Graph Architecture Research Articles

FGTN: Fragment-based graph transformer network for predicting reproductive toxicity.

Reproductive toxicity is one of the important issues in chemical safety. Traditional laboratory testing methods are costly and time-consuming with raised ethical issues. Only a few in silico models have been reported to predict human reproductive toxicity, but none of them make full use of the topological information of compounds. In addition, most existing atom-based graph neural network methods focus on attributing model predictions to individual nodes or edges rather than chemically meaningful fragments or substructures. In current studies, we develop a novel fragment-based graph transformer network (FGTN) approach to generate the QSAR model of human reproductive toxicity by considering internal topological structure information of compounds. In the FGTN model, the compound is represented by a graph architecture using fragments to be nodes and bonds linking two fragments to be edges. A super molecule-level node is further proposed to connect all fragment nodes by undirected edges, obtaining global molecular features from fragment embeddings. The FGTN model achieved an accuracy (ACC) of 0.861 and an area under the receiver operating characteristic curve (AUC) value of 0.914 on nonredundant blind tests, outperforming traditional fingerprint-based machine learning models and atom-based GCN model. The FGTN model can attribute toxic predictions to fragments, generating specific structural alerts for the positive compound. Moreover, FGTN may also have the capability to distinguish various chemical isomers. We believe that FGTN can be used as a reliable and effective tool for human reproductive toxicity prediction in contribution to the advancement of chemical safety assessment.

Contrastive meta-reinforcement learning for heterogeneous graph neural architecture search

Heterogeneous Graph Neural Networks (HGNNs) have demonstrated significant success in capturing complex interactions within heterogeneous graphs to learn graph representations. However, designing effective HGNN architecture is a challenging task that demands substantial domain knowledge and human effort. Fortunately, the advent of Heterogeneous Graph Architecture Search (HGNAS) has automated this process and yielded networks surpassing those designed by humans. Nonetheless, conventional HGNAS approaches are limited in their ability to generalize to unseen or diverse task types. To address these limitations, we propose an efficient method called Contrastive Meta-reinforcement learning-based Heterogeneous Graph Neural Architecture Search (CM-HGNAS). Our approach aims to produce high-performance networks tailored to different types of tasks. Firstly, we use gradient-based meta-learning to rapidly adapt to new tasks by leveraging the knowledge acquired from meta-training tasks. Secondly, since our meta-training tasks encompass various types of tasks, including node classification and link prediction, we employ contrastive learning to unify the evaluation metric across all tasks. This unification enhances the generalization capabilities of our model across diverse tasks. Experimental results demonstrate that our method successfully generalizes to unseen tasks and surpasses existing HGNN baselines and HGNAS techniques in terms of performance.

The equitable non-split domination number of graphs

Finding a group of dominant servers is a necessary step towards optimising service provisioning to all clients in the complex world of server-client networks. This crucial decision rests on the non-split dominance number in the graph architecture. We present the idea of an equitable non-split domination number in graphs to address the possible problem of unequal task distribution among servers. When the vertices of a dominating set, Dens, have the same colour in an equitable colouring scheme, the set is said to be equitable. The induced subgraph ⟨V −D⟩ also maintains its connectivity. In particular, the application of equitable non-split domination numbers to graphs like the Triangular Grid, Antiprism, Wheel, and Complete Graphs is explored in this study.

SecuriDN: A Modeling Tool Supporting the Early Detection of Cyberattacks to Smart Energy Systems

SecuriDN v. 0.1 is a tool for the representation of the assets composing the IT and the OT subsystems of Distributed Energy Resources (DERs) control networks and the possible cyberattacks that can threaten them. It is part of a platform that allows the evaluation of the security risks of DER control systems. SecuriDN is a multi-formalism tool, meaning that it manages several types of models: architecture graph, attack graphs and Dynamic Bayesian Networks (DBNs). In particular, each asset in the architecture is characterized by an attack graph showing the combinations of attack techniques that may affect the asset. By merging the attack graphs according to the asset associations in the architecture, a DBN is generated. Then, the evidence-based and time-driven probabilistic analysis of the DBN permits the quantification of the system security level. Indeed, the DBN probabilistic graphical model can be analyzed through inference algorithms, suitable for forward and backward assessment of the system’s belief state. In this paper, the features and the main goals of SecuriDN are described and illustrated through a simplified but realistic case study.

