FET: Small: DNA Storage and Computation with Strand Displacement Cascades

University Of Texas At Austin, Austin TX

Investigators

Abstract

DNA is incredibly information-dense (up to 6 orders of magnitude denser than optical or magnetic media) and stable (readable over millennia). Storage and retrieval of up to gigabytes of digital information in the form of text, images, and movies has already been successfully demonstrated. DNA storage is typically viewed as a form of passive storage (“cold storage”) and performing computation on the stored data currently involves sequencing the DNA, electronically computing the desired transformation, and synthesizing new DNA, which is an expensive and slow loop. This project seeks to make the next generation of dynamic DNA storage, where chemical interactions can manipulate the stored information “in memory.” Investigators will develop molecular versions of important algorithms for computation on digital data and implement them using programmable DNA-DNA interactions. The project will also contribute to undergraduate and graduate education and support undergraduate hands-on research, as well as provide training of students in applying the principles of computer science and electrical engineering to traditionally incompatible domains of biology and chemistry. This project will develop the SIMD-DNA (Single-Instruction Multiple-Data DNA) paradigm for using DNA strand displacement reactions to manipulate digital information recorded in the topological modification of the DNA substrate (location of strand breaks). Using the inherent parallelism of chemistry, this data processing scheme will be capable of parallel, in-memory computation, eliminating the need for sequencing and synthesizing new DNA on each data update. More specifically, the award will fund the development of the theoretical foundation of SIMD-DNA computation, including strand displacement programs to simulate various data-processing algorithms, as well as the experimental demonstration of the molecular algorithms and methodology. Further, software to simulate and verify SIMD-DNA programs will be developed. Theoretical and experimental studies on strand displacement on DNA of biological origin and sequence (native DNA) will be used to understand and decrease the spurious interactions that can occur in the absence of conventional sequence design. Native DNA has the potential to improve fidelity and significantly decrease cost compared to chemically synthesized DNA. Finally, new methods to improve the speed and efficiency of large strand displacement cascades will be explored, which may generalize beyond DNA storage. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.