Abstract:Abstract While the technologies of ribonucleic acid-sequence (RNA-seq) and transcript assembly analysis have continued to improve, a novel topology of RNA transcript was uncovered in the last decade and is called circular RNA (circRNA). Recently, researchers have revealed that they compete with messenger RNA (mRNA) and long noncoding for combining with microRNA in gene regulation. Therefore, circRNA was assumed to be associated with complex disease and discovering the relationship between them would contribute to medical research. However, the work of identifying the association between circRNA and disease in vitro takes a long time and usually without direction. During these years, more and more associations were verified by experiments. Hence, we proposed a computational method named identifying circRNA–disease association based on graph representation learning (iGRLCDA) for the prediction of the potential association of circRNA and disease, which utilized a deep learning model of graph convolution network (GCN) and graph factorization (GF). In detail, iGRLCDA first derived the hidden feature of known associations between circRNA and disease using the Gaussian interaction profile (GIP) kernel combined with disease semantic information to form a numeric descriptor. After that, it further used the deep learning model of GCN and GF to extract hidden features from the descriptor. Finally, the random forest classifier is introduced to identify the potential circRNA–disease association. The five-fold cross-validation of iGRLCDA shows strong competitiveness in comparison with other excellent prediction models at the gold standard data and achieved an average area under the receiver operating characteristic curve of 0.9289 and an area under the precision-recall curve of 0.9377. On reviewing the prediction results from the relevant literature, 22 of the top 30 predicted circRNA–disease associations were noted in recent published papers. These exceptional results make us believe that iGRLCDA can provide reliable circRNA–disease associations for medical research and reduce the blindness of wet-lab experiments.

Exploring potential circRNA biomarkers for cancers based on double-line heterogeneous graph representation learning

GSLCDA: An Unsupervised Deep Graph Structure Learning Method for Predicting CircRNA-Disease Association

iGRLCDA: identifying circRNA–disease association based on graph representation learning

MGRCDA: Metagraph Recommendation Method for Predicting CircRNA-Disease Association

Predicting CircRNA-Disease associations via feature convolution learning with heterogeneous graph attention network

DRGCNCDA: Predicting circRNA-disease interactions based on knowledge graph and disentangled relational graph convolutional network

GCNCDA: A new method for predicting circRNA-disease associations based on Graph Convolutional Network Algorithm

RGCNCDA: Relational graph convolutional network improves circRNA-disease association prediction by incorporating microRNAs

HGECDA: A Heterogeneous Graph Embedding Model for CircRNA-Disease Association Prediction

An in-silico method with graph-based multi-label learning for large-scale prediction of circRNA-disease associations

Identifying circRNA-disease association based on relational graph attention network and hypergraph attention network

GGAECDA: Predicting circRNA-disease associations using graph autoencoder based on graph representation learning

RDGAN: Prediction of circRNA-Disease Associations via Resistance Distance and Graph Attention Network

A multi-task prediction method based on neighborhood structure embedding and signed graph representation learning to infer the relationship between circRNA, miRNA, and cancer

Graph Convolution Networks Using Message Passing and Multi-Source Similarity Features for Predicting circRNA-Disease Association

MAGCDA: A Multi-Hop Attention Graph Neural Networks Method for CircRNA-Disease Association Prediction

A computational model of circRNA-associated diseases based on a graph neural network: prediction and case studies for follow-up experimental validation

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

LGCDA: Predicting CircRNA-Disease Association Based on Fusion of Local and Global Features

SGFCCDA: Scale Graph Convolutional Networks and Feature Convolution for circRNA-Disease Association Prediction

Integrative Graph-Based Framework for Predicting circRNA Drug Resistance Using Disease Contextualization and Deep Learning