Sparse graphs-based dynamic attention networks

In previous research, the prevailing assumption was that Graph Neural Networks (GNNs) precisely depicted the interconnections among nodes within the graph's architecture. Nonetheless, real-world graph datasets are often rife with noise, elements that can disseminate through the network and ultimately affect the outcome of the downstream tasks. Facing the complex fabric of real-world graphs and the myriad potential disturbances, we introduce the Sparse Graph Dynamic Attention Networks (SDGAT) in this research. SDGAT employs the L0 regularization technique to achieve a sparse representation of the graph structure, which eliminates noise and generates a more concise sparse graph. Building upon this foundation, the model integrates a dynamic attention mechanism, allowing it to selectively focus on key nodes and edges, filter out irrelevant data, and simultaneously facilitate effective feature aggregation with important neighbors. To evaluate the performance of SDGAT, we conducted experiments on three citation datasets and compared its performance against commonly employed models. The outcomes indicate that SDGAT excels in node classification tasks, notably on the Cora dataset, with an accuracy rate of 85.29%, marking a roughly 3% enhancement over the majority of baseline models. The experimental findings provide evidence that SDGAT delivers effective performance on all three citation datasets, underscoring the efficacy of the dynamic attention network built upon a sparse graph.

Towards multimodal graph neural networks for surgical instrument anticipation

PurposeDecision support systems and context-aware assistance in the operating room have emerged as the key clinical applications supporting surgeons in their daily work and are generally based on single modalities. The model- and knowledge-based integration of multimodal data as a basis for decision support systems that can dynamically adapt to the surgical workflow has not yet been established. Therefore, we propose a knowledge-enhanced method for fusing multimodal data for anticipation tasks.MethodsWe developed a holistic, multimodal graph-based approach combining imaging and non-imaging information in a knowledge graph representing the intraoperative scene of a surgery. Node and edge features of the knowledge graph are extracted from suitable data sources in the operating room using machine learning. A spatiotemporal graph neural network architecture subsequently allows for interpretation of relational and temporal patterns within the knowledge graph. We apply our approach to the downstream task of instrument anticipation while presenting a suitable modeling and evaluation strategy for this task.ResultsOur approach achieves an F1 score of 66.86% in terms of instrument anticipation, allowing for a seamless surgical workflow and adding a valuable impact for surgical decision support systems. A resting recall of 63.33% indicates the non-prematurity of the anticipations.ConclusionThis work shows how multimodal data can be combined with the topological properties of an operating room in a graph-based approach. Our multimodal graph architecture serves as a basis for context-sensitive decision support systems in laparoscopic surgery considering a comprehensive intraoperative operating scene.

Cloud-native systems resilience assessments based on kubernetes architecture graph

Cloud-native systems resilience assessments based on kubernetes architecture graph

Continuous-Time Quantum Walk in Glued Trees: Localized State-Mediated Almost Perfect Quantum-State Transfer.

In this work, the dynamics of a quantum walker on glued trees is revisited to understand the influence of the architecture of the graph on the efficiency of the transfer between the two roots. Instead of considering regular binary trees, we focus our attention on leafier structures where each parent node could give rise to a larger number of children. Through extensive numerical simulations, we uncover a significant dependence of the transfer on the underlying graph architecture, particularly influenced by the branching rate (M) relative to the root degree (N). Our study reveals that the behavior of the walker is isomorphic to that of a particle moving on a finite-size chain. This chain exhibits defects that originate in the specific nature of both the roots and the leaves. Therefore, the energy spectrum of the chain showcases rich features, which lead to diverse regimes for the quantum-state transfer. Notably, the formation of quasi-degenerate localized states due to significant disparities between M and N triggers a localization process on the roots. Through analytical development, we demonstrate that these states play a crucial role in facilitating almost perfect quantum beats between the roots, thereby enhancing the transfer efficiency. Our findings offer valuable insights into the mechanisms governing quantum-state transfer on trees, with potential applications for the transfer of quantum information.

GSScore: a novel Graphormer-based shell-like scoring method for protein-ligand docking.

