Optimising attractor computation in Boolean automata networks

TL;DR

This paper presents a method to optimize the size of Boolean automata networks for attractor computation using modular decomposition, analyzing complexity and practical examples.

Contribution

It introduces a novel optimization approach based on network decomposition and discusses its complexity and practical applicability.

Findings

The method effectively reduces network size for attractor computation.

Complexity analysis of most parts of the process is provided.

Some aspects of the problem remain computationally open.

Abstract

This paper details a method for optimising the size of Boolean automata networks in order to compute their attractors under the parallel update schedule. This method relies on the formalism of modules introduced recently that allows for (de)composing such networks. We discuss the practicality of this method by exploring examples. We also propose results that nail the complexity of most parts of the process, while the complexity of one part of the problem is left open.