AbstractIn the crucial process of drug discovery and repurposing, precise prediction of drug-target interactions (DTIs) is paramount. This study introduces a novel DTI prediction approach—Multi-Layer Graph Attention Neural Network (MLGANN), through a groundbreaking computational framework that effectively harnesses multi-source information to enhance prediction accuracy.

摘要在药物发现和再利用的关键过程中，精确预测药物-靶标相互作用（DTI）至关重要。这项研究引入了一种新的DTI预测方法多层图注意神经网络（MLGANN），通过一个开创性的计算框架，有效地利用多源信息来提高预测精度。

MLGANN not only strides forward in constructing a multi-layer DTI network by capturing both direct interactions between drugs and targets as well as their multi-level information but also amalgamates Graph Convolutional Networks (GCN) with a self-attention mechanism to comprehensively integrate diverse data sources.

MLGANN不仅通过捕获药物和靶标之间的直接相互作用以及它们的多级信息，在构建多层DTI网络方面取得了长足的进步，而且还将图形卷积网络（GCN）与自我注意机制相结合，以全面整合各种数据源。

This method exhibited significant performance surpassing existing approaches in comparative experiments, underscoring its immense potential in elevating the efficiency and accuracy of DTI predictions. More importantly, this study accentuates the significance of considering multi-source data information and network heterogeneity in the drug discovery process, offering new perspectives and tools for future pharmaceutical research..

该方法在比较实验中表现出优于现有方法的显着性能，强调了其在提高DTI预测的效率和准确性方面的巨大潜力。更重要的是，这项研究强调了在药物发现过程中考虑多源数据信息和网络异质性的重要性，为未来的药物研究提供了新的视角和工具。。

IntroductionDrug-Target Interaction (DTI) prediction is a computational task aimed at identifying potential interactions between chemical compounds (drugs) and biological molecules (targets, typically proteins). Prediction of DTI is a crucial component in modern drug research and development. Accurate DTI prediction can provide valuable clues for drug design and expedite drug repurposing, with approximately 75% of drugs being eligible for repurposing1.

引言药物-靶标相互作用（DTI）预测是一项计算任务，旨在识别化合物（药物）和生物分子（靶标，通常是蛋白质）之间的潜在相互作用。DTI的预测是现代药物研究和开发的关键组成部分。准确的DTI预测可以为药物设计提供有价值的线索，并加速药物再利用，大约75%的药物有资格再利用1。

Traditional in vitro experimental tests can validate DTI, but this approach is both expensive and time-consuming. In recent years, computer-assisted methods for DTI prediction have garnered widespread attention, including matrix factorization2, kernel-based methods3, graph embedding techniques4, and more.

传统的体外实验测试可以验证DTI，但这种方法既昂贵又耗时。近年来，用于DTI预测的计算机辅助方法引起了广泛关注，包括矩阵分解2，基于核的方法3，图嵌入技术4等等。

The introduction of these methods narrows the search space, reduces the workload of in vitro experiments, and accelerates the drug development process.Although the importance of DTI prediction is widely recognized, most current research in this area primarily relies on the chemical structure of drugs and the protein sequences of targets for DTI prediction5,6,7.

。虽然DTI预测的重要性已得到广泛认可，但目前该领域的大多数研究主要依赖于药物的化学结构和DTI预测靶标的蛋白质序列5,6,7。

This approach neglects other crucial multi-source information about drugs and targets, such as the physicochemical properties of drugs and the relationships between targets and diseases8,9, which are equally important for DTI prediction. With deep learning demonstrating excellent performance across various domains, some researchers have also applied it to DTI prediction and have achieved promising results10,11,12.

。随着深度学习在各个领域表现出优异的性能，一些研究人员也将其应用于DTI预测，并取得了令人鼓舞的结果10,11,12。

However, methods based on neural networks or machine learning, when integrating information about drugs and targets, can only utilize the information from individual drugs or targets3,10. The result of this approach is a loss of the unique information re.

然而，基于神经网络或机器学习的方法在整合药物和靶标信息时，只能利用单个药物或靶标的信息3,10。这种方法的结果是丢失了唯一的信息re。

We construct a multi-layer DTI network for DTI prediction, which not only encapsulates multi-level information of drugs and targets but also captures the interactions between these levels. This comprehensive integration enhances the accuracy of DTI predictions.

我们构建了一个用于DTI预测的多层DTI网络，它不仅封装了药物和靶标的多层次信息，而且还捕获了这些层次之间的相互作用。。

We design a multi-layer graph attention neural network (MLGANN) that effectively captures multi-source drug and target information within the multi-layer DTI network, as well as the interactions between them, further improving DTI prediction outcomes.

我们设计了一种多层图注意神经网络（MLGANN），可以有效地捕获多层DTI网络中的多源药物和目标信息，以及它们之间的相互作用，进一步改善DTI预测结果。

Through comparative analysis with multiple models on the DTI dataset, our experimental results demonstrate the effectiveness and superiority of the proposed method in addressing the DTI prediction problem.

