I am trying to run a group analysis on cortical thickness (two groups, no covariates). I have been trying to follow the tutorial (FsTutorialGroupAnalysis) but I run into a problem with the mri_glmfit. My script looks like this:

fspath="/dat/freesurfer"

for hemi in lh rh; do
  echo $hemi
  cd $fspath
 
    mri_glmfit \
        --y ${hemi}.norm_${hemi}_thickness.mgh \
        --fsgd /dat2/maria/scripts/final/group.fsgd dods \
        --C /dat2/maria/scripts/final/group.diff.mtx \
        --surf fsaverage ${hemi} \
        --cortex \
        --glmdir ${hemi}.group.diff.glmdir

done

I think the function is searching for " --y ${hemi}.norm_${hemi}_thickness.mgh" in the main freesurfer directory instead of going into each participant's folder. What am I missing? The first error that i get is:

" ...
FSGD /dat2/maria/scripts/final/group.fsgd
labelmask  /dat/freesurfer/subjects/fsaverage/label/lh.cortex.label
maskinv 0
glmdir lh.group.diff.glmdir
IllCondOK 0
ReScaleX 1
DoFFx 0
Creating output directory lh.group.diff.glmdir
Loading y from /dat/freesurfer/lh.norm_lh_thickness.mgh
error: mghRead(/dat/freesurfer/lh.norm_lh_thickness.mgh, -1): could not open file
   ... done reading.
ERROR: loading y /dat/freesurfer/lh.norm_lh_thickness.mgh

..."

Does it mean that something is wrong with my group.fsgd file?

A sample of this file, just in case:

"

  GroupDescriptorFile 1
  Title ThickBlinds
  Class Class1 plus blue
  Class Class2 circle green
  Input sub-NW01W Class1
  Input sub-NW02W Class1
..."

I will be grateful for your help.

Best,

Maria Czarnecka