Elaborate computation tasks can be performed by distributing the work across interconnected elementary information processing units. This principle underlies not only the operation of integrated electronic circuits, but also of many biological processes including development and, of course, the activity of the brain.
In two reports recently published in Nature, Chris Voigt’s lab and the team lead by Ricard Solé and Francesc Posas report the construction of synthetic biological circuits performing distributed multicellular computation (Tamsir et al, 2010, Regot et al 2010). In the implementation presented by Tamsir et al, individual E. coli colonies carrying a simple genetic cascade (NOR gate) are interconnected via quorum sensing signaling molecules to perform complex operations (XOR, or EQUAL). Similarly, Regot et al build multicellular circuits (e.g. multiplexer or 1-bit adder with carry) using mating pheromones to chemically ‘wire’ together engineered yeast cells that perform a variety of basic 2-input logical functions.
These works show that compartmentalizing elementary synthetic circuits enables combinatorial and flexible assembly of complex circuits and can improve the robustness of the resulting computation. What could be the next step? Looking at the operation of the brain, arguably the most powerful living computing device known, one is tempted to suggest that narrowing the diffusion range of the chemical cell-to-cell transmitter within ‘synthetic synapses’ would facilitate the miniaturization of multicellular computing networks and potentially open the door to scalable designs of arbitrary complexity.
Tamsir A, Tabor JJ, Voigt CA (2010). Robust multicellular computing using genetically encoded NOR gates and chemical ‘wires’. Nature doi: 10.1038/nature09565
Regot S, Macia J, Conde N, Furukawa K, Kjellén J, Peeters T, Hohmann S, de Nadal E, Posas F, Solé (2010). Distributed biological computation with multicellular engineered networks. Nature doi: 10.1038/nature09679