WVMDA: Predicting miRNA–Disease Association Based on Weighted Voting

An increasing number of experiments have verified that miRNA expression is associated with human diseases. The miRNA expression profile may be an indicator of clinical diagnosis and provides a new direction for the prevention and treatment of complex diseases. In this work, we present a weighted vote-based model to predict miRNA-disease association (WVMDA). In order to reasonably construct similarity network, we established credibility similarity based on the reliability of known associations and used it to improve the incomplete of original similarity. In order to eliminate noise interference as much as possible while retaining more reliable similarity information, we designed a filter. Most importantly, in order to ensure the fairness and effectiveness of weighted voting, we focus on the design of voting weight. Finally, cross-validation experiments and case studies are conducted to verify the effectiveness of the proposed model. The results showed that WVMDA could effectively identify miRNAs associated with disease.