Nowadays it is possible to understand the basic components and organization of cell machinery from the network level due to increased availability of large-scale protein-protein interaction (PPI) data. Many studies have shown that clustering of the protein interaction network (PIN) can be found as an effective approach for identifying protein complexes or functional modules. A significant number of proteins in such PIN remain uncharacterized and predicting their function remains a major challenge in system biology. We propose a protein annotation method based on clustering according to Bond Energy Algorithm (BEA), which first transforms the PIN into matrix form suitable for BEA, and after generating the resulting matrix of the BEA the AUTOCLASS algorithm is performed to obtain the PIN clusters. Protein functions are assigned based on cluster information. Experiments were performed on PPI data from the bakers' yeast and since the network is noisy and still incomplete, we use pre-processing and purifying. Results reveal improvement over previous techniques and the most prominent characteristic of the BEA clustering is that the clustering result is not dependent of the initial number of clusters.
Bond Energy Algorithm (BEA), protein interactions network, clustering methods, protein function prediction