Protein function by minimum distance classifier from protein interaction network

Protein are involved in every performed in a cell. Determining protein function using computational technique is the most challenging problem in post genomic era. Using sequence homology, phylogenetic profiles, gene expression data function of unknown protein can be predicted. Recently the large interaction networks are constructed from high throughput techniques like yeast2hybrid experiments are also used in prediction of protein function. Based on the concept that a protein performs similar function like its interacting protein, a method is proposed using minimum distance classifier which classifies the protein pairs with respect to the hypergeometric distribution and correlation coefficient as the feature of distance function. The overall accuracy is achieved.