Graph Convolutional Network and Convolutional Neural Network Based Method for Predicting lncRNA-Disease Associations
Aberrant expressions of long non-coding RNAs (lncRNAs) are often associated with diseases and identification of disease-related lncRNAs is helpful for elucidating complex pathogenesis. Recent methods for predicting associations between lncRNAs and diseases integrate their pertinent heterogeneous dat...
| Glavni autori: | , , , , |
|---|---|
| Format: | Članak |
| Jezik: | English |
| Izdano: |
MDPI AG
2019-08-01
|
| Serija: | Cells |
| Teme: | |
| Online pristup: | https://www.mdpi.com/2073-4409/8/9/1012 |
| _version_ | 1827885825501691904 |
|---|---|
| author | Ping Xuan Shuxiang Pan Tiangang Zhang Yong Liu Hao Sun |
| author_facet | Ping Xuan Shuxiang Pan Tiangang Zhang Yong Liu Hao Sun |
| author_sort | Ping Xuan |
| collection | DOAJ |
| description | Aberrant expressions of long non-coding RNAs (lncRNAs) are often associated with diseases and identification of disease-related lncRNAs is helpful for elucidating complex pathogenesis. Recent methods for predicting associations between lncRNAs and diseases integrate their pertinent heterogeneous data. However, they failed to deeply integrate topological information of heterogeneous network comprising lncRNAs, diseases, and miRNAs. We proposed a novel method based on the graph convolutional network and convolutional neural network, referred to as GCNLDA, to infer disease-related lncRNA candidates. The heterogeneous network containing the lncRNA, disease, and miRNA nodes, is constructed firstly. The embedding matrix of a lncRNA-disease node pair was constructed according to various biological premises about lncRNAs, diseases, and miRNAs. A new framework based on a graph convolutional network and a convolutional neural network was developed to learn network and local representations of the lncRNA-disease pair. On the left side of the framework, the autoencoder based on graph convolution deeply integrated topological information within the heterogeneous lncRNA-disease-miRNA network. Moreover, as different node features have discriminative contributions to the association prediction, an attention mechanism at node feature level is constructed. The left side learnt the network representation of the lncRNA-disease pair. The convolutional neural networks on the right side of the framework learnt the local representation of the lncRNA-disease pair by focusing on the similarities, associations, and interactions that are only related to the pair. Compared to several state-of-the-art prediction methods, GCNLDA had superior performance. Case studies on stomach cancer, osteosarcoma, and lung cancer confirmed that GCNLDA effectively discovers the potential lncRNA-disease associations. |
| first_indexed | 2024-03-12T19:53:18Z |
| format | Article |
| id | doaj.art-bc07ccd052cf4076859665811da6f359 |
| institution | Directory Open Access Journal |
| issn | 2073-4409 |
| language | English |
| last_indexed | 2024-03-12T19:53:18Z |
| publishDate | 2019-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Cells |
| spelling | doaj.art-bc07ccd052cf4076859665811da6f3592023-08-02T03:02:53ZengMDPI AGCells2073-44092019-08-0189101210.3390/cells8091012cells8091012Graph Convolutional Network and Convolutional Neural Network Based Method for Predicting lncRNA-Disease AssociationsPing Xuan0Shuxiang Pan1Tiangang Zhang2Yong Liu3Hao Sun4School of Computer Science and Technology, Heilongjiang University, Harbin 150080, ChinaSchool of Computer Science and Technology, Heilongjiang University, Harbin 150080, ChinaSchool of Mathematical Science, Heilongjiang University, Harbin 150080, ChinaSchool of Computer Science and Technology, Heilongjiang University, Harbin 150080, ChinaSchool of Computer Science and Technology, Heilongjiang University, Harbin 150080, ChinaAberrant expressions of long non-coding RNAs (lncRNAs) are often associated with diseases and identification of disease-related lncRNAs is helpful for elucidating complex pathogenesis. Recent methods for predicting associations between lncRNAs and diseases integrate their pertinent heterogeneous data. However, they failed to deeply integrate topological information of heterogeneous network comprising lncRNAs, diseases, and miRNAs. We proposed a novel method based on the graph convolutional network and convolutional neural network, referred to as GCNLDA, to infer disease-related lncRNA candidates. The heterogeneous network containing the lncRNA, disease, and miRNA nodes, is constructed firstly. The embedding matrix of a lncRNA-disease node pair was constructed according to various biological premises about lncRNAs, diseases, and miRNAs. A new framework based on a graph convolutional network and a convolutional neural network was developed to learn network and local representations of the lncRNA-disease pair. On the left side of the framework, the autoencoder based on graph convolution deeply integrated topological information within the heterogeneous lncRNA-disease-miRNA network. Moreover, as different node features have discriminative contributions to the association prediction, an attention mechanism at node feature level is constructed. The left side learnt the network representation of the lncRNA-disease pair. The convolutional neural networks on the right side of the framework learnt the local representation of the lncRNA-disease pair by focusing on the similarities, associations, and interactions that are only related to the pair. Compared to several state-of-the-art prediction methods, GCNLDA had superior performance. Case studies on stomach cancer, osteosarcoma, and lung cancer confirmed that GCNLDA effectively discovers the potential lncRNA-disease associations.https://www.mdpi.com/2073-4409/8/9/1012graph convolutional networkconvolutional neural networklncRNA-disease association predictionattention mechanism at node feature level |
| spellingShingle | Ping Xuan Shuxiang Pan Tiangang Zhang Yong Liu Hao Sun Graph Convolutional Network and Convolutional Neural Network Based Method for Predicting lncRNA-Disease Associations Cells graph convolutional network convolutional neural network lncRNA-disease association prediction attention mechanism at node feature level |
| title | Graph Convolutional Network and Convolutional Neural Network Based Method for Predicting lncRNA-Disease Associations |
| title_full | Graph Convolutional Network and Convolutional Neural Network Based Method for Predicting lncRNA-Disease Associations |
| title_fullStr | Graph Convolutional Network and Convolutional Neural Network Based Method for Predicting lncRNA-Disease Associations |
| title_full_unstemmed | Graph Convolutional Network and Convolutional Neural Network Based Method for Predicting lncRNA-Disease Associations |
| title_short | Graph Convolutional Network and Convolutional Neural Network Based Method for Predicting lncRNA-Disease Associations |
| title_sort | graph convolutional network and convolutional neural network based method for predicting lncrna disease associations |
| topic | graph convolutional network convolutional neural network lncRNA-disease association prediction attention mechanism at node feature level |
| url | https://www.mdpi.com/2073-4409/8/9/1012 |
| work_keys_str_mv | AT pingxuan graphconvolutionalnetworkandconvolutionalneuralnetworkbasedmethodforpredictinglncrnadiseaseassociations AT shuxiangpan graphconvolutionalnetworkandconvolutionalneuralnetworkbasedmethodforpredictinglncrnadiseaseassociations AT tiangangzhang graphconvolutionalnetworkandconvolutionalneuralnetworkbasedmethodforpredictinglncrnadiseaseassociations AT yongliu graphconvolutionalnetworkandconvolutionalneuralnetworkbasedmethodforpredictinglncrnadiseaseassociations AT haosun graphconvolutionalnetworkandconvolutionalneuralnetworkbasedmethodforpredictinglncrnadiseaseassociations |