Toward Understanding the Structure and Function of Cellular Interaction Networks

Abstract

The components of a biological cell (such as proteins and molecules) interact on several levels, forming a variety of networks. Newly-developed high-throughput molecular biology methods now allow for the construction of genome-level interaction graphs. In parallel, high-throughput molecular abundance data paired with computational algorithms can be used to infer graphs of interactions and causal relationships. Graph-theoretical measures and network models are more and more frequently used to discern the functional constraints in the organization of biological networks. Perhaps most importantly, the combination of interaction and abundance information allows the formulation of predictive dynamic models. Here we review some of the dominant theoretical and computational methods used for the inference, graph-theoretical analysis and dynamic modeling of cellular networks, and we present two specific projects that integrate these three steps. Throughout this chapter we focus on presenting the significant advances in understanding biological systems allowed by graph theory. Conversely, we hope that the new biological data and results may inspire new directions in graph theory.