Types Of Nodes Research Articles

Optical Shubnikov–de Haas oscillations in two-dimensional electron systems

We report on dynamic Shubnikov–de Haas (SdH) oscillations that are measured in the optical response, subterahertz transmittance of two-dimensional systems, and reveal two distinct types of oscillation nodes: “universal” nodes at integer ratios of radiation and cyclotron frequencies and “tunable” nodes at positions sensitive to all parameters of the structure. The nodes in both real and imaginary parts of the measured complex transmittance are analyzed using a dynamic version of the static Lifshitz-Kosevich formula. These results demonstrate that the node structure of the dynamic SdH oscillations provides an all-optical access to quantization- and interaction-induced renormalization effects, in addition to parameters one can obtain from the static SdH oscillations. Published by the American Physical Society 2024

Abstract PO4-25-01: Selective elimination of CD169+ macrophages in lymph nodes invaded by breast cancers

Abstract Lymph node metastasis is a prognostically significant factor in breast cancer, though the colonization mechanism of cancer cells in this immune cell-rich organ remains elusive. This study aimed to elucidate the impact of breast cancer on the lymph node immune cell landscape. Multiscale transcriptomic analyses were performed on both metastatic and non-metastatic lymph node samples from breast cancer patients with lymph node metastasis. Twenty laser-micro-dissected sections were obtained from 17 lymph nodes across six breast cancer patients at stages II–III, with each patient contributing both types of lymph nodes for direct comparison. Comparing the transcriptomes of paired lymph nodes with and without metastasis from the same patients revealed selective downregulation of CD169+ macrophage-related genes in metastatic lymph nodes. The spatial transcriptome indicated a potential depletion of CD169+ macrophages, initiators of anticancer immunity, from their residence (sinuses) in metastatic lymph nodes, while other principal immune cell types were unaltered. Mass spectrometry imaging revealed that the numbers of CD169+ macrophages were smaller in the metastatic lymph nodes than in the non-metastatic lymph nodes, suggesting that cancer cells uprooted CD169+ macrophages from the lymph nodes. Conversely, the count of B, T, Treg, and CD11c+ cells remained comparable in both lymph node types, albeit an enrichment of Treg cells around metastasized cancer tissues was observed. Additional immunohistochemistry analysis of 315 non-metastatic lymph nodes and 159 metastatic lymph nodes from 58 patients with breast cancer showed that a reduced CD169+ macrophage population was prevalent in various breast cancer subtypes. A subset of metastatic lymph nodes (37 out of 159) displayed a complete absence of detectable CD169+ macrophages. The data also depicted a gradual decline in CD169+ macrophages correlating with the pN classification, while no correlation was identified with pathological tumor size classification (pT) or metastasized cancer volume. The data suggest precedence of CD169+ macrophage elimination over other reported immune cell abnormalities, such as cell number and metabolic irregularities. CD169+ macrophages are a unique type of resident macrophages in the lymphoid organs that present cancer-derived antigens to CD8+ cells. The antigen-presenting role of CD169+ macrophages to T cells, a pivotal step in adaptive immunity, signifies the catastrophic implications of their suppression. This study underscores CD169+ macrophage suppression as a pronounced pathological phenotype in lymph nodes with breast cancer metastasis, thereby establishing it as a critical future therapeutic target. Citation Format: Yurina Maeshima, Tatsuki R. Kataoka, Alexis Vandenbon, Masahiro Hirata, Yasuhide Takeuchi, Yutaka Suzuki, Yukiko Fukui, Yumiko Ibi, Hironori Haga, Satoshi Morita, Masakazu Toi, Shinpei Kawaoka, Kosuke Kawaguchi. Selective elimination of CD169+ macrophages in lymph nodes invaded by breast cancers [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO4-25-01.

