The inventor of a program for simulating DNA computing systems hopes the benefits of CAD will have the same impact on the field as it has in others.
Nathanaël Aubert, a computer science PhD student at the University of Tokyo, has created a program he calls DACCAD (DNA Artificial Circuits Computer Assisted Design) to help researchers design biological computing systems from a DNA toolbox – a set of DNA-and-enzyme-based modules that can be combined to carry out computational tasks.
According to Aubert, CAD has helped progress a host of fields from integrated circuits to aeronautics by allowing the automation of error-prone or repetitive tasks and easy simulation and verification of systems, and he now hopes to bring these benefits to the field of computational biology.
The inspiration for the program came from seeing colleagues in his lab, who carry out real-life in vitro DNA computing experiments, struggling with the time-consuming and trial and error nature of the field.
“I could see it took them a lot of time to simulate even simple systems because they had to input all the equations by hand,” he said. “The idea is that it’s much faster to do a simulation on a computer than doing the experiment in real life. You can detect design problems much earlier than without computer assistance.”
The Java-based program, described in research published in the Journal of the Royal Society Interface today, allows researchers to combine modules from a DNA toolbox created by Montagne et al. to create chemical reaction networks (CRNs).
These networks can display a variety of nonlinear behaviours including oscillations and multi-stability that make them powerful systems for information processing, but their complex and cross-interacting nature makes it hard to predict outcomes when designing anything but the simplest systems.
Hours of work on complex system can yield completely unexpected results and require researchers to return to the drawing board, but the program created by Aubert allows researchers to first simulate systems on a computer using a simple graphical representation before creating them in vitro.
“You can check if your intuition was right and if not you can simply modify the design or use the tools provided with the software to optimise the parameters. Once it works the way you expect you can move on to the next step and use other existing tools to design the implementation of the system,” he said.
“Hopefully, with the help of CAD, people will be able to make much more complex and diverse DNA computing systems, because right now most systems are at the proof of concept level.”
DACCAD is not the first piece of design software created for the molecular programming field, but is the first able to design enzymatic CRN systems – a promising area of research, according to Aubert, as they are reusable and more efficient than other “conceptually simpler” and “elegant” enzyme-free systems.
The software is also able to convert any designed system to synthetic biology markup language (SBML), a common language shared by other computational biology tools, and Aubert believes it is simple enough to allow researchers unfamiliar with the field to try their hand at it.
“I think it’s the first step towards making things that are more than just academically interesting with the DNA toolbox,” he said. “We are very far from what a computer can do although we have the same computational power theoretically.
“Hopefully it will first help people who are not experts in DNA computing design new systems and get more people working in the field, but the other factor is it will help people make more interesting systems and actually implement them in wet labs.”