通过对DTI数据集上多个模型的比较分析，我们的实验结果证明了所提出的方法在解决DTI预测问题方面的有效性和优越性。

Related workDrug-target interaction (DTI) prediction is a critical task in drug discovery and development, where the goal is to identify potential interactions between drugs and target proteins. Over the years, various computational approaches have been developed to address this problem, ranging from similarity-based methods to more advanced graph-based approaches.Traditional methods for DTI prediction often rely on similarity measures, where similar drugs are assumed to interact with similar targets.

相关的工作药物-靶标相互作用（DTI）预测是药物发现和开发中的一项关键任务，其目标是确定药物与靶蛋白之间的潜在相互作用。多年来，已经开发了各种计算方法来解决这个问题，从基于相似性的方法到更先进的基于图形的方法。传统的DTI预测方法通常依赖于相似性度量，其中假设相似的药物与相似的靶标相互作用。

These methods typically compute drug-drug and target-target similarity matrices, which are then used in conjunction with machine learning models to predict interactions16. However, these approaches have limitations, such as their inability to integrate multi-source information about drugs and targets.

这些方法通常计算药物-药物和靶标-靶标相似性矩阵，然后将其与机器学习模型结合使用以预测相互作用16。然而，这些方法有局限性，例如它们无法整合有关药物和靶标的多源信息。

For example, methods like DTINet and deepDTnet fall into this category but struggle with capturing the full complexity of drug-target relationships.Recent advancements have seen the application of knowledge graphs to DTI prediction17, where biological knowledge is structured as a graph and used to infer new interactions.

例如，DTINet和deepDTnet等方法属于这一类，但难以捕捉药物-靶标关系的全部复杂性。最近的进展已经看到知识图在DTI预测中的应用17，其中生物知识被结构化为图形并用于推断新的相互作用。

Methods such as ComplEx18 and KGE_NFM19 have been developed to leverage knowledge graph embeddings for DTI prediction. While these methods provide a way to incorporate diverse biological information, they often require significant domain-specific knowledge to construct the knowledge graph, which can limit their applicability.Graph neural networks (GNNs) have gained popularity for their ability to model complex relationships in network data.

已经开发了ComplEx18和KGE\U NFM19等方法来利用知识图嵌入进行DTI预测。虽然这些方法提供了一种整合多种生物信息的方法，但它们通常需要大量特定领域的知识来构建知识图，这可能会限制其适用性。图形神经网络（GNNs）因其在网络数据中建模复杂关系的能力而广受欢迎。

Several GNN-based methods have been proposed for DTI prediction, including IMCHGAN15, SGCL-DTI14, and MHGNN20. These methods treat drug-target pairs as nodes in a heterogeneous network and use GNNs to learn embeddings that.

已经提出了几种基于GNN的DTI预测方法，包括IMCHGAN15，SGCL-DTI14和MHGNN20。这些方法将药物-靶标对视为异构网络中的节点，并使用GNN来学习嵌入。

(1)

(1)

where \(||V^M|| =N= (n_d\times m_d+n_t\times m_t)\).Fig. 1The figure illustrates a multiplex layer drug-target interaction (DTI) network, which integrates multi-level information from drugs (D) and targets (T) across several layers. On the left, different layers of drug associations are shown (labeled as \(A^{D,1},A^{D,2},A^{D,3}\)), representing various relationships among drugs \(D_1, D_2, D_3\) and \(D_4\).

其中\（| | V ^ M | |=N=（N\u d \乘以M\u d+N\u t \乘以M\u t）\）。。在左边，显示了不同层次的药物关联（标记为\（A ^{D，1}，A ^{D，2}，A ^{D，3}\）），代表药物之间的各种关系\（D\U 1，D\U 2，D\U 3 \）和\（D\U 4 \）。

On the right, target associations are depicted in similar layers (\(A^{T,1}\) and \(A^{T,2}\)) with targets \(T_1,T_2\) and \(T_3\). The central part of the diagram displays the interaction between drugs and targets, where a multi-layer network structure is used to capture the complex interplay between different layers of information.

在右边，目标关联被描述为类似的层（\（A ^{T，1}\）和\（A ^{T，2}\）与目标\（T\u 1，T\u 2 \）和\（T\u 3 \）。该图的中心部分显示了药物和靶标之间的相互作用，其中使用多层网络结构来捕获不同信息层之间的复杂相互作用。

This multi-layered approach enables for more comprehensive DTI prediction by considering both intra-layer and inter-layer interactions.Full size imageMulti-layer attention graph neural networkWe propose a model called the Multi-Layer Graph Attention Neural Network (MLGANN) for DTI prediction. In the multi-layer network of DTI, apart from the interaction between drugs and targets, there is also interaction information among various properties within drugs and targets themselves.

