Method uses linear-DNA -- compatible with standard cell enzymes -- to process signals and produce outputs

A team of researchers from the Univ. of Washington (UW), the California Institute of Technology (CalTech), the Univ. of California (UC) and Microsoft Corp. (MSFT) have come up with a "toolbox" which they say represents the most promising DNA based computer network yet.

The basic idea of a DNA (deoxyribonucleic acid) toolbox is to take inputs -- small strands of DNA or microRNA (micro-ribonucleic acid) -- perform a series of displacement reactions on "Gate" strands, and selectively put out a set of desired outputs that can be used to activate man-made molecules to deliver drugs, turn on sensors, or feed into other DNA networks.

The attractiveness of making DNA computers is that the tools to preserve and replicate your network are already on hand -- in the various cellular nucleases, ligases, topoisomerases, helicases, and polymerases that the cell uses to processes DNA.  Thus future DNA computers composed of inputs (sensors, delivered drug molecules) and outputs (releasable drug packages, selective protein transcription) can be made to be self-repairing and programmable.

Future in-situ computers may be built from DNA. [Image Source: Turbo Squid]

The trick is to come up with a programming language.

The latest work builds on an earlier 2007 paper published in Science by the authors. It uses a DNA displacement approach (like most DNA computer efforts), which implements a basic set of logic "gates" for the DNA computer as a series of DNA reactions.

DNA gate production
The team grows their gates in bacteria using natural enzymes. [Image Source: UW via Nature]

It's the implementation of these reactions that varies from study to study.  In this study the authors use a process of nicking double-stranded linear gate DNA, which undergoes displacement reactions initiated by the signal molecules.  The authors write:

Among the many proposed architectures for strand displacement computation, ours is unique in that it relies exclusively on linear, double-stranded DNA complexes (processed by ‘nicking’ one of the strands). Because this structure is compatible with natural DNA, we are able to produce our computational elements in a highly pure form by bacterial cloning. Thus, we bypass the practical limitations in the length and purity of synthetic strands.

The study's senior author, UW electrical and computer engineering professor Georg Seelig, describes the work stating:

We start from an abstract, mathematical description of a chemical system, and then use DNA to build the molecules that realize the desired dynamics.  The vision is that eventually, you can use this technology to build general-purpose tools.

I think this is appealing because it allows you to solve more than one problem.  If you want a computer to do something else, you just reprogram it. This project is very similar in that we can tell chemistry what to do.

DNA rotatingDNA rotatingDNA rotating
An animation of double-stranded DNA [Image Source: Wikimedia Commons]

The research is still a long way away from offering a solution that's compatible with human cells and can be used with manmade molecules to do something useful.  But the team believes that by laying the foundation for a DNA computer with natural maintenance capabilities, future research will be able to apply the chemical computers to improve drug delivery and watch a human's organs for signs of trouble.

Of course there's also a downside (or upside, depending on your perspective) which the authors don't mention.  The ability to build DNA computers which generate responses that can interact with the cell's natural DNA and biomolecules is highly weaponizable.  For example future researchers could make a DNA computer that lay dormant for some given amount of time, then triggered cells to become aggressive cancer tumors, and then release yet more factors that encouraged those deadly tumor cells to metastasize.  

Cancer cells
DNA computers could also be weaponized to produce signals that trigger tumorgenic behavior in cells and then trigger metastasis to up the damage. [Image Soure: NursingCrib]

Such an approach would offer difficult-to-detect assassinations or be applied on a broader scale as a means of chemical warfare.

The current work was published in the journal Nature, one of academia's most prestigious peer-reviewed journals.  It was funded in part by a $2M USD grant from the National Science Foundation (NSF) to UW electrical engineering professor Eric Klavins, a co-author on the work.

Sources: Univ. of Wash., Nature [abstract]

"It's okay. The scenarios aren't that clear. But it's good looking. [Steve Jobs] does good design, and [the iPad] is absolutely a good example of that." -- Bill Gates on the Apple iPad

Copyright 2017 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki