Target Graph Research Articles

DTA-GTOmega: Enhancing Drug-Target Binding Affinity Prediction with Graph Transformers Using OmegaFold Protein Structures

Understanding drug-protein interactions is crucial for elucidating drug mechanisms and optimizing drug development. However, existing methods have limitations in representing the three-dimensional structure of targets and capturing the complex relationships between drugs and targets. This study proposes a new method, DTA-GTOmega, for predicting drug-target binding affinity. DTA-GTOmega utilizes OmegaFold to predict protein three-dimensional structure and construct target graphs, while processing drug SMILES sequences with RDKit to generate drug graphs. By employing multi-layer graph transformer modules and co-attention modules, this method effectively integrates atomic-level features of drugs and residue-level features of targets, accurately modeling the complex interactions between drugs and targets, thereby significantly improving the accuracy of binding affinity predictions. Our method outperforms existing techniques on benchmark datasets such as KIBA, Davis, and BindingDB_Kd under cold-start setting. Moreover, DTA-GTOmega demonstrates competitive performance in real-world DTI scenarios involving DrugBank data and drug-target interactions related to cardiovascular and nervous system-related diseases, highlighting its robust generalization capabilities. Additionally, the introduced DTI evaluation metrics further validate DTA-GTOmega’s potential in handling imbalanced data.

The complexity of growing a graph

We study a new algorithmic process of graph growth which starts from a single initial vertex and operates in discrete time-steps, called slots. In every slot, the graph grows via two operations (i) vertex generation and (ii) edge activation. The process completes at the last slot where a (possibly empty) subset of the edges of the graph are removed. Removed edges are called excess edges. The main problem investigated in this paper is: Given a target graph G, design an algorithm that outputs a process that grows G, called a growth schedule. Additionally, we aim to minimize the total number of slots k and of excess edges ℓ used by the process. We provide both positive and negative results, with our main focus being either schedules with sub-linear number of slots or with no excess edges.

Structure-inclusive similarity based directed GNN: a method that can control information flow to predict drug-target binding affinity.

Exploring the association between drugs and targets is essential for drug discovery and repurposing. Comparing with the traditional methods that regard the exploration as a binary classification task, predicting the drug-target binding affinity can provide more specific information. Many studies work based on the assumption that similar drugs may interact with the same target. These methods constructed a symmetric graph according to the undirected drug similarity or target similarity. Although these similarities can measure the difference between two molecules, it is unable to analyze the inclusion relationship of their substructure. For example, if drug A contains all the substructures of drug B, then in the message-passing mechanism of the graph neural network, drug A should acquire all the properties of drug B, while drug B should only obtain some of the properties of A. To this end, we proposed a structure-inclusive similarity (SIS) which measures the similarity of two drugs by considering the inclusion relationship of their substructures. Based on SIS, we constructed a drug graph and a target graph, respectively, and predicted the binding affinities between drugs and targets by a graph convolutional network-based model. Experimental results show that considering the inclusion relationship of the substructure of two molecules can effectively improve the accuracy of the prediction model. The performance of our SIS-based prediction method outperforms several state-of-the-art methods for drug-target binding affinity prediction. The case studies demonstrate that our model is a practical tool to predict the binding affinity between drugs and targets. Source codes and data are available at https://github.com/HuangStomach/SISDTA.

Joint estimation for multisource Gaussian graphical models based on transfer learning

This study considers data from multiple sources for Gaussian graphical models with one target graph and several auxiliary graphs. We propose a method called joint estimation for multisource Gaussian graphical models (JEM-GGM) to achieve a stable and accurate estimate of the target graph. Using the information from the auxiliary graphs, the proposed method is used to effectively solve the problem of small sample sizes. In this method, equivalent regression models are developed for graphs and the difference between the auxiliary and target graphs was penalized to ensure computational efficiency and improve estimation accuracy. Simulations revealed that the proposed method always outperformed other methods in terms of estimation and prediction accuracy. The application of this method to breast and lymphatic cancer cell lines reveals that the proposed method always obtains a sparse collection of important genome pairs.