这种多层方法通过考虑层内和层间的相互作用，可以进行更全面的DTI预测。全尺寸图像多层注意图神经网络我们提出了一种称为多层图注意神经网络（MLGANN）的DTI预测模型。在DTI的多层网络中，除了药物和靶标之间的相互作用外，药物和靶标本身的各种性质之间也存在相互作用信息。

Therefore, we utilize the designed MLGANN to capture both the interaction information between drugs and targets and the multi-source information within drugs and targets.Multi-layer neighbor aggregationLet \(X\in \mathbb {R}^{N\times f}\) represent the initial features of nodes in the multi-layer DTI network, where f denotes the dimension of the embedding space.

因此，我们利用设计的MLGANN来捕获药物和靶标之间的相互作用信息以及药物和靶标内的多源信息。多层邻居聚合let \（X \ in \ mathbb{R}^{N \乘以f}\）表示多层DTI网络中节点的初始特征，其中f表示嵌入空间的维数。

We apply graph neural networks to learn embeddings for drugs and targets on the multi-layer DTI network. Specifically, in our model, we employ Graph Convolutional Networks (GCN), as the.

我们应用图神经网络来学习多层DTI网络上药物和靶标的嵌入。具体而言，在我们的模型中，我们使用图卷积网络（GCN）作为。

(2)

(2)

where \(X^{(0)} =X,\hat{A}=A^M+I^M\), and \(A^M\) is the adjacency matrix of the multi-layer DTI network, \(I^M\) is an identity matrix of the same size as \(A^M\), \(\hat{D}\) is a diagonal matrix with \(\hat{D}_{ij} = \sum \nolimits _{j = 1}\hat{A}_{ij}\), \(W^{(p)}\in \mathbb {R}^{f\times f}\) is a trainable weight matrix, \(\sigma\) is a nonlinear activation function ReLU.For a node \(v\in G^M\) (representing either a drug or a target), Eq.

其中\（X ^{（0）}=X，{A}=A ^ M+I ^ M \），并且\（A ^ M \）是多层DTI网络的邻接矩阵，\（I ^ M \）是与\（A ^ M \）大小相同的单位矩阵，\（{D}）是具有\（\ hat{D}_{ij}=\sum\n限制{j=1}\hat{A}_{ij}\），\（W ^{（p）}\ in \ mathbb{R}^{f \ times f}\）是一个可训练的权重矩阵，\（\ sigma \）是一个非线性激活函数ReLU。。

(2) updates the embedding of that node as follows:$$\begin{aligned} x_v^{(p)} = \sigma \left( W^{(p)}\sum \limits _{u}\dfrac{1}{\alpha _{vu}}x_u^{(p-1)}\right) ,\quad u\in \{v\cup N_v\cup C_v\} \end{aligned}$$.

（2） 。

(3)

(3)

where \(\alpha _{vu}\) is the normalized weight, \(N_v\) is the set of neighbors of node v in layer of \(G^M\), and \(C_v\) is the set of nodes that correspond to the same drug/target as node v. Therefore, MLGANN not only aggregates the neighbors of node v in layer of \(G^M\) (similar to what GCN does) but also embeds nodes corresponding to the same drug/target in different layers of \(G^M\).

。

This allows information to be transmitted across different layers of \(G^M\). By leveraging information from different layers of \(G^M\), MLGANN can learn better representations for each node, especially for nodes with limited interactions in a particular layer of \(G^M\). This is the main distinction between MLGANN and existing other network embedding methods.Multi-layer attention poolingWe concatenate all representations learned by the P-layer GCN to obtain the final node embedding:$$\begin{aligned} \begin{aligned} z_i^{D,k} = \left[ z_i^{D,k(0)},z_i^{D,k(1)},\cdots ,z_i^{D,k(P)}\right] \\ z_j^{T,l} = \left[ z_j^{T,l(0)},z_j^{T,l(1)},\cdots ,z_j^{T,l(P)}\right] , \end{aligned} \end{aligned}$$.

这允许信息在“G ^ M”的不同层之间传输。通过利用来自不同层（G ^ M）的信息，MLGANN可以为每个节点学习更好的表示，特别是对于在特定层（G ^ M）中交互有限的节点。这是MLGANN与现有其他网络嵌入方法的主要区别。多层注意力池将P层GCN学习到的所有表示连接起来，以获得最终的节点嵌入：$$\开始{对齐}\开始{对齐}z\u i ^{D，k}=\左[z\u i ^{D，k（0）}，z\u i ^{D，k（1）}，\cdots，z\u i ^{D，k（P）}\右]\\z\u j ^{T，l}=\左[z\u j ^{T，l（0）}，z\u j ^{T，l（1）}，\cdots，z\u j ^{T，l（P）}\右]，\end{对齐}\ end{对齐}$$。

(4)

(4)

where \(z_i^{D,k}\) denotes final embedding of ith drug in k layer of \(G^M\) and \(z_i^{D,k(p)}\) denotes \(G^M\)’s kth layer embedding of ith drug in GCN pth layer, \(z_j^{T,l}\) represents final embedding of jth target in lth layer of \(G^M\), \(z_j^{T,l(p)}\) denotes \(G^M\)’s lth layer embedding of jth drug in GCN pth layer.To obtain the final representations of drugs and targets, we have designed a self-attention mechanism to aggregate the representation vectors of drugs and targets across different layers for DTI prediction in the \(G^M\) graph.

