Abstract
An increasing number of observations support the hypothesis that the vast majority of biological functions involve interactions between proteins and that the complexity of living systems arises as a result of such interactions. In this paper we apply a multiway spectral clustering technique to protein-protein interaction networks to perform accurate link prediction between proteins. We provide experimental evidence about the accuracy and the performance of the proposed method in real datasets, compared to other known algorithms such as k-means and RWR. © 2010 IEEE.