yiman00 / meg-roi-nets Goto Github PK

A Matlab package for performing leakage-robust network inference between ROIs in MEG data

The methodology used in this pipeline is set out in

Colclough, G. L., Brookes, M., Smith, S. M. and Woolrich, M. W., "A symmetric multivariate leakage correction for MEG connectomes," NeuroImage 117, pp. 439-448 (2015)

This package provides tools for network analysis between regions of interest in source-reconstructured MEG data, analysing amplitude correlations in band-limited power using a Gaussian Markov Random Field network model, after applying a symmetric multivariate orthogonalisation correction for source leakage.

This package was originally developed by @GilesColclough, and is now maintained by the OHBA Analysis group.

• FSL
• FieldTrip
• Matlab Stats + signal processing toolboxes - though you could probably change the code to make it work
• QPAS mex files (optional)

• Source-reconstructed resting-state MEG data (you can probably get it to work for task)
• A set of ROIs or a spatial basis set, in the same space and resolution as the MEG data, saved as a nifti

This package is distributed as part of OSL; this is how it should be installed on your machine.

• The +ROInets folder is a Matlab package. Do not change the name of this folder, and do not add its contents on your path. All the functions inside it can be used with a "dot-syntax", by typing e.g. ROInets.run_network_analysis.
• For a brief summary of what each function does, type help ROInets.Contents. The help text of each function should provide more information, and examples can be found in +ROInets/+examples.
• The top-level function is ROInets.run_individual_network_analysis. View the helptext for this to view all the pipeline options.
• If you often use this package, and would like to avoid typing the prefix ROInets.* all the time, check out import.

Colclough, G. L., Brookes, M., Smith, S. M. and Woolrich, M. W., "A symmetric multivariate leakage correction for MEG connectomes, "NeuroImage 117, pp. 439-448 (2015).

For technical support or any other issues, please either open an issue, or contact the OHBA Analysis Group by email.

• Select time region for analysis
• Band-pass filter the data
• Find a time-course for each ROI using PCA
• Remove effects of leakage using an orthogonalisation process which finds the closest set of orthogonal vectors
• Find down-sampled power envelopes
• Perform network inference using partial correlation, L1 regularised using DP-glasso, optimised using cross-validation
• Convert correlations to z-statistics, by scaling relative to an empirical null, generated to share the same temporal smoothness properties as the input data.

Copyright 2015 OHBA
This program is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.

This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
GNU General Public License for more details.

You should have received a copy of the GNU General Public License
along with this program.  If not, see <http://www.gnu.org/licenses/>.