其中\（z\u i ^{D，k}\）表示第i种药物在\（G ^ M \）的k层中的最终嵌入，\（z\u i ^{D，k（p）}\）表示第i种药物在GCN pth层中的第k层嵌入，\（z\u j ^{T，l}\）表示第j种靶标在\（G ^ M \）的第l层中的最终嵌入，\（z\u j ^{T，l（p）}\）表示\（G ^ M \））在GCN pth层中嵌入第j种药物的第l层。为了获得药物和靶标的最终表示，我们设计了一种自我注意机制，以聚合不同层的药物和靶标的表示向量，以在“G ^ M”图中进行DTI预测。

The computer process is as follows:$$\begin{aligned} \begin{aligned} e_i^{D,k} = q^D \cdot LeakyReLU\left( W^Dz^{D,k}_i\right) ,\quad e_j^{T,l} = q^T \cdot LeakyReLU\left( W^Tz^{T,l}_j\right) \\ \alpha ^k_i = \frac{e_i^{D,k}}{\sum \nolimits ^{m_d}_{k'=1}e_i^{D,k'}},\quad z^D_i = \sum \limits ^{m_d}_{k=1}\alpha ^k_i z^{D,k}_i,\quad \beta ^l_j = \frac{e_i^{T,l}}{\sum \nolimits ^{m_t}_{l'=1}e_j^{T,l'}},\quad z^T_j = \sum \limits ^{m_t}_{l=1}\beta ^l_j z^{T,l}_j, \end{aligned} \end{aligned}$$.

计算机过程如下：$$\ begin{aligned}\ begin{aligned}e\u i ^{D，k}=q ^ D \ cdot LeakyReLU \ left（W ^ Dz^{D,k}_i\右），\quad e\u j ^{T，l}=q ^ T\cdot LeakyReLU\left（W ^ Tz^{T,l}_j\右）\\ alpha ^ k\u i=\ frac{e\u i ^{D，k}}{\ sum\n限制^{m_d}_{k'=1}e\u i ^{D，k'}}，quad z ^ D\u i=\ sum \ limits^{m_d}_{k=1}\α^ k\u i z^{D,k}_i，\quad\beta ^ l\u j=\frac{e\u i ^{T，l}}{\sum\nolimits^{m_t}_{l’=1}e\u j ^{T，l’}}，quad z ^ T\u j=\ sum \ limits^{m_t}_{l=1}\ beta ^ l\u j z^{T,l}_j，\ end{aligned}\ end{aligned}$$。

(5)

(5)

where \(z^D_i\in \mathbb {R}^{f'}\) and \(z^D_i\in \mathbb {R}^{f'}\) are the final representations of drugs and targets, \(W^D\in \mathbb {R}^{f'\times f'}\) and \(W^T\in \mathbb {R}^{f'\times f'}\) are trainable parameter matrices, \(q^D\in \mathbb {R}^{f'}\) and \(q^T\in \mathbb {R}^{f'}\) are trainable vectors.DTI predictionLet \(G^Y\) be the DTI network derived from the adjacency matrix \(A^Y\).

其中\（z ^ D ^ u i \ in \ mathbb{R}^{f'}\）和\（z ^ D ^ u i \ in \ mathbb{R}^ ^{f'}\）是药物和靶标的最终表示，\（W ^ D \ in \ mathbb{R}^ ^{f'}\）和\（W ^ T \ in \ mathbb{R}^ ^ ^{f'}\）是可训练的参数矩阵，\（q ^ D \ in \ mathbb{R}^{f’}和（q ^ T \ in \ mathbb{R}^{f’}）是可训练向量。DTI预测let（G ^ Y）是从邻接矩阵（A ^ Y）导出的DTI网络。

For an edge \(d_it_j\) in \(G^Y\), where \(z^D_i\) and \(z^T_j\) are final representation vectors of drug \(d_i\) and target \(t_j\), respectively. we sample a non-existing edge \(d_ut_v\) in \(G^Y\), where \(z^D_u\) and \(z^T_v\) are final representation vectors of drug \(d_u\) and target \(t_v\), respectively.

对于（G ^ Y）中的边（d ^ it ^ u j），其中（z ^ d ^ u i）和（z ^ T ^ u j）分别是药物（d ^ u i）和目标（T ^ u j）的最终表示向量。我们在（G ^ Y）中采样了一个不存在的边（d ^ u u u v），其中（z ^ d u u）和（z ^ T u v）分别是药物（d ^ u）和目标（T ^ u v）的最终表示向量。

We consider DTP \(d_it_j\) as a positive sample and \(d_ut_v\) as a negative sample. Therefore, we design the loss function based on cross-entropy as follows:$$\begin{aligned} \mathcalligra{L}=-\log \left( \sigma \left(<z^D_i,z^T_j>\right) \right) -\log \left( \sigma \left( -<z^D_u , z^T_v>\right) \right) \end{aligned}$$.

