PmiRGO is a computational method for microRNA Gene Ontology function prediction. It builds a global heterogeneous network consisting of a microRNA-microRNA co-expression network, a microRNA-target gene association network and a protein-protein interaction network. DeepWalk is employed to learn latent representations as network features. SVM classifiers models are built to predict microRNA Gene Ontology function based on the network features.

The source code and data of PmiRGO are available at: download