Protein-ligand interactions (PLIs) are essential for cellular activities and drug discovery. But due to the complexity and high cost of experimental methods, there is a great demand for computational approaches to recognize PLI patterns, such as protein-ligand docking. In recent years, more and more models based on machine learning have been developed to directly predict the root mean square deviation (RMSD) of a ligand docking pose with reference to its native binding pose. However, new scoring methods are pressingly needed in methodology for more accurate RMSD prediction. We present a new deep learning-based scoring method for RMSD prediction of protein-ligand docking poses based on a Graphormer method and Shell-like graph architecture, named GSScore. To recognize near-native conformations from a set of poses, GSScore takes atoms as nodes and then establishes the docking interface of protein-ligand into multiple bipartite graphs within different shell ranges. Benefiting from the Graphormer and Shell-like graph architecture, GSScore can effectively capture the subtle differences between energetically favorable near-native conformations and unfavorable non-native poses without extra information. GSScore was extensively evaluated on diverse test sets including a subset of PDBBind version 2019, CASF2016 as well as DUD-E, and obtained significant improvements over existing methods in terms of RMSE, $R$ (Pearson correlation coefficient), Spearman correlation coefficient and Docking power.

A cost-benefit secured DAG based analysis of distributed energy generation in urban settings

This study presents a groundbreaking approach to energy management in urban areas, focusing on smart building frameworks. The proposed method integrates a Directed Acyclic Graph (DAG) architecture to enhance data security, transparency, and integrity within the smart building ecosystem. To optimize the energy management system, advanced algorithms, namely Harmony Search (HS) and Artificial Bee Colony (ABC), are employed. Notably, the HS and ABC algorithms undergo modifications to improve their efficiency and convergence properties. The innovative use of these algorithms aims to achieve global solutions in the complex optimization landscape of energy management. The study explores the synergistic effects of DAG, HS, and ABC, providing a robust and secure solution for urban energy challenges. The proposed framework is designed to address the dynamic nature of smart buildings, allowing for adaptive and efficient energy consumption. The findings contribute to the advancement of sustainable urban development, emphasizing the importance of secure data transactions and optimized energy utilization in the context of smart cities. The modified HS and ABC algorithms play a key role in fine-tuning the energy management process, offering a novel and effective approach to address the evolving energy needs of urban areas.

Drug–target affinity prediction with extended graph learning-convolutional networks

BackgroundHigh-performance computing plays a pivotal role in computer-aided drug design, a field that holds significant promise in pharmaceutical research. The prediction of drug–target affinity (DTA) is a crucial stage in this process, potentially accelerating drug development through rapid and extensive preliminary compound screening, while also minimizing resource utilization and costs. Recently, the incorporation of deep learning into DTA prediction and the enhancement of its accuracy have emerged as key areas of interest in the research community. Drugs and targets can be characterized through various methods, including structure-based, sequence-based, and graph-based representations. Despite the progress in structure and sequence-based techniques, they tend to provide limited feature information. Conversely, graph-based approaches have risen to prominence, attracting considerable attention for their comprehensive data representation capabilities. Recent studies have focused on constructing protein and drug molecular graphs using sequences and SMILES, subsequently deriving representations through graph neural networks. However, these graph-based approaches are limited by the use of a fixed adjacent matrix of protein and drug molecular graphs for graph convolution. This limitation restricts the learning of comprehensive feature representations from intricate compound and protein structures, consequently impeding the full potential of graph-based feature representation in DTA prediction. This, in turn, significantly impacts the models’ generalization capabilities in the complex realm of drug discovery.ResultsTo tackle these challenges, we introduce GLCN-DTA, a model specifically designed for proficiency in DTA tasks. GLCN-DTA innovatively integrates a graph learning module into the existing graph architecture. This module is designed to learn a soft adjacent matrix, which effectively and efficiently refines the contextual structure of protein and drug molecular graphs. This advancement allows for learning richer structural information from protein and drug molecular graphs via graph convolution, specifically tailored for DTA tasks, compared to the conventional fixed adjacent matrix approach. A series of experiments have been conducted to validate the efficacy of the proposed GLCN-DTA method across diverse scenarios. The results demonstrate that GLCN-DTA possesses advantages in terms of robustness and high accuracy.ConclusionsThe proposed GLCN-DTA model enhances DTA prediction performance by introducing a novel framework that synergizes graph learning operations with graph convolution operations, thereby achieving richer representations. GLCN-DTA does not distinguish between different protein classifications, including structurally ordered and intrinsically disordered proteins, focusing instead on improving feature representation. Therefore, its applicability scope may be more effective in scenarios involving structurally ordered proteins, while potentially being limited in contexts with intrinsically disordered proteins.