Relation-aware heterogeneous graph neural network for entity alignment

Entity alignment refers to finding equivalent entities from different knowledge graphs. Most of the existing entity alignment methods are studied based on homogeneous graphs. However, the knowledge graph is a heterogeneous graph containing many types of nodes, such as entities, relations, and attributes. Therefore, we propose introducing a heterogeneous graph neural network to model entities and relations simultaneously and propose an iterative fusion method to enhance the interaction between these two semantic nodes. Since not all datasets contain relation information, this paper does not directly introduce a feature representation of relations. The generalizability of the approach is improved by utilizing a relation-aware strategy to obtain information about the relation. Specifically, the information propagation of the head and tail entities in the triplet is utilized to obtain the feature representation of the relation. Experimental results show that the present method performs better on three cross-lingual datasets DBP15K and two large-scale datasets DWY100K.

A Graph-Based Keyword Extraction Method for Academic Literature Knowledge Graph Construction

In this paper, we construct an academic literature knowledge graph based on the relationship between documents to facilitate the storage and research of academic literature data. Keywords are an important type of node in the knowledge graph. To solve the problem that there are no keywords in some documents for several reasons in the process of knowledge graph construction, an improved keyword extraction algorithm called TP-CoGlo-TextRank is proposed by using word frequency, position, word co-occurrence frequency, and a word embedding model. By combining the word frequency and position in the document, the importance of words is distinguished. By introducing the GloVe word-embedding model, which brings the external knowledge of documents into the TextRank algorithm, and combining the internal word co-occurrence frequency in the documents, the word-adjacency relationship is transferred non-uniformly. Finally, the words with the highest scores are combined into phrases if they are adjacent in the original text. The validity of the TP-CoGlo-TextRank algorithm is verified by experiments. On this basis, the Neo4j graph database is used to store and display the academic literature knowledge graph, to provide data support for research tasks such as text clustering, automatic summarization, and question-answering systems.

The impact of patentees assessment based on the heterogeneous patent innovation network

As a vital factor in technological innovation, patentee plays a significant role in the process of scientific and technological innovation, researching patentee has attracted the attention of experts and scholars. Previously, scholars have mainly quantified patent indicators or constructed homogeneous information networks to analyze patentees, but these methods cannot objectively measure the impact of patentees. Therefore, this study proposes a novel approach to assessing patentee impact based on a heterogeneous information network. The proposed method distinguishes the weight of different types of nodes using a weighted mechanism and extracts three types of fine-grained characteristics of network nodes. This approach results in the construction of a heterogeneous patent innovation network and the development of a new patentee impact assessment algorithm called CWAPN. Using Chinese green patents in the field of energy conservation and environmental protection as an example, experimental results show that the CWAPN algorithm can effectively assess the impact of patentees. Thereby identifying patentees who have made outstanding contributions to sustainable development in China.

Noctua2 Supercomputer

Noctua 2 is a supercomputer operated at the Paderborn Center for Parallel Computing (PC2) at Paderborn University in Germany. Noctua 2 was inaugurated in 2022 and is an Atos BullSequana XH2000 system. It consists mainly of three node types: 1) CPU Compute nodes with AMD EPYC processors in different main memory configurations, 2) GPU nodes with NVIDIA A100 GPUs, and 3) FPGA nodes with Xilinx Alveo U280 and Intel Stratix 10 FPGA cards. While CPUs and GPUs are known off-the-shelf components in HPC systems, the operation of a large number of FPGA cards from different vendors and a dedicated FPGA-to-FPGA network are unique characteristics of Noctua 2. This paper describes in detail the overall setup of Noctua 2 and gives insights into the operation of the cluster from a hardware, software and facility perspective.

