Control Strategy Identification via Trap Spaces in Boolean Networks

TL;DR

This paper introduces a novel method for identifying control strategies in biological systems modeled by Boolean networks, leveraging trap spaces to efficiently find interventions that guide the system to desired states.

Contribution

The work presents a new approach using trap spaces to find control strategies in Boolean network models, including methods that incorporate attractor information for improved control.

Findings

Effective control strategies identified for cell fate models

Trap space properties enable efficient computation of control interventions

Method allows control release and escape from traditional percolation-based limitations

Abstract

The control of biological systems presents interesting applications such as cell reprogramming or drug target identification. A common type of control strategy consists in a set of interventions that, by fixing the values of some variables, force the system to evolve to a desired state. This work presents a new approach for finding control strategies in biological systems modeled by Boolean networks. In this context, we explore the properties of trap spaces, subspaces of the state space which the dynamics cannot leave. Trap spaces for biological networks can often be efficiently computed, and provide useful approximations of attraction basins. Our approach provides control strategies for a target phenotype that are based on interventions that allow the control to be eventually released. Moreover, our method can incorporate information about the attractors to find new control strategies that would escape usual percolation-based methods. We show the applicability of our approach to two cell fate decision models.