Graph Pooling Inference Network for Text-based VQA

Effectively leveraging objects and optical character recognition (OCR) tokens to reason out pivotal scene text is critical for the challenging Text-based Visual Question Answering (TextVQA) task. Graph-based models can effectively capture the semantic relationship among visual entities (objects and tokens) and report remarkable performance in TextVQA. However, previous efforts usually leverage all visual entities and ignore the negative effect of superfluous entities. This article presents a Graph Pooling Inference Network (GPIN), which is an evolutionary graph learning method to purify the visual entities and capture the core semantics. It is observed that the dense distribution of reduplicative objects and the crowd of semantically dependent OCR tokens usually co-exist in the image. Motivated by this, GPIN adopts an adaptive node dropping strategy to dynamically downscale semantically closed nodes for graph evolution and update. To deepen the comprehension of scene text, GPIN is a dual-path hierarchical graph architecture that progressively aggregates the evolved object graph and the evolved token graph semantics into a graph vector that serves as visual cues to facilitate the answer reasoning. It can effectively eliminate object redundancy and enhance the association of semantically continuous tokens. Experiments conducted on TextVQA and ST-VQA datasets show that GPIN achieves promising performance compared with state-of-the-art methods.

Spotting Culex pipiens from satellite: modeling habitat suitability in central Italy using Sentinel-2 and deep learning techniques.

Culex pipiens, an important vector of many vector borne diseases, is a species capable to feeding on a wide variety of hosts and adapting to different environments. To predict the potential distribution of Cx. pipiens in central Italy, this study integrated presence/absence data from a four-year entomological survey (2019-2022) carried out in the Abruzzo and Molise regions, with a datacube of spectral bands acquired by Sentinel-2 satellites, as patches of 224 × 224 pixels of 20 meters spatial resolution around each site and for each satellite revisit time. We investigated three scenarios: the baseline model, which considers the environmental conditions at the time of collection; the multitemporal model, focusing on conditions in the 2 months preceding the collection; and the MultiAdjacency Graph Attention Network (MAGAT) model, which accounts for similarities in temperature and nearby sites using a graph architecture. For the baseline scenario, a deep convolutional neural network (DCNN) analyzed a single multi-band Sentinel-2 image. The DCNN in the multitemporal model extracted temporal patterns from a sequence of 10 multispectral images; the MAGAT model incorporated spatial and climatic relationships among sites through a graph neural network aggregation method. For all models, we also evaluated temporal lags between the multi-band Earth Observation datacube date of acquisition and the mosquito collection, from 0 to 50 days. The study encompassed a total of 2,555 entomological collections, and 108,064 images (patches) at 20 meters spatial resolution. The baseline model achieved an F1 score higher than 75.8% for any temporal lag, which increased up to 81.4% with the multitemporal model. The MAGAT model recorded the highest F1 score of 80.9%. The study confirms the widespread presence of Cx. pipiens throughout the majority of the surveyed area. Utilizing only Sentinel-2 spectral bands, the models effectively capture early in advance the temporal patterns of the mosquito population, offering valuable insights for directing surveillance activities during the vector season. The methodology developed in this study can be scaled up to the national territory and extended to other vectors, in order to support the Ministry of Health in the surveillance and control strategies for the vectors and the diseases they transmit.

Graph-DETR4D: Spatio-Temporal Graph Modeling for Multi-View 3D Object Detection.