GEHGAN[formula omitted] CircRNA–disease association prediction via graph embedding and heterogeneous graph attention network

There is growing proof suggested that circRNAs play a crucial function in diverse important biological reactions related to human diseases. Within the area of biochemistry, a massive range of wet experiments have been carried out to find out the connections of circRNA–disease in recent years. Since wet experiments are expensive and laborious, nowadays, calculation-based solutions have increasingly attracted the attention of researchers. However, the performance of these methods is restricted due to the inability to balance the distribution among various types of nodes. To remedy the problem, we present a novel computational method called GEHGAN to forecast the new relationships in this research, leveraging graph embedding and heterogeneous graph attention networks. Firstly, we calculate circRNA sequences similarity, circRNA RBP similarity, disease semantic similarity and corresponding GIP kernel similarity to construct heterogeneous graph. Secondly, a graph embedding method using random walks with jump and stay strategies is applied to obtain the preliminary embeddings of circRNAs and diseases, greatly improving the performance of the model. Thirdly, a multi-head graph attention network is employed to further update the embeddings, followed by the employment of the MLP as a predictor. As a result, the five-fold cross-validation indicates that GEHGAN achieves an outstanding AUC score of 0.9829 and an AUPR value of 0.9815 on the CircR2Diseasev2.0 database, and case studies on osteosarcoma, gastric and colorectal neoplasms further confirm the model’s efficacy at identifying circRNA–disease correlations.

Chaos quantum optimization‐based layered diagnosis framework for faulty sensor node diagnosis and classification in wireless sensor networks

SummaryNode faults are often influenced by factors such as physical component failure, communication module errors, battery exhaustion, and environmental factors. Various researchers have contributed to addressing the problem of detecting, diagnosing, and classifying faults in sensor networks. However, computational complexity and fault occurrence probability increase with the size of the network. In this context, an enhanced optimized layered diagnosis framework (OLDF) is proposed for detecting, diagnosing, and classifying sensor node faults. OLDF utilizes a deep belief network (DBN) with chaos quantum‐behaved particle swarm optimization (CQ‐PSO) for efficient diagnosis and classification of various sensor faults in the network. The proposed OLDF consists of two layers. In the first layer, optimized DBN with CQ‐PSO algorithm is employed to determine the fault type of the sensor nodes. Then, the fault type of sensor nodes that are classified in the first layer is given as input into the second layer of the OLDF algorithm to determine the fault severity in the network. Experimentation has been carried out by using the NS3 simulator for the OLDF framework with performance metrics such as classification accuracy, diagnosis accuracy, false positives, fault alarm rate, and average energy consumption. The validity of the proposed framework has been verified through comparisons with existing works such as MFD, INSA, FDRFC, and HFD. From the comparative analysis carried out, it is apparent that the proposed framework achieves a high classification accuracy of 89.94% and diagnosis accuracy of 93.21%, with less false positive rate, and utilizes minimum energy when compared with its counterparts.

DRTerHGAT: A drug repurposing method based on the ternary heterogeneous graph attention network

Drug repurposing is an effective method to reduce the time and cost of drug development. Computational drug repurposing can quickly screen out the most likely associations from large biological databases to achieve effective drug repurposing. However, building a comprehensive model that integrates drugs, proteins, and diseases for drug repurposing remains challenging. This study proposes a drug repurposing method based on the ternary heterogeneous graph attention network (DRTerHGAT). DRTerHGAT designs a novel protein feature extraction process consisting of a large-scale protein language model and a multi-task autoencoder, so that protein features can be extracted accurately and efficiently from amino acid sequences. The ternary heterogeneous graph of drug-protein-disease comprehensively considering the relationships among the three types of nodes, including three homogeneous and three heterogeneous relationships. Based on the graph and the extracted protein features, the deep features of the drugs and the diseases are extracted by graph convolutional networks (GCN) and heterogeneous graph node attention networks (HGNA). In the experiments, DRTerHGAT is proven superior to existing advanced methods and DRTerHGAT variants. DRTerHGAT's powerful ability for drug repurposing is also demonstrated in Alzheimer's disease.

