ADN computers handle complex calculations without the need for electricity

According to Live Science, the new biological computer is capable of performing calculations using billions of molecules in a small drop of water. The machine relies on the laws of physics to create an efficient computing system that can find answers while using less energy than other biological computers.

Damien Woods, professor of computer science at Maynooth University in Ireland, and colleagues describe the scaffolded DNA computer (SDC) in detail in research published September 16 in the journal Nature. They tested SDC on 10 programs, including 100-bit calculations. This number is the total amount of binary information involved in the entire calculation.

SDCs are made of short DNA strands interacting with a longer DNA backbone. The team placed these fibers in small amounts of salt water, then heated and cooled them. When DNA strands interact, they assemble into structures according to a set of programming rules and create calculations. DNA strands act as tiny molecular puzzle pieces, while their sequence determines which pieces can attach to each other and be placed in different positions on the longer scaffold. By designing association rules, the team can program the calculation efficiently.

Woods and colleagues' method exploits thermodynamics, in which a physical system shifts toward an energetically more favorable state. When heating and cooling the mixture, the DNA strands compete to form the most stable structure corresponding to the result of the calculation without relying on any wires or an uninterruptible power supply.

Results are displayed via fluorescence signals on the DNA structure. A low or high signal corresponds to the binary value 0 or 1. Although DNA computers are very small, the number of strands involved in the calculation process is extremely large. A drop of liquid can contain billions, even trillions, of DNA strands.

The research team used SDC to perform more than 700 calculations in experiments, including addition, multiplication by 3, division by 2, and parity checking. The machine can complete small calculations in less than a minute. However, adding 100 bits between 11 million and 34 million takes up to 14 hours. Test results show that the system can be programmed, reused many times and operates faster than other ADN computers, although still slower than conventional computers.

According to Abeer Eshra, assistant professor of computer science and co-author of the study, DNA computers cannot be applied to molecular data storage, power-saving calculations, or devices operating inside living cells.

Interesting Engineering says DNA-based computing and storage infrastructure is becoming a new direction for biological computing. Instead of writing data onto a magnetic surface or running logic through silicon transistors, this method converts digital information into specific arrangements of four basic base units: adenine, cytosine, guanine and thymine. According to New Atlas, the four chemical units that make up the DNA strand also encode information, which can serve as the foundation for algorithmic calculations. The computation process takes place in the form of competition between different chains.

Many research institutes and organizations are developing DNA-based systems. For example, a research group at the California Institute of Technology (Caltech) developed a DNA-based artificial neural network capable of recognizing handwritten numbers (0-9) through molecular interactions in solution without the need for a microprocessor.