Multi-View 3D object detection (MV3D) has made tremendous progress by leveraging multiple perspective features through surrounding cameras. Despite demonstrating promising prospects in various applications, accurately detecting objects through camera view in the 3D space is extremely difficult due to the ill-posed issue in monocular depth estimation. Recently, Graph-DETR3D presents a novel graph-based 3D-2D query paradigm in aggregating multi-view images for 3D object detection and achieves competitive performance. Although it enriches the query representations with 2D image features through a learnable 3D graph, it still suffers from limited depth and velocity estimation abilities due to the adoption of a single-frame input setting. To solve this problem, we introduce a unified spatial-temporal graph modeling framework to fully leverage the multi-view imagery cues under the multi-frame inputs setting. Thanks to the flexibility and sparsity of the dynamic graph architecture, we lift the original 3D graph into the 4D space with an effective attention mechanism to automatically perceive imagery information at both spatial and temporal levels. Moreover, considering the main latency bottleneck lies in the image backbone, we propose a novel dense-sparse distillation framework for multi-view 3D object detection, to reduce the computational budget while sacrificing no detection accuracy, making it more suitable for real-world deployment. To this end, we propose Graph-DETR4D, a faster and stronger multi-view 3D object detection framework, built on top of Graph-DETR3D. Extensive experiments on nuScenes and Waymo benchmarks demonstrate the effectiveness and efficiency of Graph-DETR4D. Notably, our best model achieves 62.0% NDS on nuScenes test leaderboard. Code is available at https://github.com/zehuichen123/Graph-DETR4D.

Network-based prediction approach for cancer-specific driver missense mutations using a graph neural network

BackgroundIn cancer genomic medicine, finding driver mutations involved in cancer development and tumor growth is crucial. Machine-learning methods to predict driver missense mutations have been developed because variants are frequently detected by genomic sequencing. However, even though the abnormalities in molecular networks are associated with cancer, many of these methods focus on individual variants and do not consider molecular networks. Here we propose a new network-based method, Net-DMPred, to predict driver missense mutations considering molecular networks. Net-DMPred consists of the graph part and the prediction part. In the graph part, molecular networks are learned by a graph neural network (GNN). The prediction part learns whether variants are driver variants using features of individual variants combined with the graph features learned in the graph part.ResultsNet-DMPred, which considers molecular networks, performed better than conventional methods. Furthermore, the prediction performance differed by the molecular network structure used in learning, suggesting that it is important to consider not only the local network related to cancer but also the large-scale network in living organisms.ConclusionsWe propose a network-based machine learning method, Net-DMPred, for predicting cancer driver missense mutations. Our method enables us to consider the entire graph architecture representing the molecular network because it uses GNN. Net-DMPred is expected to detect driver mutations from a lot of missense mutations that are not known to be associated with cancer.

A superconducting quantum processor architecture design method for improving performance and reducing frequency collisions

More physical qubits and qubit connections integrated on a superconducting quantum processor can improve the ability to execute quantum programs, but on the other hand, they might increase the probability of frequency collisions. Considering a performance and frequency collisions trade-off, it is a feasible method to optimize processor architecture aiming at running specific quantum programs. To this end, we propose a method for designing superconducting quantum processor architectures for the purpose of running specific quantum programs. Different from existing methods, our method is mainly based on graph theory to optimize processor architecture design. First, we convert the trade-off problem into the optimization of the distance between two points as well as the maximum degree in the processor architecture graph. Second, we consider the actual physical routing constraints and build a mathematical model for the optimization problem. Finally, we propose an automatic processor architecture design flow based on the mathematical model, which is implemented with an improved genetic algorithm. To show the effectiveness of our method, we selected sixteen quantum programs with different qubit numbers and different functions for comparison. Simulation results show that the architecture schemes of our method outperform IBM’s general-purpose square lattice architecture schemes. Compared to the general-purpose architecture schemes, the architecture schemes of our method have an average performance improvement of 15.61% and a minimum reduction of 21.33% in the probability of frequency collisions. Furthermore, in most of the selected quantum programs, our architecture schemes perform better than the eff-5-freq’s architecture schemes, with an average 6.58% improvement in performance and a minimum 6.45% reduction in the probability of frequency collisions. Therefore, our method can provide superconducting quantum processor architecture design with better performance and lower probability of frequency collisions for quantum programs.