Railway network delay evolution: A heterogeneous graph neural network approach

Accurate delay evolution prediction plays a pivotal role in train rescheduling decision-making for the railway network. Existing studies on delay prediction predominantly centered around predicting delays for each train in the subsequent stations (i.e., following a train-oriented perspective). Furthermore, train operations in the railway network involve different types of entities (stations, trains, etc.), making the current graph/network models with homogenous nodes (i.e., the same kind of nodes) incapable of effectively capturing the interactions between the entities. This paper develops a network-oriented model to investigate the train delay evolution on railway networks, by predicting the delays of running trains in the network after a given time interval. The proposed model combines the GraphSAGE graph neural network (GNN) and the heterogeneous graph neural network (HetGNN) architecture, thus called SAGE-Het, enabling it to capture interactions between heterogeneous nodes (i.e., different types of nodes) based on different edges (e.g., edges between trains, trains, and stations). Additionally, SAGE-Het allows for flexible inputs in contrast to conventional machine learning techniques, whose inputs must meet the consistent dimension requirement (e.g., in the form of rectangular or grid-like arrays). The performance and robustness of the suggested SAGE-Het model are assessed in experiments on the data from two sub-networks of the China railway network. The experimental results demonstrate that SAGE-Het outperforms the existing delay prediction methods and some advanced HetGNNs used for prediction tasks in other domains; SAGE-Het demonstrates excellent scalability, capable of handling various types of nodes; the predictive performances of SAGE-Het under different prediction time horizons (10/20/30 min ahead) all exhibit better performance over other baselines; the accuracies are over 90 % under the permissible 3-minute errors for the three prediction time horizons. Specifically, the impact of train interactions on delay evolution is investigated based on the flexible input characteristic of the proposed model. The results illustrate that train interactions become subtle with the increase of train headways. This finding directly contributes to decision-making in situations where conflict resolution or train-canceling actions are needed.

Examining the Completeness of Breast Cancer Pathology Reports Registered in the Population-Based Cancer Registration System in Iran during 2016 to 2018.

Ensuring the comprehensive and accurate representation of data within cancer registries holds paramount significance across various facets of public health decision-making. This study delves into the evaluation of data completeness in breast cancer (BC) pathology reports within a population-based cancer registration system in Iran, spanning the period from 2016 to 2018. Employing a retrospective and descriptive analytical approach, we harnessed secondary data extracted from pathology reports encompassing breast cancer diagnoses, which were duly recorded in the Integrated Cancer Information Management System database during 2016-2018. A total of 4000 pathology reports were thoughtfully selected from each of the three years. The spectrum of pathology information encompassed tumor type, site grade, size (T), and involvement of lymph nodes (N). Summary statistics were provided as percentages of categorical variables and mean with standard deviation of continuous variables. A comparison of categorical variables was performed using the Chi-squared test. The participants' mean age was 51.8±12.5 years. Among the 12,000 studied patients, 5744 (47.9%) were ≤ 50 years old, 5233 (43.6%) were aged 50-69 years, and 1023 (8.5%) were >60 years old. The completeness of BC pathology reports varied for different variables. Interestingly, the completeness of these variables increased with older age groups. The proportion of specific tumor types differed significantly among age groups (P = 0.001). Notably, the prevalence of invasive ductal carcinoma was higher in the ≤ 50 years age group compared to the older cohorts. Likewise, notable variations in tumor sizes were observed (P = 0.009), with a higher prevalence of missing tumor size data noted in the age group ≤ 50 years. On the other hand, pathologic T stage also demonstrated age-dependent variations (P = 0.014), indicating a higher prevalence of missing stages in the ≤ 50 years age group. Finally, tumor grade exhibited a statistically significant difference (P < 0.001), with a higher proportion of grade 1 tumors observed in the 50-69 years age group. Tumor grade had the highest completeness rate, while tumor size, pathologic T stage, and pathologic N stage had the lowest. Therefore, a good understanding of completeness of pathology reports, as well as improvement in the registration of stage, integrated system at the national level for BC is warranted.

