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

•The circRNA-disease knowledge graph is constructed by integrating different kinds of biological information.•We present a computational method based on disentangled relational graph convolutional network and tensor factorization to predict potential circRNA-disease associations. The disentangled mechanism is used to capture the hidden factor information.•The experimental results show that the proposed model achieves impressive performance and can effectively predict potential disease related circRNAs.