ArcMatch: high-performance subgraph matching for labeled graphs by exploiting edge domains

Consider a large labeled graph (network), denoted the target. Subgraph matching is the problem of finding all instances of a small subgraph, denoted the query, in the target graph. Unlike the majority of existing methods that are restricted to graphs with labels solely on vertices, our proposed approach, named can effectively handle graphs with labels on both vertices and edges. ntroduces an efficient new vertex/edge domain data structure filtering procedure to speed up subgraph queries. The procedure, called path-based reduction, filters initial domains by scanning them for paths up to a specified length that appear in the query graph. Additionally, ncorporates existing techniques like variable ordering and parent selection, as well as adapting the core search process, to take advantage of the information within edge domains. Experiments in real scenarios such as protein–protein interaction graphs, co-authorship networks, and email networks, show that s faster than state-of-the-art systems varying the number of distinct vertex labels over the whole target graph and query sizes.

GCL-Leak: Link Membership Inference Attacks against Graph Contrastive Learning

Graph contrastive learning (GCL) has emerged as a successful method for self-supervised graph learning. It involves generating augmented views of a graph by augmenting its edges and aims to learn node embeddings that are invariant to graph augmentation. Despite its effectiveness, the potential privacy risks associated with GCL models have not been thoroughly explored. In this paper, we delve into the privacy vulnerability of GCL models through the lens of link membership inference attacks (LMIA). Specifically, we focus on the federated setting where the adversary has white-box access to the node embeddings of all the augmented views generated by the target GCL model. Designing such white-box LMIAs against GCL models presents a significant and unique challenge due to potential variations in link memberships among node pairs in the target graph and its augmented views. This variability renders members indistinguishable from non-members when relying solely on the similarity of their node embeddings in the augmented views. To address this challenge, our in-depth analysis reveals that the key distinguishing factor lies in the similarity of node embeddings within augmented views where the node pairs share identical link memberships as those in the training graph. However, this poses a second challenge, as information about whether a node pair has identical link membership in both the training graph and augmented views is only available during the attack training phase. This demands the attack classifier to handle the additional “identical-membership" information which is available only for training and not for testing. To overcome this challenge, we propose GCL-LEAK, the first link membership inference attack against GCL models. The key component of GCL-LEAK is a new attack classifier model designed under the “Learning Using Privileged Information (LUPI)” paradigm, where the privileged information of “same-membership” is encoded as part of the attack classifier's structure. Our extensive set of experiments on four representative GCL models showcases the effectiveness of GCL-LEAK. Additionally, we develop two defense mechanisms that introduce perturbation to the node embeddings. Our empirical evaluation demonstrates that both defense mechanisms significantly reduce attack accuracy while preserving the accuracy of GCL models.

Graph domain adaptation with localized graph signal representations

In this paper we propose a domain adaptation algorithm designed for graph domains. Given a source graph with many labeled nodes and a target graph with few or no labeled nodes, we aim to estimate the target labels by making use of the similarity between the characteristics of the variation of the label functions on the two graphs. Our assumption about the source and the target domains is that the local behavior of the label function, such as its spread and speed of variation on the graph, bears resemblance between the two graphs. We estimate the unknown target labels by solving an optimization problem where the label information is transferred from the source graph to the target graph based on the prior that the projections of the label functions onto localized graph bases be similar between the source and the target graphs. In order to efficiently capture the local variation of the label functions on the graphs, spectral graph wavelets are used as the graph bases. Experimentation on various data sets shows that the proposed method yields quite satisfactory classification accuracy compared to reference domain adaptation methods.

