datasets, we demonstrate that MCPNet produces better quality networks as measured by AUPRC, is significantly faster than all other gene network reconstruction software, and also scales well to tens of thousands of genes and hundreds of CPU cores. Thus, MCPNet represents a new gene network reconstruction tool that simultaneously achieves quality, performance, and scalability requirements. Source code freely available for download at https://doi.org/10.5281/zenodo.6499747 and https://github.com/AluruLab/MCPNet, implemented in C++ and supported on Linux.

Landscape and perturbation of enhancer-driven core transcription regulatory circuits in cancer

Landscape and perturbation of enhancer-driven core transcription regulatory circuits in cancer

Topological properties of reconstructed credit networks and banking systemic risk

Financial bipartite networks provide channels for contagion risks and their topological properties determine financial stability. We enrich the bipartite network reconstruction methods proposed by Ramadiah et al. (2020) and extend them to the Chinese banking system. By comparing the reproducibility of the real credit market and the corresponding systemic risk, the impact of topological properties for different reconstructed bipartite networks on financial stability is analyzed. The empirical evidence shows that network reconstruction methods based on maximum entropy ensembles capture more properties in the real credit network. It also highlights that the different systemic risk level is mainly contributed by the topological properties based on common exposures. These analyses for topological properties provide regulatory insights for systemic risk prevention. It shows that reducing credit similarity across banks while increasing credit diversification in different sectors helps to control systemic risk. The results imply the possibility of increasing financial stability through the macro-regulation of the credit market structure.

Rumor source localization in social networks based on infection potential energy

The location technology of information sources in social networks is a major factor in exploring the means of information propagation. In this context, most existing methods ignore the direction of infected nodes and fail to make full use of the diffusion information, resulting in poor identification of source localization. To address such problems, a novel source location method based on infection potential energy is proposed. Its methodology consists of several steps: Firstly, a network reconstruction method is suggested, based on infection potential energy, to make the constructed information diffusion path more accurate. Afterwards, a network pruning method is identified to reduce the search space of candidate sources. Finally, considering the minor differences in the Gaussian density values of multiple nodes and the tendency of the rumor to spread, a distance centrality method is proposed. Experiments in real networks show that the suggested method achieves a better location performance.

Learning the structure of the mTOR protein signaling pathway from protein phosphorylation data

Statistical learning of the structures of cellular networks, such as protein signaling pathways, is a topical research field in computational systems biology. To get the most information out of experimental data, it is often required to develop a tailored statistical approach rather than applying one of the off-the-shelf network reconstruction methods. The focus of this paper is on learning the structure of the mTOR protein signaling pathway from immunoblotting protein phosphorylation data. Under two experimental conditions eleven phosphorylation sites of eight key proteins of the mTOR pathway were measured at ten non-equidistant time points. For the statistical analysis we propose a new advanced hierarchically coupled non-homogeneous dynamic Bayesian network (NH-DBN) model, and we consider various data imputation methods for dealing with non-equidistant temporal observations. Because of the absence of a true gold standard network, we propose to use predictive probabilities in combination with a leave-one-out cross validation strategy to objectively cross-compare the accuracies of different NH-DBN models and data imputation methods. Finally, we employ the best combination of model and data imputation method for predicting the structure of the mTOR protein signaling pathway.

Collaborative Fault Recovery and Network Reconstruction Method for Cyber-physical-systems Based on Double Layer Optimization

Collaborative Fault Recovery and Network Reconstruction Method for Cyber-physical-systems Based on Double Layer Optimization