Hybrid-attention mechanism based heterogeneous graph representation learning

Heterogeneous graph refers to a type of graph data characterized by its diverse node types and relation types, containing rich structures, features and heterogeneous information. How to fully utilize and capture these key information to generate effective node representations poses a great challenge in heterogeneous graph analysis and mining. To better tackle this problem, a heterogeneous graph representation learning model based on hybrid-attention mechanism is proposed, namely Heterogeneous Graph Relation Attention Network (HGRAN). The main contributions of HGRAN are listed as follows. First, a novel framework was proposed for better representing heterogeneous information originating from various relations and comprehensive usage of both structural and feature information instead of employing meta-path based framework. Second, a novel hybrid-attention mechanism which combines relation attention and node attention was proposed within this framework. Third, a novel feature similarity based relation attention is proposed to capture heterogeneous information originating from different relations. Fourth, in order to better implement node attention in heterogeneous graphs, a new transforming method that transforms adjacency matrices of diverse relations into a unified manner is proposed. Finally, extensive experiments on multiple real-world heterogeneous graph datasets are conducted to verify HGRAN, and the results support its superiority in comparison with the state-of-the-art methods.

TemporalHAN: Hierarchical attention-based heterogeneous temporal network embedding

Heterogeneous temporal network embedding aims to learn each node of different types of a heterogeneous temporal network in each snapshot into a low-dimensional vector representation, which can be used for various network analysis tasks such as node classification and relationship prediction. Our work proposes a novel heterogeneous temporal graph neural network embedding framework (TemporalHAN) based on hierarchical attention using a temporal convolutional network (TCN). In particular, we introduce node-level and semantic-level attention into heterogeneous graph neural networks to identify the importance of different levels between nodes. For each snapshot, we first utilise a new random walk algorithm (NRWA) to collect strongly connected heterogeneous neighbours for each node of different types and group them by node types. In addition, the algorithm utilises a damping factor to ensure that the more recent snapshots allocate more random walk steps. We then utilise node-level and semantic-level attention to learn the importance between a node and its random walk neighbour for a specific node type and learn the importance of different node-type for this node, respectively. Finally, we adopt TCN to capture the evolution information between snapshots. Experimental results on relationship prediction and node classification reveal that the TemporalHAN is competitive against diverse state-of-the-art approaches. Our code is available at https://github.com/Legendary-L/THAN.

Topic-Aware Information Coverage Maximization in Social Networks

<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Influence maximization</i> (IM) aims to identify a set of nodes <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$S$</tex-math> </inline-formula> to maximize the expected number of nodes influenced during the information propagation starting from <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$S$</tex-math> </inline-formula> . Some works had extended this problem to be <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">topic-aware</i> , where each node is associated with a topic distribution and tends to be activated with different probabilities by different topics. However, whether it is topic-aware or not, IM problem only focuses on the active nodes and overlooks all the inactive ones. Actually, an inactive node may receive the information from their active in-neighbors and become informed. Therefore, this type of nodes should also be considered when measuring the coverage of information propagation. Inspired by this, we formulate a new problem called <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">topic-aware information coverage maximization</i> (TAICM), which aims to maximize the sum of the expected number of both active and informed nodes in topic-aware social networks. Then we devise a heuristic method to solve it. Experiments on three real-world datasets demonstrate that our method can achieve similar or higher information coverage in much less or at least acceptable time than some commonly used IM algorithms.

