July 17, 2018
As the global datasphere expands, so does the need for more data storage. Enter DNA, the carrier of life’s genetic information, which offers a potential storage media of unprecedented density, durability, and efficiency.
However, using DNA as storage is currently costly and there is a lack of processing systems suited for this technology.
The University of Illinois at Urbana-Champaign is undertaking a $1.5 million effort to produce new DNA-based storage nanoscale devices using chimeric DNA, a hybrid molecule made from two different sources. Chemical and Biomolecular Engineering Professor Charles Schroeder is one of the Illinois researchers collaborating on the project. As part of the three-year project, “SemiSynBio: An on-chip nanoscale storage system using chimeric DNA,” the team will design a method to read, write, and store data in a more cost-effective way than current DNA storage techniques.
“We’re going to take cheap, native DNA and combine it with chemically modified nucleotides,” said Olgica Milenkovic, Illinois professor of electrical and computer engineering and researcher in the Coordinated Science Lab. “It will be stored and accessed using a novel implementation of a state-of-the-art semiconductor system.”
By 2025, the world will generate 163 zettabytes (one trillion gigabytes) of data a year, in part due to emerging technologies such as the Internet of Things, predicts research group IDC. While not all of that data will be stored, the demand for storage may outpace traditional storage capabilities.
DNA is an attractive solution because it can store up to 455 exabytes (1 exabyte is 1 billion gigabytes) in one gram, according to a 2015 report in the New Scientist. In addition, it is remarkably durable.
“We are still finding bones from over 100,000 years ago from which we can extract DNA,” said Milenkovic. “We don’t know of another storage medium that has such durability.”
During the course of the project, the research team will design and synthesize hybrid DNA molecules that contain non-natural chemical modifications. The goal is to expand the storage capacity enabled by DNA coding methods, essentially increasing the number of equivalent bits.
To accomplish this task, the team will employ concepts from molecular design and engineering and chemical synthesis.
The goal is to create a device that mimics the capabilities of a computer’s hard drive.
“Synthesis of chemically modified DNA essentially constitutes the ‘write’ step, and development of new methods for processing and de-encoding the chemical information stored in the DNA will constitute the ‘read’ step,” said Charles Schroeder, Professor and the Ray and Beverly Mentzer Faculty Scholar in Chemical and Biomolecular Engineering at Illinois. “I envision that the ‘reading’ and ‘storage’ steps could be built into an integrated device.”
The grant is funded through SemiSynBio, a partnership between the National Science Foundation (NSF) and the Semiconductor Research Corporation (SRC), which seeks to lay the groundwork for future information storage systems at the intersection of biology, physics, chemistry, computer science, materials science and engineering. Altogether, SemiSynBio is funding eight projects for a total of $12 million.
In addition to Milenkovic and Schroeder, Illinois collaborators include Jean-Pierre Leburton, the Gregory E. Stillman Professor of Electrical and Computer Engineering and professor of Physics, and Xiuling Li, Professor of Electrical and Computer Engineering. Leburton is also a researcher in the Beckman Institute and the Coordinated Science Lab, while Li is part of the Micro and Nanotechnology Lab.
Milenkovic also recently received a three-year, $2.5 million grant from DARPA to combine synthetic DNA with computing. Her co-PIs include Huimin Zhao and Alvaro Hernandez of Illinois, David Soloveichik of the University of Texas at Austin, and Marc Riedel of the University of Minnesota.
Written by Kim Gudeman, Coordinated Science Lab
For more information about the SemiSynBio program, please see the NSF news release.