Topological structures are consistently overestimated in functional complex networks

Abstract

Functional complex networks have meant a pivotal change in the way we understand complex systems, being the most outstanding one the human brain. These networks have classically been reconstructed using a frequentist approach that, while simple, completely disregards the uncertainty that derives from data finiteness. We provide here an alternative solution based on Bayesian inference, with link weights treated as random variables described by probability distributions, from which ensembles of networks are sampled. By using both statistical and topological considerations, we prove that the role played by links’ uncertainty is equivalent to the introduction of a random rewiring, whose omission leads to a consistent overestimation of topological structures. We further show that this bias is enhanced in short time series, suggesting the existence of a theoretical time resolution limit for obtaining reliable structures. We also propose a simple sampling process for correcting topological values obtained in frequentist networks. We finally validate these concepts through synthetic and real network examples, the latter representing the brain electrical activity of a group of people during a cognitive task.

The Human Organism as an Integrated Interaction Network: Recent Conceptual and Methodological Challenges

“…Clustering coefficient and mean shortest path are oftentimes used to decide upon a network's small-worldness (Bassett and Bullmore, 2006 ), and this property has repeatedly been reported for networks from diverse scientific disciplines. Given the many factors that impact on clustering coefficient and mean shortest path, however, these findings continue to be matter of considerable debate (Bialonski et al, 2010 ; Gastner and Ódor, 2016 ; Hilgetag and Goulas, 2016 ; Papo et al, 2016 ; Hlinka et al, 2017 ; Zanin et al, 2018 ).…”

Uncertainty in Functional Network Representations of Brain Activity of Alcoholic Patients

“…Finally, this has been shown to have important consequences in the observed topological features [27,57].…”

“…Two scenarios are compared, i.e. a standard frequentist approach, in which connectivity is represented as a single number, and a Bayesian one, yielding a complete probability distribution [57]. We firstly recover some previously reported results, i.e.…”

Principles and open questions in functional brain network reconstruction

“…Furthermore, Bayesian inference-unlike frequentist-estimates a full probability model, including hypothesis testing. Disregarding such uncertainty may lead to the detection of wrong topological structures, and specifically to an overestimation of the presence of regularities and non-trivial (i.e., non-random) structures (Zanin, Belkoura, Gomez, Alfaro, & Cano, 2018).…”