Design and Performance Study of a Six-Leg Lattice Tower for Wind Turbines

A new type of spherical node was used to design a laboratory-scale prototype of a six-leg lattice of steel tubes and concrete for application as a wind turbine tower. Repeated load tests were performed on the prototype tower for several weeks to evaluate its load-carrying capacity, deformation, energy consumption, stress distribution based on damage patterns, hysteresis curves, skeleton curves, strength, and stiffness degradation curves. The findings indicated that the prototype tower underwent thread damage to the high-strength bolts of the inclined web and weld damage between the inclined web and sealing plate. Although the stress differences between different measurement points were significant, the stress values were small at most of the measurement points. The maximum equivalent stress value was 294 MPa, which appeared in the middle layer of the BC surface. The P-Δ hysteresis curve had an inverse “S”-shape, and the bearing capacity was high. The maximum energy dissipation appeared in the 1.75 Δy loading stage. The peak load of the specimen can reach 376.2 kN, and the corresponding peak displacement is 37 mm. However, the average ductility coefficient was only 2.33, indicating little plastic deformation. The maximum strain of the tower column foot is 1800 με, and the force of the inclined web member in the middle layer is the largest. The strain of the transverse web bar increased significantly after the tower yielded, which contributed to maintaining the integrity of the structure.

Crossing Limit Cycles in Planar Piecewise Linear Systems Separated by a Nonregular Line with Node–Node Type Critical Points

In this paper, we investigate the existence and number of crossing limit cycles in a class of planar piecewise linear systems with node–node type critical points defined in two zones separated by a nonregular line formed by two rays emanated from the origin [Formula: see text], which are the positive [Formula: see text]- and [Formula: see text]-axes. We focus our attention on the existence of two-point crossing limit cycles, which intersect the switching line at two points. We obtain sufficient conditions under which the system has two two-point crossing limit cycles which intersect only one of the two rays. Moreover, we construct examples to show that this class of systems can have two, three or four two-point crossing limit cycles.

Research of the optimization problem of structure hierarchical communication network with changing its parameters

The article is devoted to the study of the problem of optimizing the hierarchical structure of a multicommodity communication network with discrete flows when changing its important parameters, such as the capacity of network arcs in transport blocks and the size of the transport block for transportation of the discrete small-lot cargo or of the data transmission in a digital communication network. The network has three levels of hierarchy – a backbone, a zonal and an internal and four types of nodes - backbone nodes of the first, second and third types, forming the backbone and zonal levels of the network, and nodes of the fourth type, which subordinate to each backbone node and forming the internal levels of the network. Types of nodes differ from each other in terms of functionality. The main task of the study is to establish how the structure of the backbone network changes (the number and location of backbone nodes of the first, second, and third types), the flow processing and distribution scheme, and the technical and economic indicators of the network's functioning for different values of its parameters. The principles of organization of sorting and distribution of flows in a three-level network and its mathematical model are given. A mathematical model of the optimization problem of the backbone network structure and flow sorting and distribution scheme is formulated. Algorithms for solving the problem are based on the discrete analogue of the local descent method proposed by the authors earlier, when the neighborhoods of the metric space of possible solutions are chosen based on heuristic considerations, taking into account the specifics of the problem being solved. Computer modeling of the problem on a network containing 10 nodes at the specified change in the both parameters is carried out. The modeling was carried out on the example of the transport network of cargo transportation using a computer program that is part of the instrumental software of the Information and Analytical Decision Support System (IA DSS), which is being developed at the Institute of Telecommunications and Global Information Space of the National Academy of Sciences of Ukraine. An experimental study of solution the problem showed that the structure of the network is weakly dependent on the change in the carrying capacity of the arcs and the size of the transport block. The proposed computer technology for solving problem when the network parameters are changed allows you to interactively modeling various options of a network, changing the topology, hierarchical structure, flows, parameters and constraints of the model and from the set of the obtained solutions to choose a best option, taking into account the selected a goal function and the accepted constraints; calculate preliminary technical and economic indicators of the network's functioning, estimate the cost of additional resources and plan the amount of investment required for the modernization and construction of its structural elements, which ultimately makes it possible to increase the efficiency functioning of the network by optimizing use of its resources and reducing the operating costs for the processing and transportation of flows.

Dual-Path Graph Neural Network with Adaptive Auxiliary Module for Link Prediction.

