Most biocomputing research to date has sought to replicate silicon-type computation involving logic gate-based architectures, with significant success. For example, researchers from ETH Zurich have used the CRISPR-Cas9 gene editing tool to create processors inside human cells.5 The Cas9 enzyme reads inputs in the form of guide RNA, and responds by expressing particular genes that then create certain proteins as the output. The result is that the cells effectively compare (or add) two inputs and deliver the result as two outputs.
Impressive as such proofs of concept are, a plethora of different functionalities occurs naturally in the operations of biological cells such as bacteria, and if we look, we may find a multiplicity of architectures for biological data processing. Research to date has tended to focus on programming DNA-based systems, for instance, creating “genetic circuits”. This is an issue of familiarity: we know how to perform genetic engineering operations well enough to make progress. However, alternative biological hardware, such as nerve cells or the cytoskeletons of cells may provide an even richer set of possibilities for biological information processing. There is also scope for performing “whole-cell biocomputations” that tap into the cell’s metabolism.6
