Researchers have created a DNA-based computing device that performs calculations across billions of molecules inside a single drop of water. The system is known as the Scaffolded DNA Computer, or SDC, and it relies on thermodynamics rather than electricity to run its operations. Rather than forcing a calculation through a step-by-step process, the design allows the molecules to naturally settle into their lowest-energy state, which then reveals the result.
A study published Sept. 16 in Nature introduced a new kind of computer, one that adds to a growing body of research where living cells and standard hardware work together. This particular design follows its own course. The SDC can be programmed and used again, though it works more slowly than silicon, it runs faster than other DNA computers.
How the SDC Works
A long DNA scaffold binds together shorter strands of DNA to form the SDC. These short strands are put into a small amount of salt water, then heated and cooled. The changes in temperature cause the strands to join into structures based on programmed rules. Each strand behaves like a puzzle piece, with its sequence deciding which pieces link up to one another and to various positions on the scaffold.
According to Damien Woods, a professor of computer science at Maynooth University in Ireland who co-authored the study, the key innovation is how the DNA molecules bind together. Rather than being assembled in a fixed sequence, they compete with one another for the right to attach to the scaffold. Through that jostling and competition, information gets processed and a computation executed, ultimately settling into an energetically-preferred state that encodes the result.
Programming the Computer
The team switches between calculations by picking fresh DNA strands from the fridge. Every program is tied to its own batch of strands, so changing the computation or altering the input means reaching for a different set from the cold storage.
Thermodynamics underpins the method, which harnesses the tendency of physical systems to reach more stable states. As the mixture is heated and then cooled, the strands contend for the most stable arrangements, making the correct configuration the energetically favorable one. Once the structure forms, it encodes the answer, letting the molecules “compute” through simple interaction.
What the Researchers Tested
The group ran the SDC on 10 pieces of software, among them 100-bit operations. Certain sums, such as 10 plus 3, took roughly 30 seconds. More complicated totals required much more time — a harder sum in the neighborhood of 11 million to 34 million items could take up to 14 hours.
Their experiments spanned over 700 computations. The programs covered addition, multiplication by 3, division by 2, and eight-bit parity detection, a familiar form of error correction in computing. Short calculations were finished in under a minute, an impressive result considering the chemical reactions involved take the same amount of time or longer.
Constantine Evans, a senior research fellow at Maynooth University and a co-author of the study, described the calculations as trivial ones that anyone could quickly complete on their own, and noted that a silicon computer would finish in an instant. He pointed out that the system his team developed relies on only a few molecules, which never follow an organized sequence of steps and never make irreversible ones, and yet still arrive at the correct answer. He explained that when it comes to computation at a molecular scale, even those seemingly simple operations are very difficult to carry out reliably.
Reusability and Future Directions
The SDC can be used more than once. Many earlier molecular computers were built for single use only, but this design allows the molecules to be employed again. The researchers pointed out that the uses remain uncertain. DNA-based systems might help with storing data on a molecular scale, with computation that takes little power, or even with devices that run inside living cells.
Abeer Eshra, an assistant professor of computer science and co-author of the study, explained that molecular computers are not meant to replace electronic ones. Instead, they could find use in biological environments, smart materials and archival DNA data storage. She added that the work opens up a fresh approach to DNA data storage, with any information kept in such a system naturally including its own error correction.
A Comparison of DNA Computers
| Feature | SDC | Other DNA computers |
|---|---|---|
| Energy input | Uses thermodynamics to settle into lowest-energy state | Requires continuous energy input |
| Reusability | Designed to be reused | Often one-time experiments |
| Speed | Faster than other DNA computers | Slower than the SDC |
| Programming | Select different DNA strands from the fridge | Varies by design |
The SDC’s pace stands out. It moves more slowly than silicon, yet it is quicker than other DNA computers. A tiny drop of fluid holds billions, and sometimes trillions, of DNA strands, as Eshra stated. The strands act upon one another to yield an answer.
Woods and Eshra described the system as programmable and reusable, and they noted that while it moves at a slower pace than silicon, it still works faster than most other DNA computers. They explained this mix of traits — programmable, reusable, and relatively quick for molecular computing — as the reason the SDC stands out.
A working model has been built to show that computing can take place at the molecular level without needing constant power. The answer appears naturally as the system settles into its lowest-energy state. Further development depends on whether researchers can make it larger and prove it reliable through repeated use. Right now, the real story is simply that it works.
Source material: “Scientists build a DNA computer that can perform calculations in a drop of water,” Live Science.
Get the Notebook.
The day's best stories and every fresh verdict, in plain English, in your inbox by seven. One email a day, no more.