Search

In recent years, Graph Neural Networks (GNNs) have garnered significant attention, with a notable focus on Graph Structure Learning (GSL), a branch dedicated to optimizing graph structures to enhance network training performance. Current GSL methods primarily involve constructing optimized graph representations by analyzing one or more initial graph sources to improve performance in subsequent application tasks. Despite these advancements, achieving high-quality graphs that accurately and robustly reflect node relationships remains challenging. This paper introduces a novel approach, termed BAB-GSL, designed to approximate an ideal graph structure through a systematic process. Specifically, two basic views are extracted from the original graph and utilized as inputs for the model, where the preliminary optimized view is generated through the view fusion module. The Attention mechanism is then applied to the optimized view to improve nodes' connectivity and expressiveness. Subsequently, the trained view is re-structured using a Bayesian optimizer to produce the final graph structure. Extensive experiments were conducted across multiple datasets, both in undisturbed and attacked scenarios, to thoroughly evaluate the proposed method, demonstrating the effectiveness and robustness of the BAB-GSL approach., Competing Interests: Declaration of competing interest No conflict of interest exits in the submission of this manuscript, and manuscript is approved by all authors for publication. I would like to declare on behalf of my co-authors that the work described was original research that has not been published previously, and not under consideration for publication elsewhere, in whole or in part. All the authors listed have approved the manuscript that is enclosed., (Copyright © 2024. Published by Elsevier Ltd.)

Traffic flow prediction is the foundation of intelligent traffic management systems. Current methods prioritize the development of intricate models to capture spatio-temporal correlations, yet they often neglect the exploitation of latent features within traffic flow. Firstly, the correlation among different road nodes exhibits dynamism rather than remaining static. Secondly, traffic data exhibits evident periodicity, yet current research lacks the exploration and utilization of periodic features. Lastly, current models typically rely solely on historical data for modeling, resulting in the limitation of accurately capturing future trend changes in traffic flow. To address these findings, this paper proposes a Periodic Dynamic Graph to Sequence Model (PDG2Seq) for traffic flow prediction. PDG2Seq consists of the Periodic Feature Selection Module (PFSM) and the Periodic Dynamic Graph Convolutional Gated Recurrent Unit (PDCGRU) to further extract the spatio-temporal features of the dynamic real-time traffic. The PFSM extracts learned periodic features using time points as indices, while the PDCGRU leverages the extracted periodic features from the PFSM and dynamic features from traffic flow to generate a Periodic Dynamic Graph for extracting spatio-temporal features. In the decoding phase, PDG2Seq utilizes periodic features corresponding to the prediction target to capture future trend changes, leading to more accurate predictions. Comprehensive experiments conducted on four large-scale datasets substantiate the superiority of PDG2Seq over existing state-of-the-art baselines. Related codes are available at https://github.com/wengwenchao123/PDG2Seq., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 Elsevier Ltd. All rights reserved.)

Embeddings derived from cell graphs hold significant potential for exploring spatial transcriptomics (ST) datasets. Nevertheless, existing methodologies rely on a graph structure defined by spatial proximity, which inadequately represents the diversity inherent in cell-cell interactions (CCIs). This study introduces STAGUE, an innovative framework that concurrently learns a cell graph structure and a low-dimensional embedding from ST data. STAGUE employs graph structure learning to parameterize and refine a cell graph adjacency matrix, enabling the generation of learnable graph views for effective contrastive learning. The derived embeddings and cell graph improve spatial clustering accuracy and facilitate the discovery of novel CCIs. Experimental benchmarks across 86 real and simulated ST datasets show that STAGUE outperforms 15 comparison methods in clustering performance. Additionally, STAGUE delineates the heterogeneity in human breast cancer tissues, revealing the activation of epithelial-to-mesenchymal transition and PI3K/AKT signaling in specific sub-regions. Furthermore, STAGUE identifies CCIs with greater alignment to established biological knowledge than those ascertained by existing graph autoencoder-based methods. STAGUE also reveals the regulatory genes that participate in these CCIs, including those enriched in neuropeptide signaling and receptor tyrosine kinase signaling pathways, thereby providing insights into the underlying biological processes., (© 2024 The Author(s). Advanced Science published by Wiley‐VCH GmbH.)

