Network Reproducibility Using Designed Building Blocks

Many real networks can repeatedly reproduce the same network topology from a small set of building blocks. To understand this phenomenon, we introduce the topic of network design. Network design studies the networks that can be formed by a set of building blocks with rules determining how they can connect to each other.

Within this framework, we introduce convergent and guided convergent assembly (GCA) to explain when a network topology can be predetermined by its design set. Specifically, in GCA we employ statistical inference methods to determine assembly pathways for a variety of systems, including protein complexes, molecules, and circuits. These assembly pathways combine building blocks such that only one network can be designed. Furthermore, we investigate the properties of design sets that determine when convergent assembly or GCA can successfully predetermine a target network topology.