MiRNA-Disease Associations Prediction based on Network Consistency Projection

Abstract

MicroRNAs (miRNAs) play an important role in the prevention, diagnosis, and treatment of human complex diseases. More and more experimental validated associations between miRNAs and diseases have been reported in recent studies, which provide useful information for new miRNA-disease association discovery. But through experiments, we have to overcome problems such as the inefficiency of methods, the need for a lot of manpower, materials, and finance… Therefore, reliable computational models are expected to be an effective supplement for inferring associations between miRNAs and diseases. In this paper, we propose a computational method based on network consistency projection to predict potential human miRNA-disease associations (MDA). This method enriches biological information and reduces prediction bias. Besides, it is not only a parameterless method but also does not require a negative sample. More importantly, it can predict miRNA without any known associated diseases. This method’s AUC value of 5-fold cross-validation (5-fold-CV) is also compared to five methods and it is shown that the AUC value of this method achieves the highest value.