The Fine-Grained Complexity of Graph Homomorphism Parameterized by Clique-Width

The generic homomorphism problem, which asks whether an input graph \(G\) admits a homomorphism into a fixed target graph \(H\) , has been widely studied in the literature. In this article, we provide a fine-grained complexity classification of the running time of the homomorphism problem with respect to the clique-width of \(G\) (denoted \({\operatorname{cw}}\) ) for virtually all choices of \(H\) under the Strong Exponential Time Hypothesis. In particular, we identify a property of \(H\) called the signature number \(s(H)\) and show that for each \(H\) , the homomorphism problem can be solved in time \(\mathcal{O^{*}}(s(H)^{{\operatorname{cw}}})\) . Crucially, we then show that this algorithm can be used to obtain essentially tight upper bounds. Specifically, we provide a reduction that yields matching lower bounds for each \(H\) that is either a projective core or a graph admitting a factorization with additional properties—allowing us to cover all possible target graphs under long-standing conjectures.

TO-UGDA: target-oriented unsupervised graph domain adaptation

Graph domain adaptation (GDA) aims to address the challenge of limited label data in the target graph domain. Existing methods such as UDAGCN, GRADE, DEAL, and COCO for different-level (node-level, graph-level) adaptation tasks exhibit variations in domain feature extraction, and most of them solely rely on representation alignment to transfer label information from a labeled source domain to an unlabeled target domain. However, this approach can be influenced by irrelevant information and usually ignores the conditional shift of the downstream predictor. To effectively address this issue, we introduce a target-oriented unsupervised graph domain adaptive framework for graph adaptation called TO-UGDA. Particularly, domain-invariant feature representations are extracted using graph information bottleneck. The discrepancy between two domains is minimized using an adversarial alignment strategy to obtain a unified feature distribution. Additionally, the meta pseudo-label is introduced to enhance downstream adaptation and improve the model’s generalizability. Through extensive experimentation on real-world graph datasets, it is proved that the proposed framework achieves excellent performance across various node-level and graph-level adaptation tasks.

Information filtering and interpolating for semi-supervised graph domain adaptation

Graph domain adaptation, which falls under the umbrella of graph transfer learning, involves transferring knowledge from a labeled source graph to improve prediction accuracy on an unlabeled target graph, where both graphs have identical label spaces but exhibit distribution discrepancies due to temporal data shifts or distinct data collection methods. This adaptation is complicated by the challenges of graph-specific domain discrepancies and cross-graph label scarcity. This paper proposes a semi-supervised Graph domain adaptation method via Information Filtering and Interpolating (GIFI). Specifically, GIFI utilizes a parameterized graph reduction module and a variational information bottleneck to adequately filter out irrelevant information from the source and target graphs to eliminate distribution discrepancy. GIFI also introduces an interpolation-enhanced pseudo-labeling strategy for cross-graph semi-supervised learning, which can mitigate model over-fitting on domain-specific features and limited labeled nodes, thus improving the model’s adaptation and discriminative capability. Experimental results on various graph domain adaptation benchmarks demonstrate GIFI’s superior performance over state-of-the-art methods. Our code is available at https://github.com/joe817/GIFI.

Rethinking Propagation for Unsupervised Graph Domain Adaptation

Unsupervised Graph Domain Adaptation (UGDA) aims to transfer knowledge from a labelled source graph to an unlabelled target graph in order to address the distribution shifts between graph domains. Previous works have primarily focused on aligning data from the source and target graph in the representation space learned by graph neural networks (GNNs). However, the inherent generalization capability of GNNs has been largely overlooked. Motivated by our empirical analysis, we reevaluate the role of GNNs in graph domain adaptation and uncover the pivotal role of the propagation process in GNNs for adapting to different graph domains. We provide a comprehensive theoretical analysis of UGDA and derive a generalization bound for multi-layer GNNs. By formulating GNN Lipschitz for k-layer GNNs, we show that the target risk bound can be tighter by removing propagation layers in source graph and stacking multiple propagation layers in target graph. Based on the empirical and theoretical analysis mentioned above, we propose a simple yet effective approach called A2GNN for graph domain adaptation. Through extensive experiments on real-world datasets, we demonstrate the effectiveness of our proposed A2GNN framework.

