Performs second level analysis using the SwE toolbox (Guillaume et al., 2014) for data, derived from a specific longitudinal, intervention study.

1. Connect to compute server and open new linux screen to run analysis and log output.

screen -dmS screenname -L -Logfile filepath/filename.log

2. Start matlab without GUI

matlab -nodisplay -nodesktop

3. In case the folder is not added to your Path for Matlab, change the working directory to the folder.

cd("path/gb_analysis")

4. Before starting the analysis, select preset to run or use select as preset to select configuration manually via a GUI.

help gb_config

5. Start to run the analysis with the seleced preset.
   • It is advised to enable parallel processing.
   • View function function documentation for more details.
   • Internal note when using MPI servers: each model may run between half an hour and several hours.

gb_config(...)

1. Matlab R2021a
2. SPM12 (Ashburner et al., 2014)
3. SwE-toolbox-2.2.2 (Guillaume et al., 2014)

1. Finished preprocessed first level contrasts
2. CSV file containing relevant information for model defintion

Estimation of all possible models that are not estimated yet or where estimation was interrupted before results could be obtained with parallel processing.

gb_config("all", "estimate", true)

• Example csv with simulated data will be uploaded soon!
• Example MRI data cannot referenced/shared before publication

Ashburner, J., Barnes, G., Chen, C., Daunizeau, J., Flandin, G., Friston, K. & Penny, W … (2014). Spm12 manual. Wellcome Trust Centre for Neu-roimaging, London, UK, 2464

Guillaume, B., Hua, X., Thompson, P. M., Waldorp, L., & Nichols, T. E.(2014). Fast and accurate modelling of longitudinal and repeated mea-sures neuroimaging data. NeuroImage, 94, 287–302. doi: 10.1016/j.neuroimage.2014.03.029