Driving the Scalability of DNA-Based Information Storage Systems

Abstract

The extreme density of DNA presents a compelling advantage over current storage media; however, in order to reach practical capacities, new approaches for organizing and accessing information are needed. Here we use chemical handles to selectively extract unique files from a complex database of DNA mimicking 5 TB of data and design and implement a nested file address system that increases the theoretical maximum capacity of DNA storage systems by five orders of magnitude. These advancements enable the development and future scaling of DNA-based data storage systems with reasonable modern capacities and file access capabilities..

Probing the physical limits of reliable DNA data retrieval

“…Previous work in this space recognized the importance of storage density for DNA to become a practical archival storage 1,3,7,10 , but the greatest complexity surrounding random access in those works reached just over 10 7 unique sequences 10 . In more recent work, a different method of random access using bead extraction of desired strands prior to PCR random access utilized 10 18 unique sequences 11 . However, in addition to different random access techniques and strand architecture, those methods differ substantially from the methods presented here and make it difficult to compare to this work.…”

Dynamic and scalable DNA-based information storage

“…ss-dsDNA strands can be efficiently created in one-pot. As future DNA databases would be comprised of upwards of 10 15 distinct strands 17 , we first asked if ss-dsDNAs could be created in a high throughput and parallelized manner. We ordered 160 nucleotide (nt) single-stranded DNAs (ssDNA) with a common 23 nt sequence that was inset 20 nt from the 3' end ( Fig.…”

“…2d, left, blue). While it is possible to increase the number of distinct strands per address (i.e., information per file) to make up for the loss of addresses, this would result in files too large to be sequenced and decoded in a single sequencing run 17 . It is also important to note that our simulations were based upon very cycles of PCR generated the optimal amount of 160 nt ss-dsDNAs while minimizing excess ssDNA production.…”

“…The creation of ss-dsDNA strands is simple and high throughput, it is compatible with existing file system architectures including hierarchical addresses 11,17 , and it facilitates scaling of capacity. While the need to include the T7 promoter in every strand does occupy valuable data payload space, it is a worthwhile tradeoff: the T7 promoter decreases data density and capacity in a linear fashion, yet it more than compensates by simultaneously improving both metrics exponentially by allowing many sequences to appear in the data payload that normally would have to be avoided in PCR-based systems (or conversely by allowing the full set of mutually nonconflicting addresses to be used) 11,17 . It is also important to note that DORIS may help solve scalability issues with the encoding process as well: cross-comparing all address sequences with all data payload sequences is computationally intractable, but the need to do this is eliminated with DORIS as addresses will not physically interact with data payload sequences.…”

“…While these criteria have not yet been achieved in aggregate, we were inspired by the creative use of molecular biology approaches in prior work to address some of these challenges. For example, polymerase chain reaction (PCR) is the predominant method for information access in DNA storage systems 7,12 and is scalable, especially with some modifications 17 , while single-stranded DNA toeholds and strand displacement have been used for DNA computation [18][19][20] , DNA search 21 , detection 22 , and rewritable 12,[23][24][25] information storage. The challenge is that in their current form these technologies have inherent limitations and tradeoffs either in physical scalability, encoding density, or reusability.…”