Connectomes of phase-synchronization generated by Wilson-Cowan network model
Description
This dataset contains model-generated group-level connectomes of phase-synchronization 1.) between 74 left-hemispheric regions of the Destrieux brain atlas, and 2.) between 74 right-hemispheric regions of the Destrieux brain atlas (Destrieux et al. (2010)). Model dynamics were generated by simulating a model of 74 Wilson-Cowan oscillators (Wilson & Cowan (1972)), with links between oscillators specified by a diffusion MRI (Magnetic Resonance Imaging)-based connectome. Source-reconstructed model dynamics were obtained by projecting simulated data to MEG sensors and then, projecting sensor-level data back to brain regions using subject-specific forward and inverse operators respectively (N=23). Subject-specific source-reconstructed model dynamics were bandpass filtered in the alpha frequency band (8-12 Hz) using Morlet filtering (peak-frequency=9.83 Hz, m=5). Subject-specific matrices of phase-synchronization were then estimated between every pair of left-hemispheric regions (or right-hemispheric regions), using weighted Phase Lag Index (wPLI) (Vinck et al. (2011)). These 23 subject-specific matrices of phase-synchronization were averaged to produce model-generated group-level connectomes of phase-synchronization.
Files
Steps to reproduce
MATLAB code to produce connectomes of phase-synchronization generated by Wilson-Cowan network model, is available from 'megmod' folder at: https://github.com/nitinwilliams/eeg_meg_analysis/. WCnetworkmodel_examplesimulation.m contains example code to produce model-generated connectome of phase-synchronization for a single subject. NOTE: Code requires Brain Dynamics Toolbox (https://github.com/breakspear/bdtoolkit) to be on MATLAB path.