What is the role of detailed brain network connectome in the emergence of cortex-wide scale-free dynamical patterns?

This collaborative project is in line with worldwide brain-research initiatives into the connectivity and activity of large-scale complex brain cortical networks. Advanced imaging technologies have allowed to obtain both the whole brain connectome structure and cortex-wide complex dynamical patterns, for example for mice, but how the relatively static network connectivity shapes the vast repertoire of dynamics remains an elusive fundamental question. Examining the structure–dynamics relationship is important to elucidate the mechanisms of normal brain function and disorders and to develop brain-inspired computing methods and devices.

Structurally, brain connectomes is a complex network with a hierarchy of modules linked by large-degree hubs and a wide range of connection strength (weight) covering orders of magnitude. Dynamically, cortical activity in intact, awake brain states displays striking scale-free avalanches characterized by a power-law distribution of activation sizes ranging from neuronal spiking to cortex-wide voltage fluctuations.

We propose to study the interaction between local cortical states and the complex network connectome for the emergence of the cortex-wide scale-invariant dynamics patterns. This goal will be pursued through complementary collaborations between the expertise of the PI Dr. C.S. Zhou in the analysis and modeling of brain network connectivity and avalanches and the unique cortex-wide voltage imaging of the Co-I Prof. Thomas Knöpfel at Imperial College London.

We aim to develop a dynamical network framework for data analysis and modeling to link the local cortical states to the cortex-wide scale-invariant patterns via the mouse brain connectome (data from the Allen Institute for Brain Science). From the experimental data, we hope to elucidate how the activity patterns are related to the spatial eigenmodes inherent in the underlying connectome. To obtain deeper insight into the dynamical mechanisms, we will build a large-scale cortical network model by linking biologically plausible excitation-inhibition (E-I)–balanced regional neuronal networks through the brain connectome and investigate how the spatial eigenmodes are activated differently when the local neural circuits operate in supercritical, critical, and subcritical regimes by changing the E-I balance.

Hence, the proposed collaborative study addresses important and timely open issues in the dynamical network principles underlying the large-scale structure-activity relationship in the brain. A framework combining data analysis based on eigenmodes of network structure and activity patterns and the E-I network model would be useful to study the effect of altering local cortical states and network connectivity in neurodegenerative diseases or brain disorders and to search for biomarkers based on large-scale brain network dynamics.