Despite the often misreported news articles on biology, there is no clear cut function of a single biological component such as a protein, gene or metabolite. There is no “gene” for lung cancer, rather, there is a genetic network that when perturbed, causes lung cancer. What is more important is a biological component’s relationship with the enormous regulatory systems of proteins, genes and metabolites that we call life. The importance of the systemic relationship of a node over the individual characteristics is similar to that in social networks and the internet - no node is an island. In reality, cellular function such as programmed cell-death is a product of several different biological components coming together.

Albert-László Barabási and Zoltán N. Oltvai wrote a seminal review paper entitled, “Network Biology: Understanding The Cell’s Functional Organization” in Nature Reviews Genetics (2004). Barabási and Oltvai connect many concepts from graph theory to biological sciences, one of which is a network that follows a power law distribution of connectivity, known as a scale-free network. Protein-protein interactions, flux of metabolic reactions and gene regulatory networks have been shown to form scale-free network, in which there is an inner “core” of highly connected nodes and a large “crust” of sparsely connected nodes. The advantage of biological scale-free networks is that they are incredibly robust against common accidental failure - a gene network can still persist with a large percentage of randomly mutated genes. In reality, biological networks are more than just scale-free networks - they are hiearchical networks with interconnected modules of nodes which together perform a specific function.

There is evolutionarily evidence that gene duplication, a major driver of genome evolution, as contributing to the evolution of scale-free networks in life.Barabási and Zoltán cite evidence that the highly connected proteins modules are evolutionarlily more conserved from yeast to higher organisms like humans, indicating there is value in the highly connected networks. The model is that (1) duplicated genes increase the connections of highly connected genes (”rich-get-richer” mechanism) and (2) genes that are duplicated are most likely sparesely connected, nonessential genes.

With the existence of set of highly connected pathways, there may be a way of using VCG applied to network routing to judge the quantitative importance of a critical gene to serve as a hub, in the style of the Google PageRank mechanism discussed in class.