Ivan I. Garibay and Annie S. Wu
The problem of designing basic building blocks that self-reproduce, self-assemble, and self-organize into increasingly complex functionalities has already been solved by nature. Proteins are the basic building blocks of biological functionality. The genome, proteome, transcriptome and metabolome interact via regulatory networks, protein interaction networks, and metabolic webs, to build functional modules and ultimately achieve large-scale functional biological organization. Since proteomics seeks to understand the functions of the proteins as well as their complex interaction, the fundamental question that we post is: does the science of proteomics have something to teach us about the design competent functional building blocks and about the procedures for combining them to achieve high-level complex functionality? We believe that, as in many other cases, cross-fertilization among computational synthesis and proteomics would be productive. In particular, we provide a brief discussion on a relatively new and surprising proteomic result that may provide some insights on how to effectively organize building blocks: biochemical interaction networks share large-scale topology and properties. These networks are scale-free, modular, hierarchical, error and attack tolerant, their elementary building blocks selforganize into small recurrent patterns (i.e., motifs in gene regulatory networks, and pathways in metabolic networks), and they follow a simple self-organization premise: new nodes, if possible, prefer to connect to highly connected existing nodes. These results suggest that along the road from basic building blocks to high level functionality, we may find it useful to make a stop in the town of proteomics.