我们认为DTP \（d\u it\u j \）是正样本，而\（d\u ut\u v \）是负样本。因此，我们基于交叉熵设计损失函数如下：$$\ begin{aligned}\ mathcalligra{L}=-\ log \ left（\ sigma \ left（<z ^ D\u i，z ^ T\u j>\ right）\ right）\ log \ left（\ sigma \ left（<z ^ D\u，z ^ T\u v>\ right）\ right）\ end{aligned}$$。

(6)

(6)

where \(\sigma\) is a nonlinear activation function Sigmoid, \(<\cdot ,\cdot>\) is the inner product in Euclidean space.ExperimentsExperimental setupDatasets We collected data on DTIs from DrugBank (Version5.1.8), as well as relevant data on drug chemical structural similarity, drug-side effect relationships, drug-disease relationships, target interaction relationships, and target-disease relationships.

其中\（\ sigma \）是一个非线性激活函数Sigmoid，\（\cdot，\cdot>\）是欧几里德空间中的内积。实验实验设置数据集我们从DrugBank（版本5.1.8）收集了有关DTI的数据，以及有关药物化学结构相似性，药物副作用关系，药物-疾病关系，靶标相互作用关系和靶标-疾病关系的相关数据。

All of this data is available on DrugBank. By taking the intersection of these datasets, we obtained a total of 201 drugs, 252 targets, and 907 interactions between them. We used 607 of these interactions for training and 300 interactions for testing. The statistical information for these datasets is summarized in Table 1, where D denotes drug, T denotes target, D-D-1 denotes drug chemical structural similarity, D-D-2 denotes similarity based on drug-side effect relationships, D-D-3 denotes similarity based on drug-disease relationships, T-T-1 denotes target interaction relationships, T-T-2 denotes similarity based on target-disease relationships.Table 1 Statistics for the drug target interaction dataset.Full size tableBaselines The baselines can be categorized into three categories: (1) Similarity-based methods, mainly including DTINet12, deepDTnet26 and NEDTP4; (2) KG-based models which use TransE27, ComplEx18 and KGE_NFM19; (3) Graph neural network (GNN)-based methods, such as DTI-MGNN8, SGCL-DTI14, MHGNN20 and IMCHGAN15.Parameter Configuration The hidden node embedding dimension of our models is set to 256 and the output embedding dimension is set to 64.

所有这些数据都可以在DrugBank上找到。通过这些数据集的交叉，我们获得了总共201种药物，252个靶标以及它们之间的907种相互作用。我们使用其中的607个交互进行训练，300个交互进行测试。这些数据集的统计信息总结在表1中，其中D表示药物，T表示靶标，D-D-1表示药物化学结构相似性，D-D-2表示基于药物副作用关系的相似性，D-D-3表示基于药物-疾病关系的相似性，T-T-1表示靶标相互作用关系，T-T-2表示基于靶标-疾病关系的相似性。表1药物-靶标相互作用数据集的统计数据。全尺寸表基线基线基线可分为三类：（1）基于相似性的方法，主要包括DTINet12，deepDTnet26和NEDTP4；（2） 使用TransE27，ComplEx18和KGE\u NFM19的基于KG的模型；（3） 基于图神经网络（GNN）的方法，例如DTI-MGNN8，SGCL-DTI14，MHGNN20和IMCHGAN15。参数配置我们模型的隐藏节点嵌入维度设置为256，输出嵌入维度设置为64。

We stack two convolutional layers. For the model optimization, our model will be trained 200 epochs with a learning rate of 0.01 and a batch size of 256. For HAN and MAGNN, the dimension of the attent.

我们堆叠两个卷积层。对于模型优化，我们的模型将训练200个时期，学习率为0.01，批量大小为256。对于HAN和MAGNN来说，参与者的规模。

Data availability

数据可用性

The data on drug-target interactions underpinning this study were sourced from DrugBank (Version 5.1.8). For more details, please visit https://go.drugbank.com/.

支持这项研究的药物-靶标相互作用的数据来自DrugBank（版本5.1.8）。有关更多详细信息，请访问https://go.drugbank.com/.

ReferencesNosengo, N. et al. Can you teach old drugs new tricks? Nature 534, 314–316 (2016).Article

参考文献Nosengo，N。等人。你能教老药新把戏吗？自然534314-316（2016）。文章

ADS

广告

PubMed

PubMed

Google Scholar

谷歌学者

Ezzat, A., Zhao, P., Wu, M., Li, X.-L. & Kwoh, C.-K. Drug-target interaction prediction with graph regularized matrix factorization. IEEE/ACM Trans. Comput. Biol. Bioinform. 14, 646–656 (2016).Article

Ezzat，A.，Zhao，P.，Wu，M.，Li，X.-L.＆Kwoh，C.-K.使用图正则化矩阵分解进行药物-靶标相互作用预测。IEEE/ACM Trans。计算机。生物。生物信息。14646-656（2016）。文章

PubMed

PubMed

Google Scholar

谷歌学者

Ding, Y., Tang, J. & Guo, F. Identification of drug-target interactions via dual Laplacian regularized least squares with multiple kernel fusion. Knowl.-Based Syst. 204, 106254 (2020).Article

