Computational capabilities of a multicellular reservoir computing system.

Area of Science:

• Synthetic Biology
• Computational Biology
• Biophysics

Background:

• Cells process information for ecological, industrial, and biomedical applications.
• Single-cell information processing is limited by molecular complexity and metabolic burden.
• Multicellular systems offer a solution by combining cells with designed subfunctions.

Purpose of the Study:

• To introduce reservoir computing (RC) for enhanced information processing in synthetic multicellular systems.
• To extend RC to multicellular populations using diffusion-based cell-to-cell signaling.
• To demonstrate the feasibility of diffusion-based multicellular reservoirs for temporal computing tasks.

Main Methods:

• Simulated a 3D community of cells communicating via diffusible molecules as a reservoir.
• Applied the multicellular reservoir to approximate binary signal processing tasks, including median and parity functions.
• Compared the computational advantage of multicellular reservoirs over single-cell reservoirs.

Main Results:

• Demonstrated that diffusion-based multicellular reservoirs can perform complex temporal computing tasks.
• Showed a computational advantage of multicellular reservoirs over single-cell reservoirs.
• Identified biological properties influencing the computational performance of these systems.

Conclusions:

• Diffusion-based multicellular reservoirs represent a feasible synthetic framework for advanced temporal computing.
• This approach overcomes limitations of single-cell information processing.
• Further research can optimize biological properties for enhanced computational performance.