Digital logic and signal processing computations with molecular reactions.

Abstract

Just as electronic systems implement computation in terms of voltage (energy per unit charge), molecular systems compute in terms of chemical concentrations (em molecules per unit volume). Broadly, the field strives for molecular implementations of computational processes -- that is to say processes that transform input concentrations of chemical types into output concentrations of chemical types. In this dissertation, we present methodologies to implement digital signal processing (DSP) operations, such as filtering and signal transformation, and digital logic operations, such as latching and flip-flopping, with molecular reactions. Molecular reactions that produce time-varying output quantities of molecules as a function of time-varying input quantities are designed according to a DSP or logic specification. Unlike all previous schemes for molecular computation, the methodology produces designs that are dependent only on coarse rate categories for the reactions ("fast" and "slow"). Given such categories, the computation is exact and independent of the specific reaction rates. We first present a methodology for implementing DSP through a globally synchronous, locally asynchornous scheme we call the RGB scheme. We then present a general methodology for implementing synchronous sequential computation. We generate a four-phase clock signal through robust, sustained chemical oscillations. We implement memory elements by transferring concentrations between molecular types in alternating phases of the clock. Thirdly, we propose a general methodology for implementing asynchronous sequential computation, including a method to schedule data flow for feed-forward systems and a method to implement systems with feedback loops. Finally, we present a methodology for systematic synthesis of various types of sequential digital logic. Given a system specification, a chemical reaction network is synthesized to perform the input/output logic functions. Synthesized systems are concise and robust in that computation accuracy does not depend on specific values of rate constants. All designs are mapped into DNA strand displacement reactions and validated through transient simulations of the chemical kinetics at the DNA reactions level.