Efficient management of wastewater treatment plants (WWTPs) necessitates accurate forecasting of influent water quality parameters (WQPs) and flow rate (Q) to reduce energy consumption and mitigate carbon emissions. The time series of WQPs and Q are highly non-linear and influenced by various factors such as temperature (T) and precipitation (Precip). Conventional models often struggle to account for long-term temporal patterns and overlook the complex interactions of parameters within the data, leading to inaccuracies in detecting WQPs and Q. This work introduced the Pre-training enhanced Spatio-Temporal Graph Neural Network (PT-STGNN), a novel methodology for accurately forecasting of influent COD, ammonia nitrogen (NH 3 -N), total phosphorus (TP), total nitrogen (TN), pH and Q in WWTPs. PT-STGNN utilizes influent data of the WWTP, air quality data and meteorological data from the service area as inputs to enhance prediction accuracy. The model employs unsupervised Transformer blocks for pre-training, with efficient masking strategies to effectively capture long-term historical patterns and contextual information, thereby significantly boosting forecasting accuracy. Furthermore, PT-STGNN integrates a unique graph structure learning mechanism to identify dependencies between parameters, further improving the model's forecasting accuracy and interpretability. Compared with the state-of-the-art models, PT-STGNN demonstrated superior predictive performance, particularly for a longer-term prediction (i.e., 12 h), with MAE, RMSE and MAPE at 12-h prediction horizon of 2.737 ± 0.040, 4.209 ± 0.060 and 13.648 ± 0.151 %, respectively, for the algebraic mean of each parameter. From the results of graph structure learning, it is observed that there are strong dependencies between NH 3 -N and TN, TP and Q, as well as Precip, etc. This study innovatively applies STGNN, not only offering a novel approach for predicting influent WQPs and Q in WWTPs, but also advances our understanding of the interrelationships among various parameters, significantly enhancing the model's interpretability., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 Elsevier B.V. All rights reserved.)

The gene regulatory network (GRN) plays a vital role in understanding the structure and dynamics of cellular systems, revealing complex regulatory relationships, and exploring disease mechanisms. Recently, deep learning (DL)-based methods have been proposed to infer GRNs from single-cell transcriptomic data and achieved impressive performance. However, these methods do not fully utilize graph topological information and high-order neighbor information from multiple receptive fields. To overcome those limitations, we propose a novel model based on multiview graph attention network, namely, scMGATGRN, to infer GRNs. scMGATGRN mainly consists of GAT, multiview, and view-level attention mechanism. GAT can extract essential features of the gene regulatory network. The multiview model can simultaneously utilize local feature information and high-order neighbor feature information of nodes in the gene regulatory network. The view-level attention mechanism dynamically adjusts the relative importance of node embedding representations and efficiently aggregates node embedding representations from two views. To verify the effectiveness of scMGATGRN, we compared its performance with 10 methods (five shallow learning algorithms and five state-of-the-art DL-based methods) on seven benchmark single-cell RNA sequencing (scRNA-seq) datasets from five cell lines (two in human and three in mouse) with four different kinds of ground-truth networks. The experimental results not only show that scMGATGRN outperforms competing methods but also demonstrate the potential of this model in inferring GRNs. The code and data of scMGATGRN are made freely available on GitHub (https://github.com/nathanyl/scMGATGRN)., (© The Author(s) 2024. Published by Oxford University Press.)

Scene graphs can enhance the understanding capability of intelligent ships in navigation scenes. However, the complex entity relationships and the presence of significant noise in contextual information within navigation scenes pose challenges for navigation scene graph generation (NSGG). To address these issues, this paper proposes a novel NSGG network named SGK-Net. This network comprises three innovative modules. The Semantic-Guided Multimodal Fusion (SGMF) module utilizes prior information on relationship semantics to fuse multimodal information and construct relationship features, thereby elucidating the relationships between entities and reducing semantic ambiguity caused by complex relationships. The Graph Structure Learning-based Structure Evolution (GSLSE) module, based on graph structure learning, reduces redundancy in relationship features and optimizes the computational complexity in subsequent contextual message passing. The Key Entity Message Passing (KEMP) module takes full advantage of contextual information to refine relationship features, thereby reducing noise interference from non-key nodes. Furthermore, this paper constructs the first Ship Navigation Scene Graph Simulation dataset, named SNSG-Sim, which provides a foundational dataset for the research on ship navigation SGG. Experimental results on the SNSG-sim dataset demonstrate that our method achieves an improvement of 8.31% (R@50) in the PredCls task and 7.94% (R@50) in the SGCls task compared to the baseline method, validating the effectiveness of our method in navigation scene graph generation.

