Fig. 1From: A tensor-based bi-random walks model for protein function prediction- Flowchart of RWRT method. A Constructing the functional similarity tensor T according to functional association analysis based on original PPI network, domain context and protein complex information. B Given a testing protein pi, running a bi-random walks algorithm on the tensor T to obtain the functional similarity vector Xi between pi and the other proteins. C Removing redundant partners that have more external functional similarity than internal functional similarity within the module consisted of the target protein and its partners, and picking out top K of all functions of the remaining partnersBack to article page