Inferring hidden connections in complex networks using collective patterns

What is it about?

Many complex systems—such as brain networks, ecosystems, and social structures—have unknown connections. Traditional approaches infer missing links by analyzing pairwise interactions (correlations). Here we present a novel method that focuses on how collective processes evolve across the entire network. We find that emergent patterns, such as Turing patterns in reaction-diffusion systems, encode structural information that can reveal missing links. Tested across different networks, this approach provides a powerful tool for uncovering hidden structures in complex systems, with applications in neuroscience, ecology, and beyond.

Featured Image

Photo by Alina Grubnyak on Unsplash

Photo by Alina Grubnyak on Unsplash

Why is it important?

We formulate a new paradigm of network inference by evaluating data as self-organized collective patterns, highlighting how emergent structures can reveal hidden network properties, addressing key challenges in network science. We illustrate this approach for Turing patterns in reaction-diffusion dynamics on graphs, a case where patterns are related to eigenvectors of the graph’s Laplacian matrix. This approach is particularly timely given the growing interest in Turing patterns on networks and their role in biological, ecological, and technological systems.