Objective : Decoding auditory attention from brain signals is essential for the development of neuro-steered hearing aids. This study aims to overcome the challenges of extracting discriminative feature representations from electroencephalography (EEG) signals for auditory attention detection (AAD) tasks, particularly focusing on the intrinsic relationships between different EEG channels. Approach : We propose a novel attention-guided graph structure learning network, AGSLnet, which leverages potential relationships between EEG channels to improve AAD performance. Specifically, AGSLnet is designed to dynamically capture latent relationships between channels and construct a graph structure of EEG signals. Main result : We evaluated AGSLnet on two publicly available AAD datasets and demonstrated its superiority and robustness over state-of-the-art models. Visualization of the graph structure trained by AGSLnet supports previous neuroscience findings, enhancing our understanding of the underlying neural mechanisms. Significance : This study presents a novel approach for examining brain functional connections, improving AAD performance in low-latency settings, and supporting the development of neuro-steered hearing aids., (© 2024 IOP Publishing Ltd.)

Accurately predicting Drug-Drug Interaction (DDI) is a critical and challenging aspect of the drug discovery process, particularly in preventing adverse reactions in patients undergoing combination therapy. However, current DDI prediction methods often overlook the interaction information between chemical substructures of drugs, focusing solely on the interaction information between drugs and failing to capture sufficient chemical substructure details. To address this limitation, we introduce a novel DDI prediction method: Multi-layer Adaptive Soft Mask Graph Neural Network (MASMDDI). Specifically, we first design a multi-layer adaptive soft mask graph neural network to extract substructures from molecular graphs. Second, we employ an attention mechanism to mine substructure feature information and update latent features. In this process, to optimize the final feature representation, we decompose drug-drug interactions into pairwise interaction correlations between the core substructures of each drug. Third, we use these features to predict the interaction probabilities of DDI tuples and evaluate the model using real-world datasets. Experimental results demonstrate that the proposed model outperforms state-of-the-art methods in DDI prediction. Furthermore, MASMDDI exhibits excellent performance in predicting DDIs of unknown drugs in two tasks that are more aligned with real-world scenarios. In particular, in the transductive scenario using the DrugBank dataset, the ACC and AUROC and AUPRC scores of MASMDDI are 0.9596, 0.9903, and 0.9894, which are 2% higher than the best performing baseline., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2024 Lin, Hong, Cai, Lu and Lin.)

Health insurance fraud is becoming more common and impacting the fairness and sustainability of the health insurance system. Traditional health insurance fraud detection primarily relies on recognizing established data patterns. However, with the ever-expanding and complex nature of health insurance data, it is difficult for these traditional methods to effectively capture evolving fraudulent activity and tactics and keep pace with the constant improvements and innovations of fraudsters. As a result, there is an urgent need for more accurate and flexible analytics to detect potential fraud. To address this, the Multi-channel Heterogeneous Graph Structured Learning-based health insurance fraud detection method (MHGSL) was proposed. MHGSL constructs a graph of health insurance data from various entities, such as patients, departments, and medicines, and employs graph structure learning to extract topological structure, features, and semantic information to construct multiple graphs that reflect the diversity and complexity of the data. We utilize deep learning methods such as heterogeneous graph neural networks and graph convolutional neural networks to combine multi-channel information transfer and feature fusion to detect anomalies in health insurance data. The results of extensive experiments on real health insurance data demonstrate that MHGSL achieves a high level of accuracy in detecting potential fraud, which is better than existing methods, and is able to quickly and accurately identify patients with fraudulent behaviors to avoid loss of health insurance funds. Experiments have shown that multi-channel heterogeneous graph structure learning in MHGSL can be very helpful for health insurance fraud detection. It provides a promising solution for detecting health insurance fraud and improving the fairness and sustainability of the health insurance system. Subsequent research on fraud detection methods can consider semantic information between patients and different types of entities., Competing Interests: The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (© 2024 The Author(s).)

