Prediction of circRNA-disease associations based on multiple biological data

Abstract

Circular RNA (circRNA) plays a significant role in gene expression, splicing, and transcription. More and more evidence indicates that circRNA is related to the pathogenesis and development of diseases. In this paper, a non-negative matrix factorization algorithm based on circRNA expression profiles data and disease similarity data (EDNMF) is proposed to predict circRNA-disease associations. The EDNMF algorithm firstly preprocesses the circRNA-disease associations to solve the impact of too little the number of known circRNA-disease associations. Then, the EDNMF algorithm converts circRNA expression profile and cancer similarity data into constraints. Finally, we can obtain the final scores for circRNA-disease associations by improved NMF algorithm based on pre-processed circRNA-disease associations. The performance results of 5-fold and 10-fold cross-validation indicate that the EDNMF algorithm achieves satisfactory performance comparing with other algorithms. Besides, the case study shows that EDNMF can mine new circRNA-disease associations very well, which can provide a reference for studying circRNA-disease associations.