Link prediction, which has important applications in many fields, predicts the possibility of the link between two nodes in a graph. Link prediction based on Graph Neural Network (GNN) obtains node representation and graph structure through GNN, which has attracted a growing amount of attention recently. However, the existing GNN-based link prediction approaches possess some shortcomings. On the one hand, because a graph contains different types of nodes, it leads to a great challenge for aggregating information and learning node representation from its neighbor nodes. On the other hand, the attention mechanism has been an effect instrument for enhancing the link prediction performance. However, the traditional attention mechanism is always monotonic for query nodes, which limits its influence on link prediction. To address these two problems, a Dual-Path Graph Neural Network (DPGNN) for link prediction is proposed in this study. First, we propose a novel Local Random Features Augmentation for Graph Convolution Network as a baseline of one path. Meanwhile, Graph Attention Network version 2 based on dynamic attention mechanism is adopted as a baseline of the other path. And then, we capture more meaningful node representation and more accurate link features by concatenating the information of these two paths. In addition, we propose an adaptive auxiliary module for better balancing the weight of auxiliary tasks, which brings more benefit to link prediction. Finally, extensive experiments verify the effectiveness and superiority of our proposed DPGNN for link prediction.

Learning Association Characteristics by Dynamic Hypergraph and Gated Convolution Enhanced Pairwise Attributes for Prediction of Disease-Related lncRNAs.

As the long non-coding RNAs (lncRNAs) play important roles during the incurrence and development of various human diseases, identifying disease-related lncRNAs can contribute to clarifying the pathogenesis of diseases. Most of the recent lncRNA-disease association prediction methods utilized the multi-source data about the lncRNAs and diseases. A single lncRNA may participate in multiple disease processes, and multiple lncRNAs usually are involved in the same disease process synergistically. However, the previous methods did not completely exploit the biological characteristics to construct the informative prediction models. We construct a prediction model based on adaptive hypergraph and gated convolution for lncRNA-disease association prediction (AGLDA), to embed and encode the biological characteristics about lncRNA-disease associations, the topological features from the entire heterogeneous graph perspective, and the gated enhanced pairwise features. First, the strategy for constructing hyperedges is designed to reflect the biological characteristic that multiple lncRNAs are involved in multiple disease processes. Furthermore, each hyperedge has its own biological perspective, and multiple hyperedges are beneficial for revealing the diverse relationships among multiple lncRNAs and diseases. Second, we encode the biological features of each lncRNA (disease) node using a strategy based on dynamic hypergraph convolutional networks. The strategy may adaptively learn the features of the hyperedges and formulate the dynamically evolved hypergraph topological structure. Third, a group convolutional network is established to integrate the entire heterogeneous topological structure and multiple types of node attributes within an lncRNA-disease-miRNA graph. Finally, a gated convolutional strategy is proposed to enhance the informative features of the lncRNA-disease node pairs. The comparison experiments indicate that AGLDA outperforms seven advanced prediction methods. The ablation studies confirm the effectiveness of major innovations, and the case studies validate AGLDA's ability in application for discovering potential disease-related lncRNA candidates.

Sterling: Synergistic Representation Learning on Bipartite Graphs

A fundamental challenge of bipartite graph representation learning is how to extract informative node embeddings. Self-Supervised Learning (SSL) is a promising paradigm to address this challenge. Most recent bipartite graph SSL methods are based on contrastive learning which learns embeddings by discriminating positive and negative node pairs. Contrastive learning usually requires a large number of negative node pairs, which could lead to computational burden and semantic errors. In this paper, we introduce a novel synergistic representation learning model (STERLING) to learn node embeddings without negative node pairs. STERLING preserves the unique local and global synergies in bipartite graphs. The local synergies are captured by maximizing the similarity of the inter-type and intra-type positive node pairs, and the global synergies are captured by maximizing the mutual information of co-clusters. Theoretical analysis demonstrates that STERLING could improve the connectivity between different node types in the embedding space. Extensive empirical evaluation on various benchmark datasets and tasks demonstrates the effectiveness of STERLING for extracting node embeddings.