Ding，Y.，Tang，J。＆Guo，F。通过具有多核融合的双拉普拉斯正则化最小二乘法鉴定药物-靶标相互作用。诺尔-。204106254（2020）。文章

Google Scholar

谷歌学者

An, Q. & Yu, L. A heterogeneous network embedding framework for predicting similarity-based drug-target interactions. Brief. Bioinform. 22, bbab275 (2021).Article

An，Q。＆Yu，L。一种用于预测基于相似性的药物-靶标相互作用的异质网络嵌入框架。简介。生物信息。22，bbab275（2021）。文章

PubMed

PubMed

Google Scholar

谷歌学者

Gao, K. Y. et al. Interpretable drug target prediction using deep neural representation. In IJCAI Vol. 2018, 3371–3377 (2018).Huang, K., Xiao, C., Glass, L. M. & Sun, J. Moltrans: Molecular interaction transformer for drug-target interaction prediction. Bioinformatics 37, 830–836 (2021).Article .

Gao，K.Y.等人。使用深度神经表示的可解释药物靶标预测。在IJCAI第2018卷中，3371-3377（2018）。Huang，K.，Xiao，C.，Glass，L.M。和Sun，J.Moltrans：用于药物-靶标相互作用预测的分子相互作用变压器。生物信息学37830-836（2021）。。

CAS

中科院

PubMed

PubMed

Google Scholar

谷歌学者

Chu, Y. et al. DTI-MLCD: Predicting drug-target interactions using multi-label learning with community detection method. Brief. Bioinform. 22, bbaa205 (2021).Article

Chu，Y。等人。DTI-MLCD：使用多标记学习和社区检测方法预测药物-靶标相互作用。简介。生物信息。22，bbaa205（2021）。文章

PubMed

PubMed

Google Scholar

谷歌学者

Liu, B., Pliakos, K., Vens, C. & Tsoumakas, G. Drug-target interaction prediction via an ensemble of weighted nearest neighbors with interaction recovery. Appl. Intell. 52, 1–23 (2022).

Liu，B.，Plakos，K.，Vens，C。＆Tsoumakas，G。通过加权最近邻与相互作用恢复的集合进行药物-靶标相互作用预测。应用。因特尔。52，1-23（2022）。

Google Scholar

谷歌学者

Pliakos, K., Vens, C. & Tsoumakas, G. Predicting drug-target interactions with multi-label classification and label partitioning. IEEE/ACM Trans. Comput. Biol. Bioinform. 18, 1596–1607 (2019).Article

Plakos，K.，Vens，C。＆Tsoumakas，G。通过多标签分类和标签分配预测药物-靶标相互作用。IEEE/ACM Trans。计算机。生物。生物信息。。文章

Google Scholar

谷歌学者

Olayan, R. S., Ashoor, H. & Bajic, V. B. DDR: Efficient computational method to predict drug-target interactions using graph mining and machine learning approaches. Bioinformatics 34, 1164–1173 (2018).Article

Olayan，R.S.，Ashoor，H。＆Bajic，V.B。DDR：使用图挖掘和机器学习方法预测药物-靶标相互作用的有效计算方法。生物信息学341164-1173（2018）。文章

CAS

中科院

PubMed

PubMed

Google Scholar

谷歌学者

Zhang, Z.-C. et al. A graph regularized generalized matrix factorization model for predicting links in biomedical bipartite networks. Bioinformatics 36, 3474–3481 (2020).Article

Zhang，Z.-C.等人。用于预测生物医学二分网络中链接的图正则化广义矩阵分解模型。生物信息学363474-3481（2020）。文章

CAS

中科院

PubMed

PubMed

Google Scholar

谷歌学者

Luo, Y. et al. A network integration approach for drug-target interaction prediction and computational drug repositioning from heterogeneous information. Nat. Commun. 8, 573 (2017).Article

Luo，Y。等人。一种用于药物-靶标相互作用预测和计算药物从异质信息重新定位的网络集成方法。。8573（2017）。文章

ADS

广告

PubMed

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Wan, F., Hong, L., Xiao, A., Jiang, T. & Zeng, J. NeoDTI: Neural integration of neighbor information from a heterogeneous network for discovering new drug-target interactions. Bioinformatics 35, 104–111 (2019).Article

Wan，F.，Hong，L.，Xiao，A.，Jiang，T。＆Zeng，J。NeoDTI：来自异构网络的邻居信息的神经整合，用于发现新的药物-靶标相互作用。生物信息学35104-111（2019）。文章

CAS

中科院

PubMed

PubMed

Google Scholar

谷歌学者

Li, Y., Qiao, G., Gao, X. & Wang, G. Supervised graph co-contrastive learning for drug-target interaction prediction. Bioinformatics 38, 2847–2854 (2022).Article