Open-Set Graph Domain Adaptation via Separate Domain Alignment

Domain adaptation has become an attractive learning paradigm, as it can leverage source domains with rich labels to deal with classification tasks in an unlabeled target domain. A few recent studies develop domain adaptation approaches for graph-structured data. In the case of node classification task, current domain adaptation methods only focus on the closed-set setting, where source and target domains share the same label space. A more practical assumption is that the target domain may contain new classes that are not included in the source domain. Therefore, in this paper, we introduce a novel and challenging problem for graphs, i.e., open-set domain adaptive node classification, and propose a new approach to solve it. Specifically, we develop an algorithm for efficient knowledge transfer from a labeled source graph to an unlabeled target graph under a separate domain alignment (SDA) strategy, in order to learn discriminative feature representations for the target graph. Our goal is to not only correctly classify target nodes into the known classes, but also classify unseen types of nodes into an unknown class. Experimental results on real-world datasets show that our method outperforms existing methods on graph domain adaptation.

SAT-Based Algorithms for Regular Graph Pattern Matching

Graph matching is a fundamental problem in pattern recognition, with many applications such as software analysis and computational biology. One well-known type of graph matching problem is graph isomorphism, which consists of deciding if two graphs are identical. Despite its usefulness, the properties that one may check using graph isomorphism are rather limited, since it only allows strict equality checks between two graphs. For example, it does not allow one to check complex structural properties such as if the target graph is an arbitrary length sequence followed by an arbitrary size loop. We propose a generalization of graph isomorphism that allows one to check such properties through a declarative specification. This specification is given in the form of a Regular Graph Pattern (ReGaP), a special type of graph, inspired by regular expressions, that may contain wildcard nodes that represent arbitrary structures such as variable-sized sequences or subgraphs. We propose a SAT-based algorithm for checking if a target graph matches a given ReGaP. We also propose a preprocessing technique for improving the performance of the algorithm and evaluate it through an extensive experimental evaluation on benchmarks from the CodeSearchNet dataset.

Relational Graph Convolution Network with Multi Features for AntiCOVID-19 Drugs Discovery using 3CLpro Potential Target

Background: The potential of graph neural networks (GNNs) to revolutionize the analysis of non-Euclidean data has gained attention recently, making them attractive models for deep machine learning. However, insufficient compound or moleculargraphs and feature representations might significantly impair and jeopardize their full potential. Despite the devastating impacts of ongoing COVID-19 across the globe, for which there is no drug with proven efficacy that has been shown tobe effective. As various stages of drug discovery and repositioning require the accurate prediction of drugtarget interactions(DTI), here, we propose a relational graph convolution network using multi-features based on the developed drug chemicalcompound-coronavirus target graph representation and combination of features. During the implementation of the model, we further introduced the use of not only the feature module to understand the topological structure of drugs but also the structure of the proven drug target (i.e., 3CLpro) for SARS-Cov-2 that shares a genome sequence similar to that of other members of the beta-coronavirus group such as SARS-Cov, MERS-CoV, bat coronavirus. Our feature comprises topologicalinformation in molecular SMILES and local chemical context in the SMILES sequence for the drug chemical compound and drug target. Our proposed method prevailed with high and compelling performance accuracy of 97.30% which could beprioritized as the potential and promising prediction route for the development of novel oral antiviral medicine for COVID-19 drugs. Objective: Forecasting DTI stands as a pivotal aspect of drug discovery. The focus on computational methods in DTI prediction has intensified due to the considerable expense and time investment associated with conducting extensive in vitro and in vivo experiments. Machine learning techniques, particularly deep learning, have found broad applications in DTI prediction. We are convinced that this study could be prioritized and utilized as the promising predictive route for the development of novel oral antiviral treatments for COVID-19 and other variants of coronaviruses. Methods: This study addressed the problem of COVID-19 drugs using proposed RGCN with multifeatures as an attractive and potential route. This study focused mainly on the prediction of novel antiviral drugs against coronaviruses using graph-based methodology, namely RGCN. This research further utilized the features of both drugs and common potential drug targets found in betacoronaviruses group to deepen understanding of their underlying relation. Results: Our suggested approach prevailed with a high and convincing performance accuracy of 97.30%, which may be utilizedas a top priority to support and advance this field in the prediction and development of novel antiviral treatments against coronaviruses and their variants. Conclusion: We recursively performed experiments using the proposed method on our constructed DCCCvT graph dataset from our collected dataset with various single and multiple combinations of features and found that our model had achieved comparable best-averaged accuracy performance on T7 features followed by a combination of T7, R6, and L8. The proposed model implemented in this investigation turns out to outperform the previous related works.

