On the Simulation Power of Surface Chemical Reaction Networks

TL;DR

This paper explores the computational capabilities of surface chemical reaction networks (sCRNs), demonstrating their equivalence with other distributed models like cellular automata and tile automata, highlighting their potential for pattern simulation.

Contribution

It establishes the mutual simulation power of sCRNs and other models, connecting chemical reaction networks to well-studied distributed computation frameworks.

Findings

sCRNs can simulate tile automata, cellular automata, and amoebot models

All models can simulate each other with initial configuration

A coloring technique enables global orientation in sCRNs

Abstract

The Chemical Reaction Network (CRN) is a well-studied model that describes the interaction of molecules in well-mixed solutions. In 2014, Qian and Winfree [22] proposed the abstract surface chemical reaction network model (sCRN), which takes the advantage of spatial separation by placing molecules on a structured surface, limiting the interaction between molecules. In this model, molecules can only react with their immediate neighbors. Many follow up works study the computational and pattern-construction power of sCRNs. In this work, our goal is to describe the power of sCRN by relating the model to other well-studied models in distributed computation. In this work, our main result is to show that, given the same initial configuration, sCRN, affinity strengthening tile automata, cellular automata, and amoebot can all simulate each other (up to unavoidable rotation and reflection of the pattern). One of our techniques is a coloring on-the-fly, which allows all molecules in sCRN to have a global orientation.