Li，Y.，Qiao，G.，Gao，X。＆Wang，G。用于药物-靶标相互作用预测的监督图对比学习。生物信息学382847-2854（2022）。文章

CAS

中科院

PubMed

PubMed

Google Scholar

谷歌学者

Li, J., Wang, J., Lv, H., Zhang, Z. & Wang, Z. IMCHGAN: Inductive matrix completion with heterogeneous graph attention networks for drug-target interactions prediction. IEEE/ACM Trans. Comput. Biol. Bioinform. 19, 655–665 (2021).Article

Li，J.，Wang，J.，Lv，H.，Zhang，Z。＆Wang，Z。IMCHGAN：用于药物-靶标相互作用预测的异质图注意网络的归纳矩阵完成。IEEE/ACM Trans。计算机。生物。生物信息。19655-665（2021）。文章

Google Scholar

谷歌学者

Adasme, M. F., Parisi, D., Sveshnikova, A. & Schroeder, M. Structure-based drug repositioning: Potential and limits. In Seminars in cancer biology Vol. 68 192–198 (Elsevier, 2021).

Adasme，M.F.，Parisi，D.，Sveshnikova，A。＆Schroeder，M。基于结构的药物重新定位：潜力和局限性。在癌症生物学研讨会第68卷192-198（Elsevier，2021）中。

Google Scholar

谷歌学者

Mohamed, S. K., Nounu, A. & Nováček, V. Drug target discovery using knowledge graph embeddings. In Proceedings of the 34th ACM/SIGAPP Symposium on Applied Computing, 11–18 (2019).Trouillon, T., Welbl, J., Riedel, S., Gaussier, É. & Bouchard, G. Complex embeddings for simple link prediction.

Mohamed，S.K.，Nounu，A。＆Nováček，V。使用知识图嵌入发现药物靶标。在第34届ACM/SIGAPP应用计算研讨会论文集，11-18（2019）。特鲁伊隆（T.Trouillon），韦尔布尔（J.Welbl），里德尔（S.Riedel），高斯（Gaussier）Bouchard，G。用于简单链接预测的复杂嵌入。

In International Conference on Machine Learning, 2071–2080 (PMLR, 2016).Ye, Q. et al. A unified drug-target interaction prediction framework based on knowledge graph and recommendation system. Nat. Commun. 12, 6775 (2021).Article .

在2071-2080年国际机器学习会议上（PMLR，2016）。Ye，Q。等人。基于知识图和推荐系统的统一药物-靶标相互作用预测框架。。126775（2021）。。

ADS

广告

CAS

中科院

PubMed

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Li, M., Cai, X., Xu, S. & Ji, H. Metapath-aggregated heterogeneous graph neural network for drug-target interaction prediction. Brief. Bioinform. 24, bbac578 (2023).Article

Li，M.，Cai，X.，Xu，S。＆Ji，H。Metapath聚合异质图神经网络用于药物-靶标相互作用预测。简介。生物信息。24，bbac578（2023）。文章

PubMed

PubMed

Google Scholar

谷歌学者

Fan, Y. et al. SGCLDGA: Unveiling drug-gene associations through simple graph contrastive learning. Brief. Bioinform. 25, bbae231 (2024).Article

Fan，Y。等。SGCLDGA：通过简单的图形对比学习揭示药物-基因关联。简介。生物信息。。文章

PubMed

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Cao, J. et al. NGCN: Drug-target interaction prediction by integrating information and feature learning from heterogeneous network. J. Cell Mol. Med. 28, e18224 (2024).Article

Cao，J。等人。NGCN：通过整合来自异构网络的信息和特征学习来预测药物-靶标相互作用。J。Cell Mol。Med。28，e18224（2024）。文章

PubMed

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Dehghan, A., Razzaghi, P., Abbasi, K. & Gharaghani, S. TripletMultiDTI: Multimodal representation learning in drug-target interaction prediction with triplet loss function. Expert Syst. Appl. 232, 120754 (2023).Article

Dehghan，A.，Razzaghi，P.，Abbasi，K。＆Gharaghani，S。TripletMultiDTI：具有三重态损失功能的药物-靶标相互作用预测中的多模式表示学习。专家系统。应用。232120754（2023）。文章

Google Scholar

谷歌学者

Rafiei, F. et al. DeepTraSynergy: Drug combinations using multimodal deep learning with transformers. Bioinformatics 39, btad438 (2023).Article

Rafiei，F。等人，《DeepTraSynergy：使用变压器的多模式深度学习的药物组合》。生物信息学39，btad438（2023）。文章

CAS

中科院

PubMed

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Gharizadeh, A., Abbasi, K., Ghareyazi, A., Mofrad, M. R. & Rabiee, H. R. HGTDR: Advancing drug repurposing with heterogeneous graph transformers. Preprint at arXiv:2405.08031 (2024).Zeng, X. et al. Target identification among known drugs by deep learning from heterogeneous networks. Chem.

Gharizadeh，A.，Abbasi，K.，Ghareyazi，A.，Mofrad，M.R。＆Rabiee，H.R。HGTDR：用异构图形转换器推进药物再利用。arXiv预印本：2405.08031（2024）。Zeng，X。等人。通过从异构网络中进行深度学习来识别已知药物中的靶标。化学。

Sci. 11, 1775–1797 (2020).Article .

