Integration of Network Information for Protein Function Prediction

Abstract

Protein function prediction is a canonical problem in computational biology. For protein activities described by terms in databases such as the Gene Ontology (GO), this task is typically pursued as a binary classification problem. As a result of an astonishing increase in the available genome-wide protein information, integrating different protein data sets in terms of networks has become a significant opportunity and a major focus to infer functionality. This chapter, beginning with some background, introduces a network-based probabilistic framework relying on the protein-protein interaction (PPI) network to classify proteins into functional categories, continuing with extensions to the Gene Ontology hierarchical structure and to categorical protein features, and finishing with a discussion of issues with this framework and future work.