Skip to main content
Fig. 1 | BMC Bioinformatics

Fig. 1

From: Protein complexes identification based on go attributed network embedding

Fig. 1

The basic idea of GANE to predict protein complexes from protein-protein interaction networks. The GANE method for protein complex prediction is a two-step procedure. Firstly, it learns the vector representation for each protein from the GO attributed PPI network. Based on the pair-wise vector representation similarity, a weighted adjacency matrix is constructed. Secondly, it uses a clique mining method to generate candidate cores. A set of seed cores are generated from the set of candidate cores with density-based clique ranking and redundancy-based clique updating. For each seed core, its attachments are those proteins with correlation scores that are larger than a threshold. The seed cores with attachments are the predicted protein complexes

Back to article page