Form, Function, and Evolvability in Biological Networks

A driving problem in systems biology for several years has been exploring the extent to which the topology of a small biological network constrains or guides its function. The absence of such constraint would allow a given network to evolve without rewiring its underlying form. We introduce a quantitative measure of this evolvability that does not rely on pre-defining the preferred function of a given topology. We then study the stochastic description of the experimental setup of Guet [1], treating chemical inducers as functional inputs and the expression of a reporter gene as the functional output. We take an information-theoretic approach, allowing the system to set parameters that optimize signal processing ability, thus enumerating the highest-fidelity functions. We find that, while all networks studied are highly evolvable by our measure--meaning that the function has little dependence on location in parameter space--the evolvability is correlated with individual topological features. Certain topological attributes, then, are shown (with statistical significance) to convey evolvability to biological networks. [1] C. C. Guet et al., Science \textbf{296}, 1466 (2002).