Skip to main content
Fig. 3 | BMC Systems Biology

Fig. 3

From: HGPEC: a Cytoscape app for prediction of novel disease-gene and disease-disease associations and evidence collection based on a random walk on heterogeneous network

Fig. 3

A workflow for prediction of novel breast cancer-associated genes and diseases. This task is completed after five following steps: 1) Construct a heterogeneous network by selecting a phenotypic disease similarity network and a network of genes/proteins. 2) Select breast cancer (OMIM ID: 114480, a disease of interest), then identify training genes (i.e., known breast cancer-associated genes) and training disease (i.e., breast cancer). 3) Select a set of candidate genes; all remaining diseases in the network are selected as candidate diseases by default. 4) Rank/prioritize all candidate genes and diseases by the RWRH-based method. 5) Examine ranked genes and diseases by two means: ii) network- and rank-based visualization and ii) collection of annotations and association evidences for highly ranked candidate genes and diseases

Back to article page