Heterogeneous Graph Convolutional Networks and Matrix Completion for miRNA-Disease Association Prediction

Abstract

Due to the cost and complexity of biological experiments, many computational methods have been proposed to predict potential miRNA-disease associations by utilizing known miRNA-disease associations and other related information. However, there are some challenges for these computational methods. First, the relationships between miRNAs and diseases are complex. The computational network should consider the local and global influence of neighborhoods from the network. Furthermore, predicting disease-related miRNAs without any known associations is also very important. This study presents a new computational method that constructs a heterogeneous network composed of a miRNA similarity network, disease similarity network, and known miRNA-disease association network. The miRNA similarity considers the miRNAs and their possible families and clusters. The information of each node in heterogeneous network is obtained by aggregating neighborhood information with graph convolutional networks (GCNs), which can pass the information of a node to its intermediate and distant neighbors. Disease-related miRNAs with no known associations can be predicted with the reconstructed heterogeneous matrix. We apply 5-fold cross-validation, leave-one-disease-out cross-validation, and global and local leave-one-out cross-validation to evaluate our method. The corresponding areas under the curves (AUCs) are 0.9616, 0.9946, 0.9656, and 0.9532, confirming that our approach significantly outperforms the state-of-the-art methods. Case studies show that this approach can effectively predict new diseases without any known miRNAs.

DGAMDA: Predicting miRNA‐disease association based on dynamic graph attention network

“…8 Considering neighborhoods' local and global influence from the interaction network, Zhu et al offer a method that uses a graph convolution network to extract the node information and matrix completion to predict the association score. 9 Chen et al have proposed a novel method, named MDSCMF, which combines matrix decomposition and similarityconstrained matrix factorization. The proposed model decomposes the miRNA-disease association matrix into low-rank matrices and then applies the similarity constraints on them to further improve the performance of prediction.…”

“…The model uses dual Laplacian regularization to better exploit the available information 8 . Considering neighborhoods' local and global influence from the interaction network, Zhu et al offer a method that uses a graph convolution network to extract the node information and matrix completion to predict the association score 9 . Chen et al have proposed a novel method, named MDSCMF, which combines matrix decomposition and similarity‐constrained matrix factorization.…”

LRGCPND: Predicting Associations between ncRNA and Drug Resistance via Linear Residual Graph Convolution

“…GCMDR [ 24 ] established a three-layer latent factor model to predict miRNA-disease associations introducing features such as miRNA expression profile and drug PubChem substructure fingerprints into the model. Zhu et al [ 25 ] utilized the matrix completion method. SDLDA [ 26 ] introduced singular value decomposition and ILNCRNADIS-FB [ 27 ] calculated the three-dimensional feature blocks to capture characteristics.…”

Network Analyses Based on Machine Learning Methods to Quantify Effects of Peptide–Protein Complexes as Drug Targets Using Cinnamon in Cardiovascular Diseases and Metabolic Syndrome as a Case Study

“…Due to the characters of peptide-protein interactions, it is reasonable to perform network analyses based on machine learning methods since the relationships between those peptides and proteins could be illustrated clearly in the form of graphs (Zhu et al, 2020;Ji et al, 2021;Yingying et al, 2021). Similarly, some complex diseases are found to be similar based on network analyses, indicating that more relationships between different diseases could be predicted using bioinformatics pipelines (Wang et al, 2019).…”