Semi-supervised domain adaptation on graphs with contrastive learning and minimax entropy

Label scarcity in a graph is frequently encountered in real-world applications due to the high cost of data labeling. To this end, semi-supervised domain adaptation (SSDA) on graphs aims to leverage the knowledge of a labeled source graph to aid in node classification on a target graph with limited labels. SSDA tasks need to overcome the domain gap between the source and target graphs. However, to date, this challenging research problem has yet to be formally considered by the existing approaches designed for cross-graph node classification. This paper proposes a novel method called SemiGCL to tackle the graph Semi-supervised domain adaptation with Graph Contrastive Learning and minimax entropy training. SemiGCL generates informative node representations by contrasting the representations learned from a graph’s local and global views. Additionally, SemiGCL is adversarially optimized with the entropy loss of unlabeled target nodes to reduce domain divergence. Experimental results on benchmark datasets demonstrate that SemiGCL outperforms the state-of-the-art baselines on the SSDA tasks.

INGNN-DTI: prediction of drug-target interaction with interpretable nested graph neural network and pretrained molecule models.

Drug-target interaction (DTI) prediction aims to identify interactions between drugs and protein targets. Deep learning can automatically learn discriminative features from drug and protein target representations for DTI prediction, but challenges remain, making it an open question. Existing approaches encode drugs and targets into features using deep learning models, but they often lack explanations for underlying interactions. Moreover, limited labeled DTIs in the chemical space can hinder model generalization. We propose an interpretable nested graph neural network for DTI prediction (iNGNN-DTI) using pre-trained molecule and protein models. The analysis is conducted on graph data representing drugs and targets by using a specific type of nested graph neural network, in which the target graphs are created based on 3D structures using Alphafold2. This architecture is highly expressive in capturing substructures of the graph data. We use a cross-attention module to capture interaction information between the substructures of drugs and targets. To improve feature representations, we integrate features learned by models that are pre-trained on large unlabeled small molecule and protein datasets, respectively. We evaluate our model on three benchmark datasets, and it shows a consistent improvement on all baseline models in all datasets. We also run an experiment with previously unseen drugs or targets in the test set, and our model outperforms all of the baselines. Furthermore, the iNGNN-DTI can provide more insights into the interaction by visualizing the weights learned by the cross-attention module. The source code of the algorithm is available at https://github.com/syan1992/iNGNN-DTI.

The Realizability of Theta Graphs as Reconfiguration Graphs of Minimum Independent Dominating Sets

Abstract The independent domination number i(G) of a graph G is the minimum cardinality of a maximal independent set of G, also called an i(G)-set. The i-graph of G, denoted ℐ (G), is the graph whose vertices correspond to the i(G)-sets, and where two i(G)-sets are adjacent if and only if they differ by two adjacent vertices. Not all graphs are i-graph realizable, that is, given a target graph H, there does not necessarily exist a source graph G such that H = ℐ (G). We consider a class of graphs called “theta graphs”: a theta graph is the union of three internally disjoint nontrivial paths with the same two distinct end vertices. We characterize theta graphs that are i-graph realizable, showing that there are only finitely many that are not. We also characterize those line graphs and claw-free graphs that are i-graphs, and show that all 3-connected cubic bipartite planar graphs are i-graphs.