Reconfigurable and Privacy-Preserving Economical Operation for Interconnected DC Microgrid Clusters

This article presents a new distributed and reconfigurable strategy for global economical consensus operation of multiple, interconnected dc microgrid clusters in a leader–follower graph architecture. The generation costs associated with the distributed energy resources are optimized globally by observer-based controllers in a multilayered hierarchical control structure, equipped with privacy-preserving feature. Self-rearranging and event-based reconfiguration of graphs are also incorporated for obtaining feasible solution to the economic dispatch problem. This is an important aspect of the proposed work as, unlike previously reported research works, it equally prioritizes both voltage regulation and optimal operation. Moreover, continuity of supply to critical loads is a challenging criterion in autonomous systems in which the pinning or leader node plays a pivotal role. Therefore, the multifold benefits of choosing the highest generation unit as the leader in improving the power quality is highlighted for the first time, which remained undiscovered in previous works in literature. The feasibility of the proposed scheme is tested by time-domain simulations in MATLAB/Simulink. Moreover, the results obtained from the controller hardware in loop environment using a real-time digital simulator and field-programmable-gate-array-based WAVECT controller demonstrate the economical operation of the clusters with renewable energy sources.

The Intel Programmable and Integrated Unified Memory Architecture Graph Analytics Processor

The Intel Programmable and Integrated Unified Memory Architecture Graph Analytics Processor

TCM syndrome classification using graph convolutional network

IntroductionTraditional Chinese Medicine (TCM) diagnosis is a reasoning process through expert knowledge, in which syndrome classification is a key step for prescription recommendation and the treatment of patients. Doctors generally differentiate syndrome types according to patients’ symptoms and state elements. This paper proposes a syndrome classification method based on graph convolutional network with residual structure, to exploit the potential relationship between symptoms and state elements. MethodsWe constructed a graph convolutional network by combining symptoms and state elements for syndrome classification, called Symptoms-State elements Graph Convolutional Network (SSGCN), embedding the inherent logic of TCM diagnosis and treatment with a prescription graph. This graph architecture wherein contained the relationship between symptoms and state elements, and a multi-layer perceptron (MLP) was trained to classify different syndromes. ResultsExperiments were conducted on two self-built datasets according to two classic TCM books, i.e., Theories on Febrile Diseases and Traditional Chinese Medicine Prescription Dictionary. Accuracy, precision, recall and F1-score were adopted to evaluate the syndrome classificaiton results. Our proposed SSGCN method achieved accuracy of 75.59%, 69.63%, precision of 69.10%, 76.33%, recall of 75.63%, 66.67% and F1-score of 71.26%, 65.84% in the above two datasets, respectively. The proposed method for syndrome classification outperformed several popular methods including support vector machine, random forest, extreme gradient boosting and convolutional neural network. ConclusionsBy constructing a prescription graph in which symptoms are used as nodes and state elements are taken into account for edges, graph convolution is implemted to capture the relationship of symptoms and state elements. This model improves the performance of syndrome classification and can be further extened for some other related applications in TCM.

Triplet adversarial Learning-driven graph architecture search network augmented with Probsparse-attention mechanism for fault diagnosis under Few-shot & Domain-shift

The consistent probability distribution between training & testing data is one of the prerequisites for valid intelligent diagnosis models. Nevertheless, the ineluctable distribution discrepancies produced by variable working conditions cause intense conflicts with the above. Besides, many existing models fail to explicitly characterize the correlation between data, which makes them difficult to mine discriminative features in limited samples. Fortunately, the proposal of graph neural networks provides a potential solution to overcome the above issues. Inspired by this, this paper creatively proposes a graph architecture search network (GASN) for cross-domain fault diagnosis. Its main framework consists of graph-feature extraction modules, graph-architecture search modules and differentiated classifiers, which can effectively reduce the computational complexity and accelerate the inferential efficiency of GASN, with two distinctive characteristics: (1) Specially designed search module utilizes multi-head probsparse-attention mechanism to search the optimal graph-architecture, thus suppressing the redundancy and exclusivity among candidate graph-data; (2) Triplet loss-driven domain adversarial training strategy is proposed to enhance the domain adaptability of GASN, which can assist itself to achieve fine-grained adaptation of distinguishable architectures in response to unknown domains. Comparative results on four case studies indicate that the GASN can achieve superior performance to existing state-of-art models even under imbalanced data.