EAGER: Exploratory Software Development & Experiments of Dynamic DNA Nanosystems

Duke University, Durham NC

Investigators

Abstract

There is important need for a synergistic combination of (i) software development that provides rapid development of the experimental DNA nanosystems and (ii) experiments of novel DNA nanosystems, which will guide and provide feedback to the software development efforts. The project will develop software that will provide an integrated design/specification/simulation/optimization environment for designing molecular devices that self-assemble as DNA nanostructures and change state in reaction to their environment, based on a programmable series of hybridization reactions. The integrated methodology will include a molecular programming language that allows high-level specification of states of DNA nanostructures and their state-transitions (involving key hybridization and strand displacement reactions), without fully specifying underlying strand sequences or nanostructures, and provide a WYSIWYG (visual) method for input. A molecular compiler will compile the molecular program into a detailed specification of component DNA stands and hybridization reactions. A thermodynamic and kinetic simulator will calculate equilibrium densities of assembled nanostructures and key kinetic reaction properties; speeding its computations by novel use of decompositions of the strands into component subsequence domains and tensor product decompositions of the state space. Specification (sequence domain-level) and sequence compilation tools will be integrated with thermodynamic and kinetic simulators, so simulations can be done at sequence-domain level, rather than base-pair level, resulting in considerable speedups. A novel experiment design subsystem will give suggested design of experimental demonstrations for verifying assembly of the intended nanostructures, and key reaction steps. Support by the EAGER program will allow a two-year early development phase to test and demonstrate novel concepts and designs. Educational Objectives include cross-disciplinary training of graduate students, carefully supervised mentoring for undergraduates, and summer internships for high school students and teachers. There is substantial multidisciplinary impact to nanoscience, biochemistry and chemistry, which will profit from the introduction of key methodologies derived from mainstream computer science, such as programming languages and compilers, and compiler optimization, particularly since these will be highly specialized and tailored for the specific needs of DNA nanostructures and hybridization reactions. The integration of a molecular programming language, with a molecular compiler, and simulator, with feedback to provide optimization will provide unique enhanced design capabilities for these science disciplines. To maximize impact and use of this software, there is a planned staged external distribution of prototype software system that will incorporate feedback from users.