Skip to content

MATLAB code for identifying various linear and nonlinear dynamical systems models from resting state neural time series, and more!

Notifications You must be signed in to change notification settings

enozari/rest-system-id

Repository files navigation

This repository contains the MATLAB codes used for data processing and simulations in the article referenced below. The methods of system identification discussed therein are available under 'System ID Methods', the codes used for the simulation of the linearizing effects of macroscopic neurodynamics and neuroimaging can be found in 'Macroscopic Linearity Simulation', and the piece of code used for generating arbitrarily distinct linear systems with almost identical functional connectivity is placed in 'Generating Systems with Identical FC'. The utility codes are available in 'Utilities'.

Licensing and Reference

The software available in this repository is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. If you use codes from this repository for your work, please provide reference to the article

  • Nozari A, Bertolero M A, Stiso J, Caciagli L, Cornblath E J, He X, Mahadevan A, Pappas G J, and Bassett D S, "Is the brain macroscopically linear? A system identification of resting state dynamics", 2021.

External packages and files:

The following MATLAB packages should be downloaded and installed (added to MATLAB path using the addpath() function) before running the codes in this repository:

Also, the following files need to be downloaded and available to the MATLAB functions (either placed in the same directory or in a directory on the MATLAB path):

Finally, the file main_fmri.m assumes that preprocessed HCP S1200 rsfMRI time series are available in an 'HCP' subdirectory. The data is not included in this distribution but is publicly available at https://db.humanconnectome.org and our preprocessing pipeline is described in the Methods section of the above reference. Similarly, the file main_ieeg.m assumes that preprocessed RAM rsiEEG time series are available in the rs_5min/rand_segments_xxx subdirectory, for xxx = 1, 5, and 25. These are not included in this distribution but are publicly available at http://memory.psych.upenn.edu/RAM (modulo a simple subsampling required for the xxx = 5, 25).

About

MATLAB code for identifying various linear and nonlinear dynamical systems models from resting state neural time series, and more!

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages