Abstract:Key PointsUnderstanding the network organization of the brain has been a long-standing challenge for neuroscience. In the past decade, developments in graph theory have provided many new methods for topologically analysing complex networks, some of which have already been translated to the characterization of anatomical and functional brain networks.Anatomical networks at whole-brain and cellular scales in several species consistently demonstrate conservation of wiring costs and small-world topology (high clustering and short path length). Human brain anatomical networks, derived from MRI or diffusion tensor imaging data, have high-degree cortical 'hubs' and modular and hierarchical properties.Functional networks also demonstrate small-world properties at whole-brain and cellular spatial scales. Additionally, complex network properties including small-worldness and the existence of hubs are conserved over different frequency scales in functional MRI and electrophysiological data.Convergent experimental and computational data suggest that there is interdependence in the organization of structural and functional networks. The topology, synchronizability and other dynamic properties of functional networks are strongly affected by small-world and other metrics of structural connectivity. Conversely, over a slower timescale the dynamics can modulate structural network topology.Neuropsychiatric disorders can be thought of as dysconnectivity syndromes, and graph theory has already been used to quantify abnormality of structural and functional network properties in schizophrenia, Alzheimer's disease and other disorders. Graph theory can help us to understand the vulnerability of brain networks to lesions and could in future be used to provide markers of genetic risk for disorders or to measure therapeutic effects of drug treatments on functional networks.The network organization of the brain, as it is beginning to be revealed by graph theory, is compatible with the hypothesis that the brain, perhaps in common with other complex networks, has evolved both to maximize the efficiency of information transfer and to minimize connection cost, at all scales of space and time. Key issues for future work include clarifying the relationship between the brain's network properties and its emergent cognitive behaviours in health and disease.

Extreme-value statistics of brain networks: importance of balanced condition

Critical excitation-inhibition balance in dense neural networks

Emergence of Irregular Activity in Networks of Strongly Coupled Conductance-Based Neurons

A toy model for brain criticality: self-organized excitation/inhibition ratio and the role of network clustering

Balanced Networks of Spiking Neurons with Spatially Dependent Recurrent Connections

Effect of Small-World Connectivity on Fast Sparsely Synchronized Cortical Rhythms

Highly connected neurons spike less frequently in balanced networks

Graph theoretical analysis of complex networks in the brain

Extreme variability in convergence to structural balance in frustrated dynamical systems

Impact of random links on neuronal extreme events

A biophysical model of dynamic balancing of excitation and inhibition in fast oscillatory large-scale networks

Scale-free behavior of weight distributions of connectomes

Balance of excitation and inhibition determines 1/f power spectrum in neuronal networks

Eigenvalue ratio statistics of complex networks: Disorder vs. Randomness

Nonnormal amplification in random balanced neuronal networks

The topology of large Open Connectome networks for the human brain

Consequences of Dale's law on the stability-complexity relationship of random neural networks

The effects of inhibitory and excitatory neurons on the dynamics and control of avalanching neural networks

A statistical model for brain networks inferred from large-scale electrophysiological signals

How Chaotic is the Balanced State?

Complex brain networks: graph theoretical analysis of structural and functional systems