Motivation: Drug-target interaction prediction is an important area of research to predict whether there is an interaction between a drug molecule and its target protein. It plays a critical role in drug discovery and development by facilitating the identification of potential drug candidates and expediting the overall process. Given the time-consuming, expensive, and high-risk nature of traditional drug discovery methods, the prediction of drug-target interactions has become an indispensable tool. Using machine learning and deep learning to tackle this class of problems has become a mainstream approach, and graph-based models have recently received much attention in this field. However, many current graph-based Drug-Target Interaction (DTI) prediction methods rely on manually defined rules to construct the Drug-Protein Pair (DPP) network during the DPP representation learning process. However, these methods fail to capture the true underlying relationships between drug molecules and target proteins., Results: We propose GSL-DTI, an automatic graph structure learning model used for predicting drug-target interactions (DTIs). Initially, we integrate large-scale heterogeneous networks using a graph convolution network based on meta-paths, effectively learning the representations of drugs and target proteins. Subsequently, we construct drug-protein pairs based on these representations. In contrast to previous studies that construct DPP networks based on manual rules, our method introduces an automatic graph structure learning approach. This approach utilizes a filter gate on the affinity scores of DPPs and relies on the classification loss of downstream tasks to guide the learning of the underlying DPP network structure. Based on the learned DPP network, we transform the prediction of drug-target interactions into a node classification problem. The comprehensive experiments conducted on three public datasets have shown the superiority of GSL-DTI in the tasks of DTI prediction. Additionally, GSL-DTI provides a fresh perspective for advancing research in graph structure learning for DTI prediction., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 Elsevier Inc. All rights reserved.)

Graph Convolutional Networks (GCNs) with naive message passing mechanisms have limited performance due to the isotropic aggregation strategy. To remedy this drawback, some recent works focus on how to design anisotropic aggregation strategies with tricks on feature mapping or structure mining. However, these models still suffer from the low ability of expressiveness and long-range modeling for the needs of high performance in practice. To this end, this paper proposes a tree-guided anisotropic GCN, which applies an anisotropic aggregation strategy with competitive expressiveness and a large receptive field. Specifically, the anisotropic aggregation is decoupled into two stages. The first stage is to establish the path of the message passing on a tree-like hypergraph consisting of substructures. The second one is to aggregate the messages with constrained intensities by employing an effective gating mechanism. In addition, a novel anisotropic readout mechanism is constructed to generate representative and discriminative graph-level features for downstream tasks. Our model outperforms baseline methods and recent works on several synthetic benchmarks and datasets from different real-world tasks. In addition, extensive ablation studies and theoretical analyses indicate the effectiveness of our proposed method., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2023 Elsevier Ltd. All rights reserved.)

To learn the embedding representation of graph structure data corrupted by noise and outliers, existing graph structure learning networks usually follow the two-step paradigm, i.e., constructing a "good" graph structure and achieving the message passing for signals supported on the learned graph. However, the data corrupted by noise may make the learned graph structure unreliable. In this paper, we propose an adaptive graph convolutional clustering network that alternatively adjusts the graph structure and node representation layer-by-layer with back-propagation. Specifically, we design a Graph Structure Learning layer before each Graph Convolutional layer to learn the sparse graph structure from the node representations, where the graph structure is implicitly determined by the solution to the optimal self-expression problem. This is one of the first works that uses an optimization process as a Graph Network layer, which is obviously different from the function operation in traditional deep learning layers. An efficient iterative optimization algorithm is given to solve the optimal self-expression problem in the Graph Structure Learning layer. Experimental results show that the proposed method can effectively defend the negative effects of inaccurate graph structures. The code is available at https://github.com/HeXiax/SSGNN., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2023 Elsevier Ltd. All rights reserved.)