Counting List Homomorphisms from Graphs of Bounded Treewidth: Tight Complexity Bounds

The goal of this work is to give precise bounds on the counting complexity of a family of generalized coloring problems (list homomorphisms) on bounded-treewidth graphs. Given graphs G , H , and lists L (v) ⊆ V(H) for every v ∈ V(G) , a f:V(G) → V(H) that preserves the edges (i.e., uv ∈ E(G) implies f(u)f(v) ∈ E(H) ) and respects the lists (i.e., f(v) ∈ L(v) ). Standard techniques show that if G is given with a tree decomposition of width t , then the number of list homomorphisms can be counted in time |V(H)| t ⋅ n 𝒪(1) . Our main result is determining, for every fixed graph H , how much the base |V(H)| in the running time can be improved. For a connected graph H , we define irr( H ) in the following way: if H has a loop or is nonbipartite, then irr( H ) is the maximum size of a set S⊆ V(H) where any two vertices have different neighborhoods; if H is bipartite, then irr( H ) is the maximum size of such a set that is fully in one of the bipartition classes. For disconnected H , we define irr( H ) as the maximum of irr( C ) over every connected component C of H . It follows from earlier results that if irr( H )=1, then the problem of counting list homomorphisms to H is polynomial-time solvable, and otherwise it is #P-hard. We show that, for every fixed graph H , the number of list homomorphisms from (G,L) to H — can be counted in time \(\operatorname{irr}(H)^t\cdot n^{\mathcal {O}(1)}\) if a tree decomposition of G having width at most t is given in the input, and, — given that \(\operatorname{irr}(H)\ge 2\) , cannot be counted in time \((\operatorname{irr}(H)-\varepsilon)^t\cdot n^{\mathcal {O}(1)}\) for any \(\varepsilon \gt 0\) , even if a tree decomposition of G having width at most t is given in the input, unless the Counting Strong Exponential-Time Hypothesis (#SETH) fails. Thereby, we give a precise and complete complexity classification featuring matching upper and lower bounds for all target graphs with or without loops.

GALA: Graph Diffusion-based Alignment with Jigsaw for Source-free Domain Adaptation.

Source-free domain adaptation is a crucial machine learning topic, as it contains numerous applications in the real world, particularly with respect to data privacy. Existing approaches predominantly focus on Euclidean data, such as images and videos, while the exploration of non-Euclidean graph data remains scarce. Recent graph neural network (GNN) approaches could suffer from serious performance decline due to domain shift and label scarcity in source-free adaptation scenarios. In this study, we propose a novel method named Graph Diffusion-based Alignment with Jigsaw (GALA) tailored for source-free graph domain adaptation. To achieve domain alignment, GALA employs a graph diffusion model to reconstruct source-style graphs from target data. Specifically, a score-based graph diffusion model is trained using source graphs to learn the generative source styles. Then, we introduce perturbations to target graphs via a stochastic differential equation instead of sampling from a prior, followed by the reverse process to reconstruct source-style graphs. We feed them into an off-the-shelf GNN and introduce class-specific thresholds with curriculum learning, which can generate accurate and unbiased pseudo-labels for target graphs. Moreover, we develop a simple yet effective graph mixing strategy named graph jigsaw to combine confident graphs and unconfident graphs, which can enhance generalization capabilities and robustness via consistency learning. Extensive experiments on benchmark datasets validate the effectiveness of GALA. The source code is available at https://github.com/luo-junyu/GALA.

Graph-theoretical optimization of fusion-based graph state generation

Graph states are versatile resources for various quantum information processing tasks, including measurement-based quantum computing and quantum repeaters. Although the type-II fusion gate enables all-optical generation of graph states by combining small graph states, its non-deterministic nature hinders the efficient generation of large graph states. In this work, we present a graph-theoretical strategy to effectively optimize fusion-based generation of any given graph state, along with a Python package OptGraphState. Our strategy comprises three stages: simplifying the target graph state, building a fusion network, and determining the order of fusions. Utilizing this proposed method, we evaluate the resource overheads of random graphs and various well-known graphs. Additionally, we investigate the success probability of graph state generation given a restricted number of available resource states. We expect that our strategy and software will assist researchers in developing and assessing experimentally viable schemes that use photonic graph states.