Skip to main content
Fig. 7 | BMC Bioinformatics

Fig. 7

From: DeepciRGO: functional prediction of circular RNAs through hierarchical deep neural networks using heterogeneous network features

Fig. 7

Flowchart of DeepciRGO, which includes three steps: (a) build the global heterogeneous network by integrating three networks (circRNA co-expression network, circRNA–protein interaction network, and PPI network); (b) employ HIN2Vec to learn the latent representations of the nodes (circRNAs and proteins); (c) train each GO class with the neural network model and annotate circRNAs

Back to article page