Search

Fifty years after the publication of Philip Anderson’s landmark essay ‘More is different’ that crystallized the idea of emergence, eight scientists describe the most interesting phenomena that emerge in their fields.

A challenging problem is to identify the most central agents in interconnected multilayer networks. Here, De Domenico et al. present a mathematical framework to calculate centrality in such networks—versatility—and rank nodes accordingly.

Multilayer networks have been used to capture the structure of complex systems with different types of interactions, but often contain redundant information. Here, De Domenico et al. present a method based on quantum information, to identify the minimal configuration of layers to retain.

The topology of interactions is shaping dynamics of complex systems. Here, the authors develop a quantitative method to determine how much higher-order structure can be reduced without affecting dynamical behavior, revealing when higher-order interactions matter.

Information dynamics are key in complex systems. The authors apply graph neural networks to group components for coarse-graining networks, reducing computational complexity. Their approach preserves information flow under compression, being effective for large networks.

Network dismantling allows to find minimum set of units attacking which leads to system’s break down. Grassia et al. propose a deep-learning framework for dismantling of large networks which can be used to quantify the vulnerability of networks and detect early-warning signals of their collapse.

Topological features such as modularity and small-worldness are common in real-world networks. The emergence of such features may be driven by a trade-off between information exchange and response diversity that resembles thermodynamic efficiency.

This Review explains the advances in complexity science for smart cities, showing how the logic of this field can be applied to increasingly complex features of cities.

Network density significantly influences coefficient estimates in functional connectomes, posing challenges for consistent analysis. Here, the authors analyze multimodal neuroimaging data and synthetic networks, revealing that statistical approaches affect metric estimates, especially at varying densities, and caution against over-reliance on thresholding, impacting graph-based studies across neuroscience and related fields.

Network robustness is usually assessed following topological criteria, but disregards the role played by non-topological information. Artime et al. propose a flexible percolation framework that overcomes this limitation and combines both dimensions, offering new ways to protect real systems.

An analysis of news shared on Twitter estimates the level of infodemic risk associated with COVID-19 across countries. Epidemic spread and infodemic risk co-evolve, with reliable information becoming more dominant as infection rates rise locally.

Uno sguardo ai primi innovativi lavori del Premio Nobel, e all'ambiente accademico che li ha resi possibili.

A look at the Nobel laurate’s ground-breaking early work, and the academic environment that made it possible.

Complex biological, social and engineering systems operate through intricate connectivity patterns. Understanding their robustness and resilience against disturbances is crucial for applications. This Review addresses systemic breakdown, cascading failures and potential interventions, highlighting the importance of research at the crossroad of statistical physics and machine learning.

Human mobility data is crucial for many applications, but researchers often rely on single datasets assuming universal validity. Comparing 7 diverse sources across 145 countries, we find significant differences in mobility patterns and networks, impacting applications like epidemic modeling and emphasizing the need for transparent data processing.

Reshaping network theory to describe the multilayered structures of the real world has formed a focus in complex networks research in recent years. Progress in our understanding of dynamical processes is but one of the fruits of this labour.

Describing interdependencies and coupling between complex systems requires tools beyond what the framework of single networks offers. This Review covers recent developments in the study and modelling of multilayer networks.

Theoretical models and structures recovered from measured data serve for analysis of complex networks. The authors discuss here existing gaps between theoretical methods and real-world applied networks, and potential ways to improve the interplay between theory and applications.

Understanding information flow in networks, from regulatory dynamics within cells to epidemic spreading in social networks, is fundamental to predict their behavior. We introduce a multi-pathways temporal distance which encodes the concerted behavior of paths in propagating perturbations and predicts the latent geometry induced by the dynamics.

Geometric insights into the structure and function of complex networks have led to exciting developments in network science. This Review Article summarizes progress in network geometry, its theory, and applications to biological, sociotechnical and other real-world networks.

In this DREAM challenge, 75 methods for the identification of disease-relevant modules from molecular networks are compared and validated with GWAS data. The authors provide practical guidelines for users and establish benchmarks for network analysis.

Characterizing the interactions between viral and human proteins is key to understand the function and structure of viruses such as SARS-CoV-2 and for informing drug design and repurposing strategies. Here, the authors use statistical physics techniques to perform a systematic multiscale comparison of the effects on the human interactome of SARS-CoV-2 with respect to other viruses, and find that COVID-19 exhibits properties typical of systemic diseases.

Real-world networks are typically characterised by a non-trivial organization at the mesoscale, such that groups of nodes are preferentially connected within distinguishable network regions known as communities. In this work the authors define unipartite, bipartite, and multilayer network representations of hypergraph flows to extract the community structure of social and biological systems with higher-order interactions.

The dynamics of information within complex networks can be captured by a set of operators where the effect of detachment of a node defines the node-networks entanglement, which can be used as a multiscale centrality measure. Here, the authors show that the nodes with high entanglement centrality, which are critical for information dynamics, are also the ones responsible for keeping the network integrated.

While the global efficiency measures how easy it is to travel or exchange information concurrently between any two nodes in a network, this might be difficult to compute when networks are not embedded into space and edge weights do not encode physical distances, but instead represent flows. Here, the authors propose and analyse an efficiency measure based on the flow across least resistance pathways that can be computed without any knowledge on the system except for its weighted representation.

The time scales of individual contagion and human mobility are both relevant to understand the evolution of epidemics. Here, the authors study a reaction-diffusion process for an epidemic metapopulation model, showing that modulating the relative time scale of these two processes produces different epidemic outcomes and can reproduce the effect of non-pharmaceutical interventions.

Quantum communication and computing is now in a data-intensive domain where a classical network describing a quantum system seems no longer sufficient to yield a generalization of complex networks methods to the quantum domain. The authors review recent progress into this paradigm shift that drives the creation of a network theory based fundamentally on quantum effects.