科学。111775-1797（2020）。。

CAS

中科院

PubMed

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Bordes, A., Usunier, N., Garcia-Duran, A., Weston, J. & Yakhnenko, O. Translating embeddings for modeling multi-relational data. Adv. Neural Inf. Processi. Syst. 26 (2013).Download referencesAuthor informationAuthors and AffiliationsCollege of Science, China Agricultural University, Beijing, 100083, ChinaQi WangSDU-ANU Joint Science College, Shandong University, Weihai, 264209, Shandong, ChinaQianwen LuAcademy of Mathematics and Systems Science, Chinese Academy of Sciences, Beijing, 100190, ChinaZhiheng ZhouSchool of Mathematical Sciences, University of Chinese Academy of Sciences, Beijing, 100190, ChinaZhiheng ZhouAuthorsQianwen LuView author publicationsYou can also search for this author in.

Bordes，A.，Usunier，N.，Garcia-Duran，A.，Weston，J。＆Yakhnenko，O。翻译嵌入以建模多关系数据。高级神经信息处理。系统。26（2013）。下载参考文献作者信息作者和所属机构中国农业大学理学院，北京，100083，中国齐王SDU ANU联合科学学院，山东大学威海，264209，山东，中国前文鲁中国科学院数学与系统科学研究院，北京，100190，中国周志恒中国科学院大学数学科学院，北京，100190，中国周志恒作者前文鲁维作者出版物你也可以在中搜索这位作者。

PubMed Google ScholarZhiheng ZhouView author publicationsYou can also search for this author in

PubMed谷歌学者Zhiheng ZhouView作者出版物您也可以在

PubMed Google ScholarQi WangView author publicationsYou can also search for this author in

PubMed Google ScholarQi WangView作者出版物您也可以在

PubMed Google ScholarContributionsQ.L. conceived the project, developed the prediction method, designed the experiments, implemented the experiments, and wrote the paper. Z.Z. conceived the project, designed the experiments, analyzed the result, and wrote the paper. Q.W. conceived the project, analyzed the result, and revised the paper.

PubMed谷歌学术贡献。五十、 构思了该项目，开发了预测方法，设计了实验，实施了实验，并撰写了论文。Z、 Z.构思了该项目，设计了实验，分析了结果，并撰写了论文。Q、 W.构思了这个项目，分析了结果，并修改了论文。

All authors have reviewed and approved the manuscript.Corresponding authorCorrespondence to.

所有作者都审阅并批准了手稿。对应作者对应。

Qi Wang.Ethics declarations

齐旺。道德宣言

Competing interests

相互竞争的利益

The authors declare no competing interests.

作者声明没有利益冲突。

Additional informationPublisher’s noteSpringer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.Rights and permissions

Additional informationPublisher的noteSpringer Nature在已发布地图和机构隶属关系中的管辖权主张方面保持中立。权限和权限

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material.

开放获取本文是根据知识共享署名非商业性NoDerivatives 4.0国际许可证授权的，该许可证允许以任何媒介或格式进行任何非商业性使用，共享，分发和复制，只要您对原始作者和来源给予适当的信任，提供知识共享许可证的链接，并指出您是否修改了许可材料。

You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

根据本许可证，您无权共享源自本文或其部分的改编材料。本文中的图像或其他第三方材料包含在文章的知识共享许可证中，除非该材料的信用额度中另有说明。如果材料未包含在文章的知识共享许可中，并且您的预期用途不受法律法规的许可或超出许可用途，则您需要直接获得版权所有者的许可。

To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/..

要查看此许可证的副本，请访问http://creativecommons.org/licenses/by-nc-nd/4.0/..

Reprints and permissionsAbout this articleCite this articleLu, Q., Zhou, Z. & Wang, Q. Multi-layer graph attention neural networks for accurate drug-target interaction mapping.

转载和许可本文引用本文Lu，Q.，Zhou，Z。＆Wang，Q。多层图注意神经网络用于精确的药物-靶标相互作用映射。

Sci Rep 14, 26119 (2024). https://doi.org/10.1038/s41598-024-75742-1Download citationReceived: 02 February 2024Accepted: 08 October 2024Published: 30 October 2024DOI: https://doi.org/10.1038/s41598-024-75742-1Share this articleAnyone you share the following link with will be able to read this content:Get shareable linkSorry, a shareable link is not currently available for this article.Copy to clipboard.

科学报告1426119（2024）。https://doi.org/10.1038/s41598-024-75742-1Download引文接收日期：2024年2月2日接受日期：2024年10月8日发布日期：2024年10月30日OI：https://doi.org/10.1038/s41598-024-75742-1Share本文与您共享以下链接的任何人都可以阅读此内容：获取可共享链接对不起，本文目前没有可共享的链接。复制到剪贴板。

Provided by the Springer Nature SharedIt content-sharing initiative

由Springer Nature SharedIt内容共享计划提供

Subjects

主题

Computational modelsData miningMachine learningNetwork topology

计算模型数据挖掘机器学习网络拓扑