BCMCMI: A Fusion Model for Predicting circRNA-miRNA Interactions Combining Semantic and Meta-path

Meng-Meng Wei,Chang-Qing Yu,Li-Ping Li,Zhu-Hong You,Lei Wang

DOI: https://doi.org/10.1021/acs.jcim.3c00852

IF: 6.162

2023-08-03

Journal of Chemical Information and Modeling

Abstract:More and more evidence suggests that circRNA plays a vital role in generating and treating diseases by interacting with miRNA. Therefore, accurate prediction of potential circRNA-miRNA interaction (CMI) has become urgent. However, traditional wet experiments are time-consuming and costly, and the results will be affected by objective factors. In this paper, we propose a computational model BCMCMI, which combines three features to predict CMI. Specifically, BCMCMI utilizes the bidirectional encoding capability of the BERT algorithm to extract sequence features from the semantic information of circRNA and miRNA. Then, a heterogeneous network is constructed based on cosine similarity and known CMI information. The Metapath2vec is employed to conduct random walks following meta-paths in the network to capture topological features, including similarity features. Finally, potential CMIs are predicted using the XGBoost classifier. BCMCMI achieves superior results compared to other state-of-the-art models on two benchmark datasets for CMI prediction. We also utilize t-SNE to visually observe the distribution of the extracted features on a randomly selected dataset. The remarkable prediction results show that BCMCMI can serve as a valuable complement to the wet experiment process.

chemistry, multidisciplinary, medicinal,computer science, interdisciplinary applications, information systems