The rapid accumulation of electronic health records (EHRs) and the advancements in data analysis technology have laid the foundation for research and clinical decision-making in the healthcare community. Graph neural networks (GNNs), a deep learning model family for graph embedding representations, have been widely used in the field of smart healthcare. However, traditional GNNs rely on the basic assumption that the graph structure extracted from the complex interactions among the EHRs must be a real topology. Noisy connections or false topology in the graph structure leads to inefficient disease prediction. We devise a new model named PM-GSL to improve diabetes clinical assistant diagnosis based on patient multi-relational graph structure learning. Specifically, we first build a patient multi-relational graph based on patient demographics, diagnostic information, laboratory tests, and complex interactions between medicines in EHRs. Second, to fully consider the heterogeneity of the patient multi-relational graph, we consider the node characteristics and the higher-order semantics of nodes. Thus, three candidate graphs are generated in the PM-GSL model: original subgraph, overall feature graph, and higher-order semantic graph. Finally, we fuse the three candidate graphs into a new heterogeneous graph and jointly optimize the graph structure with GNNs in the disease prediction task. The experimental results indicate that PM-GSL outperforms other state-of-the-art models in diabetes clinical assistant diagnosis tasks.

Spatial-temporal graph modeling has been widely studied in many fields, such as traffic forecasting and energy analysis, where data has time and space properties. Existing methods focus on capturing stable and dynamic spatial correlations by constructing physical and virtual graphs along with graph convolution and temporal modeling. However, existing methods tending to smooth node features may obscure the spatial-temporal patterns among nodes. Worse, the graph structure is not always available in some fields, while the manually constructed stable or dynamic graphs cannot necessarily reflect the true spatial correlations either. This paper proposes a Subgraph-Aware Graph Structure Revision network (SAGSR) to overcome these limitations. Architecturally, a subgraph-aware structure revision graph convolution module (SASR-GCM) is designed, which revises the learned stable graph to obtain a dynamic one to automatically infer the dynamics of spatial correlations. Each of these two graphs is separated into one homophilic subgraph and one heterophilic subgraph by a subgraph-aware graph convolution mechanism, which aggregates similar nodes in the homophilic subgraph with positive weights, while keeping nodes with dissimilar features in the heterophilic subgraph mutually away with negative aggregation weights to avoid pattern obfuscation. By combining a gated multi-scale temporal convolution module (GMS-TCM) for temporal modeling, SAGSR can efficiently capture the spatial-temporal correlations and extract complex spatial-temporal graph features. Extensive experiments, conducted on two specific tasks: traffic flow forecasting and energy consumption forecasting, indicate the effectiveness and superiority of our proposed approach over several competitive baselines., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2022 Elsevier Ltd. All rights reserved.)

Currently, most Graph Structure Learning (GSL) methods, as a means of learning graph structure, improve the robustness of GNN merely from a local view by considering the local information related to each edge and indiscriminately applying the mechanism across edges, which may suffer from the local structure heterogeneity of the graph (i.e., the uneven distribution of inter-class connections over nodes). To overcome the drawbacks, we extract the graph structure as a learnable parameter and jointly learn the structure and common parameters of GNN from the global view. Excitingly, the common parameters contain the global information for nodes features mapping, which is also crucial for structure optimization (i.e., optimizing the structure relies on global mapping information). Mathematically, we apply a generic structure extractor to abstract the graph structure and transform GNNs in the form of learning structure and common parameters. Then, we model the learning process as a novel bi-level optimization, i.e., Generic Structure Extraction with Bi-level Optimization for Graph Structure Learning (GSEBO) , which optimizes GNN parameters in the upper level to obtain the global mapping information and graph structure is optimized in the lower level with the global information learned from the upper level. We instantiate the proposed GSEBO on classical GNNs and compare it with the state-of-the-art GSL methods. Extensive experiments validate the effectiveness of the proposed GSEBO on four real-world datasets.

Clustering methods for multivariate data exploiting the underlying geometry of the graphical structure between variables are presented. As opposed to standard approaches for graph clustering that assume known graph structures, the edge structure of the unknown graph is first estimated using sparse regression based approaches for sparse graph structure learning. Subsequently, graph clustering on the lower dimensional projections of the graph is performed based on Laplacian embeddings using a penalized k-means approach, motivated by Dirichlet process mixture models in Bayesian nonparametrics. In contrast to standard algorithmic approaches for known graphs, the proposed method allows estimation and inference for both graph structure learning and clustering. More importantly, the arguments for Laplacian embeddings as suitable projections for graph clustering are formalized by providing theoretical support for the consistency of the eigenspace of the estimated graph Laplacians. Fast computational algorithms are proposed to scale the method to large number of nodes. Extensive simulations are presented to compare the clustering performance with standard methods. The methods are applied to a novel pan-cancer proteomic data set, and protein networks and clusters are evaluated across multiple different cancer types.