DNA-based Infrastructure for Storage and Computation

Introduction

Throughout evolution, Biology has found an almost-perfect molecule to pass information down through the millennia. Whether for a simple bacterium or the largest whale, DNA achieves high information density while permitting cellular machinery rapid random-access retrieval to decode everything from a single protein to the development of an entire human brain.

Humanity's rapidly growing data storage and computing needs make it tempting to exploit optimised biological molecules and processes, but biology is messy, poorly understood, and much more specialised for life than it is for our needs. We propose to use DNA for storage together with a rationally designed, well-characterised, and robust molecular computing architecture.

Project Overview

The DISCO project will address the challenge of engineering a programmable and robust DNA storage and computing platform. DISCO combines powerful molecular-algorithmic ideas from DNA computing with notions of thermodynamic stability from DNA nanostructures to provide an expressive and robust system design.

Methodology

1. DNA Scaffold Strands: The project proposes the use of long DNA scaffold strands, upon which hundreds of smaller strands bind to store data.
2. Data Operations: These strands can be later read, erased, rewritten, and computed upon.
3. Prototype Development: DISCO begins with the aim of trialing a small, but thoroughly characterised and modelled prototype that facilitates storage and computation on 10 bits.

Scaling Up

Then, DISCO will scale up the system to hundreds of bits by leveraging recent technological advances to enable high-throughput techniques at both the experimental and data-readout stages.

Future Directions

Finally, DISCO will be extended to two-dimensional DNA origami structures, each carrying about a hundred bits, but with millions of structures in the same test tube. This will exhibit DNA storage and computation in a distributed system at a massively parallel scale.