From Architecture to Function (and Back) in Bio-networks

Abstract

Living organisms exhibit systemic or emergent behavior due in large part to the combinatorial interdependencies among the components in the complex molecular systems that comprise them. Biological networks encode these interdependencies into convenient structural representations, enabling their analysis and understanding. Examples include: synergistic behavior of transcription factors during gene regulation, where several of them act together atomically to elicit gene expression response; the modular architecture of biological networks and its relation to function; and the enrichment or certain small topological features in the networks, e.g. triangles, called network motifs. Studying the connection between biological network architecture and function is made possible by novel methods enabling correct statistical and combinatorial accounting of topological features in the network systems.

My focus in the past few years has been on characterizing biological networks in terms of the underlying local topologies or building blocks, and linking those building blocks to observable function in living organisms. Here I describe a set of tools and techniques that help us do that, together with results and insights from our studies. The techniques described come from various scientific disciplines that have had to deal with networked systems for some time, like social sciences, statistical mechanics, and complex network theory among others. In all case we have adapted these tools to biological network analysis and either extended or improved upon them.