AUTHOR=Gao Yuanxu , Jia Kaiwen , Shi Jiangcheng , Zhou Yuan , Cui Qinghua 

TITLE=A Computational Model to Predict the Causal miRNAs for Diseases

JOURNAL=Frontiers in Genetics

VOLUME=Volume 10 - 2019

YEAR=2019

URL=https://www.frontiersin.org/journals/genetics/articles/10.3389/fgene.2019.00935

DOI=10.3389/fgene.2019.00935

ISSN=1664-8021

ABSTRACT=MicroRNAs (miRNAs) are one class of important noncoding RNA molecules and their dysfunction is associated with a number of diseases. Currently, a series of databases and algorithms have been developed for dissecting human miRNA-disease associations. However, these tools only presented the associations between miRNAs and disease but did not address whether the associations are causal or not, a key biomedical issue which is critical for understanding the roles of candidate miRNAs in the mechanisms of specific diseases. Here we first manually curated causal miRNA-disease association information and updated the human miRNA disease database (HMDD) accordingly. Then we built a computational model, MDCAP, to predict novel causal miRNA-disease associations. As a result, we collected 6,667 causal miRNA-disease associations between 616 miRNAs and 440 diseases, which accounts for ~20% of the total data in HMDD. The MDCAP model achieved an AUC of 0.928 for ROC analysis by independent test and an AUC of 0.925 for ROC analysis by 10-fold cross validation. Finally, case studies conducted on myocardial infarction (MI) and hsa-mir-498 further suggested the biomedical significance of the predictions.