This web server provides a method to infer potential interaction between microRNA and disease for further confirmation using biological experiments. First, we use MeSH tree structure and disease related genes to measure the disease similarity, and the subnetwork of microRNA similarity is constructed using the microRNA–target gene and microRNA-long non-coding RNA associations. A heterogeneous network was constructed by connecting the disease similarity subnetwork and the microRNA similarity subnetwork using the experimentally verified microRNA–disease associations. We extend the framework of random walk with restart to predict microRNA–disease associations in the heterogeneous network.
The area under the curve (AUC) is 0.8049.
For 15 human diseases, an AUC from 0.7970 to 0.9249.
Application to disease with no known related microRNAs
Cases studies: lung neoplasms and breast neoplasms
|HMDD||human microRNA-disease database||http://www.cuilab.cn/hmdd|
|MeSH||Medical Subject Headings||http://www.ncbi.nlm.nih.gov/mesh|
|DisGeNET||a database of gene-disease associations||http://www.disgenet.org/|
|miRTarBase||experimentally validated microRNA-target gene interactions||http://mirtarbase.mbc.nctu.edu.tw/|
|starBase v2.0||microRNA-lncRNA interactions||http://starbase.sysu.edu.cn/mirLncRNA.php|
|LncRNADisease||the experimentally supported lncRNA-disease association||http://www.cuilab.cn/lncrnadisease|
|HumanNet||a probabilistic functional gene network||http://www.functionalnet.org/humannet/|