Skip to content
 
 

Repository files navigation

Networks Gradients Sampling Toolbox

This is a MATLAB toolbox for sampling regional timeseries data with either network or gradient constraints. The toolbox is based on nullspace sampling methods. Although the toolbox is primarily designed with analyiss of neuroimaging data in mind, the methods can be applied to other neural timeseries data.

The toolbox has two main algorithms (functions):

  1. sampling timeseries with preserved intra- or all-network correlations (network_sampler.m).
  2. sampling timeseries with preserved k principal components (gradient_sampler.m).

Getting started

Download or clone this repository into your preferred location. To download, simply press the Code button and select Download ZIP. To clone, enter git clone https://github.com/AdityaNanda/Networks-Gradients-Sampling-Toolbox on the command line.

To add the toolbox to the MATLAB path, use the command addpath(genpath(path_to_toolbox)) or use the Set path button in the Environment section of the Home ribbon, and click add with subfolders. Now you can directly access the relevant functions.

Demo

The scripts demo_networks.m and demo_gradients include demos of the two main functions, and use the enclosed file hcp_1subj.mat, which includes:

  • A parcellated regional timeseries for one resting-state recording of one subject from the Human Connectome Project.
  • A partition of this vector into 17 * 2 = 34 networks (the parcellation and partition are from Schaefer et al. (2018) https://doi.org/10.1093/cercor/bhx179).

Licence

This software is free to use for all academic and research purposes. See the License file for details.

Reference

Nanda, Aditya, and Mikail Rubinov. "Unbiased and efficient sampling of timeseries reveals redundancy of brain network and gradient structure." NeuroImage 274 (2023): 120110.

Nanda A and Rubinov M (2023) Unbiased and Efficient Sampling of Timeseries Reveals Redundancy of Brain Network and Gradient Structure, OSF Preprints, https://doi.org/10.31219/osf.io/ce9xv .

About

A toolbox for generating synthetic timeseries with preserved network structure (as defined on pairwise correlations) or gradient structure (as defined by PCAs).

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages