Chapter 7. Topological properties of protein interaction networks

Sergey Maslov, Brookhaven National Laboratory

Protein interaction networks lack the top-down design.

Instead, selective forces of biological evolution shape them from the raw material provided by gene duplications followed by functional divergence of paralogous proteins. This is reflected in large-scale topological properties of these networks. This chapter will review recent research related to the interplay between topology and biological function in these networks.

Outline:

Protein interaction networks lack the top-down design. Instead, selective forces of biological evolution shape them from the raw material provided by gene duplications followed by functional divergence of paralogous proteins. This is reflected in large-scale topological properties of these networks characterized by

(i) the existence of "hub"-proteins having a disproportionately large number of binding partners

(ii) "small world" architecture in which most pairs of nodes could be linked to each other by a relatively short chain of interactions involving several intermediate proteins. While such architecture facilitates meaningful signaling it also presents a potential problem by providing a conduit for propagation of undesirable cross-talk between individual functional units/pathways.

(iii) densely interconnected modules correlated with biological function.

In this chapter I will review recent research related to the interplay between topology and biological function in these networks. In particular, I will concentrate on:

(i) Algorithms aimed at detection of

· functionally significant topological patterns such as

· degree-degree correlations reflecting the hierarchical structure of PPI networks

· small motifs involving PPI and other protein networks (transcriptional regulation, protein modification, etc.)

(ii) How the topology of these networks affects the propagation of signals and noise.

In particular, I explore how large (several-fold) changes in protein levels of a small number of proteins shift the Law of Mass Action (LMA) equilibrium between bound and unbound concentrations in the whole PPI network. Such changes have a potential to cascade down a small-world network. Most often such “action at a distance” would represent an undesirable effect which has to be either tolerated or corrected by the cell. However, under particular circumstances reversible changes in the LMA equilibrium could be used for regulation and signaling between or within individual pathways.

Even in the absence of large systematic changes protein levels are subject to random fluctuations due to a stochastic nature of their production and degradation. It is important to understand how these “raw” fluctuations translate into fluctuations in biologically relevant bound and unbound concentrations.