"Topology and Function in Protein Interaction Networks"

Protein interaction data are an important source of information for addressing the questions of what proteins do and how they
 work together to perform diverse tasks.  Protein interactions, which, taken together, can be represented as networks or graphs,
 have been determined on a large scale for several organisms. In my work, I study the relationship between protein function and 
interaction network topology, focusing on protein-protein physical interaction networks. In the first part of the talk, I will 
discuss a novel network analysis paradigm in which known information about protein features, such as molecular function or domain
 content is used to uncover the organizational principles of interaction networks, as revealed in recurring patterns of interaction 
among different types of proteins. I will then address the use of physical interaction networks for predicting protein function. 
 I will begin by discussing which topological properties of interaction networks should be taken into account by network-based 
function prediction algorithms and will then introduce an original network-flow based algorithm for predicting protein function.  
This algorithm, FunctionalFlow, takes advantage of both network topology and some measure of locality, and, as a result, has improved 
performance over earlier methods.