Heat-rechargeable computation in DNA logic circuits and neural networks

Area of Science:

• Molecular engineering
• Biophysics
• Chemical systems

Background:

• Biological systems utilize metabolism for energy, while artificial molecular machines lack a universal power source like ATP or electricity.
• DNA has been used as fuel for nanodevices and reactions, but requires specific sequences, limiting its universality.
• Current artificial molecular systems struggle with sustained operation and require adaptable energy solutions.

Purpose of the Study:

• To investigate heat as a universal energy source for enzyme-free DNA circuits.
• To demonstrate the ability of heat to drive molecular systems from equilibrium to out-of-equilibrium states.
• To enable sustained computation and advanced behaviors in artificial chemical systems.

Main Methods:

• Utilizing temperature ramps (heating and cooling) to induce secondary structures in nucleic acids.
• Employing kinetically trapped states in DNA to provide energy for computation.
• Designing and implementing complex logic circuits and neural networks with over 200 molecular species.

Main Results:

• Heat successfully restored enzyme-free DNA circuits from equilibrium to out-of-equilibrium states.
• Complex logic circuits and neural networks responded to temperature ramps, enabling computation.
• Circuits recharged within minutes, allowing at least 16 rounds of computation with sequential inputs.
• Demonstrated sustained operation and consistent performance without problematic waste build-up.

Conclusions:

• Heat serves as a universal, ATP-like energy source for enzyme-free DNA circuits.
• This strategy enables sustained, iterative computation and autonomous behaviors in artificial chemical systems.
• The approach is scalable and opens new avenues for advanced molecular machine functionalities.