GATCDA: Predicting circRNA-Disease Associations Based on Graph Attention Network

Bian, Chen; Lei, Xiujuan; Wu, Fang‐Xiang

doi:10.3390/cancers13112595

Cited by 23 publications

(22 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Graph attention network (GAT) is an effective graph representation learning tool, which represents the importance of neighbor nodes to the target node by assigning different weights to different neighbor nodes ( Bian et al, 2021 ). Here, for target node i and metapath M related to i , we first use GAT to assign weights (attention coefficients) to metapath instances in M , thereby learning the importance of different metapath instances to target nodes.…”

Section: Methodsmentioning

confidence: 99%

Metapath Aggregated Graph Neural Network and Tripartite Heterogeneous Networks for Microbe-Disease Prediction

Chen

Lei

2022

Front. Microbiol.

View full text Add to dashboard Cite

More and more studies have shown that understanding microbe-disease associations cannot only reveal the pathogenesis of diseases, but also promote the diagnosis and prognosis of diseases. Because traditional medical experiments are time-consuming and expensive, many computational methods have been proposed in recent years to identify potential microbe-disease associations. In this study, we propose a method based on heterogeneous network and metapath aggregated graph neural network (MAGNN) to predict microbe-disease associations, called MATHNMDA. First, we introduce microbe-drug interactions, drug-disease associations, and microbe-disease associations to construct a microbe-drug-disease heterogeneous network. Then we take the heterogeneous network as input to MAGNN. Second, for each layer of MAGNN, we carry out intra-metapath aggregation with a multi-head attention mechanism to learn the structural and semantic information embedded in the target node context, the metapath-based neighbor nodes, and the context between them, by encoding the metapath instances under the metapath definition mode. We then use inter-metapath aggregation with an attention mechanism to combine the semantic information of all different metapaths. Third, we can get the final embedding of microbe nodes and disease nodes based on the output of the last layer in the MAGNN. Finally, we predict potential microbe-disease associations by reconstructing the microbe-disease association matrix. In addition, we evaluated the performance of MATHNMDA by comparing it with that of its variants, some state-of-the-art methods, and different datasets. The results suggest that MATHNMDA is an effective prediction method. The case studies on asthma, inflammatory bowel disease (IBD), and coronavirus disease 2019 (COVID-19) further validate the effectiveness of MATHNMDA.

show abstract

Section: Methodsmentioning

confidence: 99%

Metapath Aggregated Graph Neural Network and Tripartite Heterogeneous Networks for Microbe-Disease Prediction

Chen

Lei

2022

Front. Microbiol.

View full text Add to dashboard Cite

show abstract

“…Since these methods adopt various datasets and evaluation metrics, we apply the same dataset and AUC as the metrics to compare the predictive capacity of models fairly and reasonably. In this part, the GATGCN is compared with several state-of-the-art methods, including KATZHCDA (Fan et al, 2018a), DWNN-RLS (Yan et al, 2018), PWCDA (Lei et al, 2018), GCNCDA (Wang et al, 2020b), and GATCDA (Bian et al, 2021). KATZHCDA is a graph-based method that uses the walking lengths and number of walks among nodes to measure the similarity among nodes in the heterogenous network.…”

Section: Comparison With Other Methodsmentioning

confidence: 99%

“…Wang et al designed an approach named GCNCDA to identify diseaserelated circRNAs, which extracts high-level features contained in the circRNA-disease heterogenous network through graph convolutional networks to calculate association scores (Wang et al, 2020b). GATCDA is a novel model for discovering the correlation between diseases and circRNAs, which learns the latent representation of nodes by assigning corresponding weights to each neighbor node (Bian et al, 2021). Xiao et al designed a computational model named NSL2CD that adopts network embedding by adaptive subspace learning (Xiao et al, 2021).…”

Section: Introductionmentioning

confidence: 99%

Using Graph Attention Network and Graph Convolutional Network to Explore Human CircRNA–Disease Associations Based on Multi-Source Data

Li,

Wang,

Zhang

et al. 2022

Front. Genet.

View full text Add to dashboard Cite

Cumulative research studies have verified that multiple circRNAs are closely associated with the pathogenic mechanism and cellular level. Exploring human circRNA–disease relationships is significant to decipher pathogenic mechanisms and provide treatment plans. At present, several computational models are designed to infer potential relationships between diseases and circRNAs. However, the majority of existing approaches could not effectively utilize the multisource data and achieve poor performance in sparse networks. In this study, we develop an advanced method, GATGCN, using graph attention network (GAT) and graph convolutional network (GCN) to detect potential circRNA–disease relationships. First, several sources of biomedical information are fused via the centered kernel alignment model (CKA), which calculates the corresponding weight of different kernels. Second, we adopt the graph attention network to learn latent representation of diseases and circRNAs. Third, the graph convolutional network is deployed to effectively extract features of associations by aggregating feature vectors of neighbors. Meanwhile, GATGCN achieves the prominent AUC of 0.951 under leave-one-out cross-validation and AUC of 0.932 under 5-fold cross-validation. Furthermore, case studies on lung cancer, diabetes retinopathy, and prostate cancer verify the reliability of GATGCN for detecting latent circRNA–disease pairs.

show abstract

“…As a result, it achieved an AUC value of 0.909 in fivefold CV based on circR2Disease dataset. Bian et al developed GATCDA method based on graph attention network to obtain representation of circRNAs and diseases, calculated the probability score by dot production [32], and yielded an AUC value of 0.9011.…”

Section: Introductionmentioning

confidence: 99%

GATNNCDA: A Method Based on Graph Attention Network and Multi-Layer Neural Network for Predicting circRNA-Disease Associations

Liu

Wang

et al. 2021

IJMS

View full text Add to dashboard Cite

Circular RNAs (circRNAs) are a new class of endogenous non-coding RNAs with covalent closed loop structure. Researchers have revealed that circRNAs play an important role in human diseases. As experimental identification of interactions between circRNA and disease is time-consuming and expensive, effective computational methods are an urgent need for predicting potential circRNA–disease associations. In this study, we proposed a novel computational method named GATNNCDA, which combines Graph Attention Network (GAT) and multi-layer neural network (NN) to infer disease-related circRNAs. Specially, GATNNCDA first integrates disease semantic similarity, circRNA functional similarity and the respective Gaussian Interaction Profile (GIP) kernel similarities. The integrated similarities are used as initial node features, and then GAT is applied for further feature extraction in the heterogeneous circRNA–disease graph. Finally, the NN-based classifier is introduced for prediction. The results of fivefold cross validation demonstrated that GATNNCDA achieved an average AUC of 0.9613 and AUPR of 0.9433 on the CircR2Disease dataset, and outperformed other state-of-the-art methods. In addition, case studies on breast cancer and hepatocellular carcinoma showed that 20 and 18 of the top 20 candidates were respectively confirmed in the validation datasets or published literature. Therefore, GATNNCDA is an effective and reliable tool for discovering circRNA–disease associations.

show abstract

GATCDA: Predicting circRNA-Disease Associations Based on Graph Attention Network

Cited by 23 publications

References 49 publications

Metapath Aggregated Graph Neural Network and Tripartite Heterogeneous Networks for Microbe-Disease Prediction

Metapath Aggregated Graph Neural Network and Tripartite Heterogeneous Networks for Microbe-Disease Prediction

Using Graph Attention Network and Graph Convolutional Network to Explore Human CircRNA–Disease Associations Based on Multi-Source Data

GATNNCDA: A Method Based on Graph Attention Network and Multi-Layer Neural Network for Predicting circRNA-Disease Associations

Contact Info

Product

Resources

About