From Static to Dynamic: Exploring Temporal Networks in Systems Biology

TL;DR

This paper reviews the use of temporal networks in systems biology, emphasizing their importance for modeling dynamic biological systems and providing an overview of analytical methods and tools for researchers.

Contribution

It offers an integrative overview of dynamic network analysis in biology, clarifying concepts and presenting multi-scale analytical strategies for temporal networks.

Findings

Highlights the importance of temporal networks in biological research

Provides a comprehensive overview of analytical methods and tools

Discusses challenges and future directions in dynamic network analysis

Abstract

Network science has become an essential interdisciplinary tool for understanding complex biological systems. However, because these systems undergo continuous, often stimulus-driven changes in both structure and function, traditional static network approaches frequently fall short in capturing their dynamic nature. Dynamic network analysis (DNA) addresses this limitation and offers a powerful framework to investigate these evolving relationships. This work focuses on temporal networks, a central paradigm within DNA, as an effective approach for modelling time-resolved changes in biological systems. While DNA has gained traction in domains like social and communication sciences, its integration in biology has been more gradual, hindered by data limitations and the need for domain-specific adaptations. Aimed at supporting researchers, particularly those new to the field, the review offers an integrative overview of the diverse and multidisciplinary landscape of DNA, with a focus on temporal networks in systems biology. I begin by clarifying foundational terminology and concepts, then present a multi-scale perspective spanning microscale (nodes and edges), mesoscale (motifs and communities), and macroscale (global topology) analyses. Finally, I explore analytical strategies and computational tools suited to various research objectives, including methods for detecting structural shifts, assessing network similarity, tracking module evolution, and predictive modelling of future network states.