Dear Freesurfer experts,

I'm having a persistent error with mri_glimfit. The fsgd file isn't being read correctly, despite following the exact requirements of the wiki. Also, the output mentions this "carriage returns warning", and despite running this conversion: cat fsgd_file.fsgd | sed 's/\r/\n/g' > new.fsgd_file.fsgd, the new fsgd still doesn't work (this code somehow alters the Group Descriptor at the top of the file, giving another error).

I've attached my fsgd file (that I used while running the code below)

Also, please see the full error of the glim_fit below:

WARNING: carriage returns have been detected in file fsgd_file.fsgd

Was it created on a Windows computer?

This may cause an error in reading the FSGD file.

If so, try running:

cat fsgd_file.fsgd | sed 's/\r/\n/g' > new.fsgd_file.fsgd

Then use new.fsgd_file.fsgd

INFO: ignoring tag Tile

INFO: ignoring tag nput

INFO: gd2mtx_method is dods

Reading source surface /Applications/freesurfer/BAP/group_analysis_data/qdec/fsaverage/surf/rh.white

Number of vertices 163842

Number of faces 327680

Total area 65020.839844

AvgVtxArea 0.396851

AvgVtxDist 0.717994

StdVtxDist 0.193566

$Id: mri_glmfit.c,v 1.241.2.4 2016/12/08 22:02:40 zkaufman Exp $

cwd /Applications/freesurfer/BAP/group_analysis_data/qdec

cmdline mri_glmfit.bin --y fsgd_RH_thickness.mgh --fsgd fsgd_file.fsgd dods --C contrasts.mtx --surf fsaverage rh --glmdir rhBAP_thickness.glmdir

sysname Darwin

hostname CCASPH-A1900447.cloud.ccas.gwu.edu

machine x86_64

user ajobsaid

FixVertexAreaFlag = 1

UseMaskWithSmoothing 1

OneSampleGroupMean 0

y /Applications/freesurfer/BAP/group_analysis_data/qdec/fsgd_RH_thickness.mgh

logyflag 0

usedti 0

FSGD fsgd_file.fsgd

labelmask /Applications/freesurfer/BAP/group_analysis_data/qdec/fsaverage/label/rh.cortex.label

maskinv 0

glmdir rhBAP_thickness.glmdir

IllCondOK 0

ReScaleX 1

DoFFx 0

Creating output directory rhBAP_thickness.glmdir

Loading y from /Applications/freesurfer/BAP/group_analysis_data/qdec/fsgd_RH_thickness.mgh

... done reading.

INFO: gd2mtx_method is dods

Saving design matrix to rhBAP_thickness.glmdir/Xg.dat

Computing normalized matrix

Normalized matrix condition is 1e+10

Design matrix ------------------

--------------------------------

ERROR: matrix is ill-conditioned or badly scaled, condno = 1e+10

--------------------------------

Possible problem with experimental design:

Check for duplicate entries and/or lack of range of

continuous variables within a class.

If you seek help with this problem, make sure to send:

1. Your command line:

mri_glmfit.bin --y fsgd_RH_thickness.mgh --fsgd fsgd_file.fsgd dods --C contrasts.mtx --surf fsaverage rh --glmdir rhBAP_thickness.glmdir

2. The FSGD file (if using one)

3. And the design matrix above

Attempting to diagnose further

SumSq: Min=99999997952.000000 (col 0), Max=0.000000 (col 0)

... could not determine the cause of the problem

Thanks!

Alex Job