Computation via synthetic gene circuits is a foundational paradigm for programming cellular behavior. Analogizing such circuits as logic circuits composed of a hierarchy of “biological gates” connected by “molecular wires,” has been a dominant theme in synthetic biology. The roughly linear number (as a function of circuit size) of such orthogonal wires, however, and the characterization effort and engineering expertise needed to use them, challenges the scalability of this approach.
Our recent work on boolean logic and arithmetic circuits through DNA excision (BLADE) offers an alternative programming principle: a BLADE circuit is single-layer, using input recombinases as the only molecular wires, leveraging their unique ability to toggle gene expression anywhere in the circuit (Weinberg et al. http://dx.doi.org/10.1038/nbt.3805). In contrast to primitives like NAND/NOR as basis for synthesizing more complex programs, a BLADE device resembles a field-programmable device based on “if-then- else” statements. This design alternative allows scaling to complex circuits while avoiding some of the challenges of tuning and computational search during design. It enabled us to create 100s of complex circuits (e.g. full adder) as per specification, with 96% of them measuring close to specification. It enabled us to decouple circuit outputs and assembly from logic design, recursively scale up to n inputs/outputs, and abstract logic similarly as in electronic circuit design, exemplifying some of the fundamental principles of synthetic biology (Way et al.http://dx.doi.org/10.1016/j.cell.2014.02.039).
Our work highlights the novelty of computational architectures enabled by a diversity of genetic components with unique behaviors. It illustrates the advantages of exploiting DNA recombinatorics which makes a large state space accessible to computation. Biological applications demand complex and robust circuits, which we believe will necessitate scalable device architectures accompanied by sophisticated design-build software, informed by user-defined specifications and available biochemical information.
I welcome thoughtful comments here and on Twitter (@pipette_ninja).
(I thank Wilson Wong and Ben Weinberg for their contributions to this post. A derivative of this post was published by Cell Systems’ Principles of Systems Biology #16 here.)
See Bhatia's original blog post here.