Deep Reinforcement Learning for Stabilization of Large-scale Probabilistic Boolean Networks

TL;DR

This paper presents a model-free deep reinforcement learning approach for controlling large-scale Probabilistic Boolean Networks, enabling targeted state stabilization without relying on transition probabilities, demonstrated on networks with up to 200 nodes.

Contribution

It introduces a scalable, model-free deep RL framework for stabilizing large PBNs, applicable to various control scenarios, and demonstrates its effectiveness on networks with hundreds of nodes.

Findings

Successful control of large PBNs including a 200-node melanoma network

Linear time complexity in training interactions

Effective control without using probability transition matrices

Abstract

The ability to direct a Probabilistic Boolean Network (PBN) to a desired state is important to applications such as targeted therapeutics in cancer biology. Reinforcement Learning (RL) has been proposed as a framework that solves a discrete-time optimal control problem cast as a Markov Decision Process. We focus on an integrative framework powered by a model-free deep RL method that can address different flavours of the control problem (e.g., with or without control inputs; attractor state or a subset of the state space as the target domain). The method is agnostic to the distribution of probabilities for the next state, hence it does not use the probability transition matrix. The time complexity is linear on the time steps, or interactions between the agent (deep RL) and the environment (PBN), during training. Indeed, we explore the scalability of the deep RL approach to (set) stabilization of large-scale PBNs and demonstrate successful control on large networks, including a metastatic melanoma PBN with 200 nodes.