.cfg format

group-analysis definition file

A group-analysis definition file (.cfg) lists the source of data and configuration options for a General Linear Model group-level analysis. Cfg files are text files always containing fieldname/fieldvalues pairs using the syntax:

#fieldname

fieldvalue1

fieldvalue2

...

The group-analysis definition files (*.cfg) may have a combination of the following fields:

#data : source of data

list of nifti files entered into second-level analysis (Nsubjects x Nmeasures) defining one or multiple outcome / dependent measures

note: when entering multiple files per subject (e.g. repeated measures) enter first all files (one per subject) for measure#1, followed by all files for measure#2, etc.

note: NIFTI files may contain 3d volume-level data, fsaverage surface-level data, or ROI-to-ROI data (see conn_surf_write and conn_mtx_write to create surface/matrix nifti files)

note: enter a single 4d file or a filename with wildcards (e.g. vol_*.nii) to simplify the specification of multiple files

alternatively list of SPM.mat files containing first-level analyses (Nsubjects x 1, or Nsubjects x Nmeasures)

alternatively list of folder names containing SPM.mat first-level analyses (Nsubjects x 1, or Nsubjects x Nmeasures)

#design : design matrix

(Neffects x Nsubjects) values defining different explanatory / independent measures or subject-effects

enter one row for each modeled effect

each row should contain one value/number per subject

#contrast_between: between-subjects contrast

between-subjects contrast vector/matrix (Nc1 x Neffects)

#contrast_within : within-subjects contrast

within-subjects contrast vector/matrix (Nc2 x Nmeasures)

#contrast_names : 1st-level contrast labels

(optional, only when entering SPM.mat files in #data field) list of contrast names to select from first-level analysis files (Nmeasures x 1)

#mask : analysis mask

(optional) analysis-mask file

#analysistype : analysis type

(optional) analysis type 1: include both parametric and non-parametric stats; 2: include only parametric stats (Random Field Theory assumptions); 3: include only non-parametric stats (permutation/randomization analyses)

#folder : output folder

(optional) folder where analysis are stored; default current folder

#data_labels : data labels

(optional) labels of columns of data matrix (for display purposes)

#design_labels : design labels

(optional) labels of columns of design matrix (for display purposes)

#design_file : design file

(optional) (alternative to design_matrix field) file containing design_matrix data (.csv/.tsv/.txt/.json/.mat with Nsubjects x Neffects matrix of explanatory / independent measures)

#design_matrix : (alternative format) design matrix

(Nsubjects x Neffects) values defining different explanatory / independent measures or subject-effects

enter one row for each subject

each row should contain one value/number per modeled effect/covariate

analysis.cfg example:

#data

/data/beta_sub*.nii

/data/beta_sub02.nii

/data/beta_sub03.nii

/data/beta_sub04.nii

/data/beta_sub05.nii

/data/beta_sub06.nii

/data/beta_sub07.nii

/data/beta_sub08.nii

/data/beta_sub09.nii

/data/beta_sub10.nii

/data/beta_sub11.nii

/data/beta_sub12.nii

/data/beta_sub13.nii


#design

1 1 1 1 1 1 1 0 0 0 0 0 0

0 0 0 0 0 0 0 1 1 1 1 1 1


#contrast_between

-1 1


#folder

/data/analysis01