Noise as a Computational Resource

Download

PDF Download (87Kb)

Description/Abstract

In systems far from equilibrium, such as cellular biomolecular assemblies, energetic input is converted into systematic execution of function. The functional machinery comprises transport and interconversion of matter, as well as signalling systems and the regulation of other functional components. Within the microscopic dimensions of the cell, these processes are carried out by discrete co-ordinated interactions among molecules in a noisy environment. We take the position that given the pronounced effects noise can have in such small volumes having low copy numbers of molecular species, cells have harnessed evolutionary pressures into making productive use of noise. Correspondingly, given the drive towards miniaturisation in future computational hardware, we can view the attendant concerns about “taming” the noise inherent to this regime as an opportunity to learn from the way cells fulfil their transport and information processing needs. In particular, we shall look at how molecular ratchets exploit thermal noise, how signalling processes may exploit fluctuations in the number of enzymes, and how the ability to read out from conformational substates of enzymes can enable targeted low-pass filtering to guide computational steps through a suitably mapped state space.