Not sure why you could not load the fsgd, but the other stuff looks correct. Smaller condition numbers are better, with the best being 1 and something like 1000 being pretty bad. DOF=8 is probably correct if you had two groups (which would preclude a one-sample group mean). It should not take more than a minute to run unless you are smoothing or simulating.

doug

Lars M. Rimol wrote:

Hi,
I did a group analysis on 10 subjects (as a test run) following the steps of the updated tutorial. I get no error messages, however, when I load the y.fsgd, there are no data points. These 10 subjects were chosen randomly, solely for the purpose of running through the group analysis procedure, so I don't necessarily expect significant findings but at least there shold be some datapoints, right?

Make_average_subject, mris_preproc and mri_surf2surf went fine without error messages but the mri_glmfit script took less than 1 minute to run, which I think seems a little suspicious. What should "Matrix condition" be? Here, it was 1. And DOF? (It was 8). Is "OneSampleGroupMean 0" a sign that something has gone wrong?

There was a warning message during mri_glmfit, but the fsgdf is in accordance with the tutorial:

[lmr@ip45 ovelse]$ mri_glmfit --y rh.IQ.txt.thickness.10.mgh --fsgd IQ.txt --glmdir rh.IQ.txt.glmdir --pca --surf average rh --C contrast.mat
gdfReadHeader: reading IQ.txt
WARNING: gdfReadV1: no variables on 'Variables' line found
INFO: gd2mtx_method is dods
Reading source surface /home/subjects/average/surf/rh.white
reading group avg surface area 948 cm^2 from file
Reading in average area /home/subjects/average/surf/rh.white.avg.area.mgh
Number of vertices 163842
Number of faces    327680
Total area         78857.140625
AvgVtxArea       0.481300
AvgVtxDist       0.814244
StdVtxDist       0.276731

$Id: mri_glmfit.c,v MailScanner has detected a possible fraud attempt from "1.78.2.5" claiming to be MailScanner has detected a possible fraud attempt from "1.78.2.5" claiming to be MailScanner warning: numerical links are often malicious: 1.78.2.5 2006/03/10 23:27:29 greve Exp $
cwd /home/subjects/
cmdline mri_glmfit --y rh.IQ.txt.thickness.10.mgh --fsgd IQ.txt --glmdir rh.IQ.txt.glmdir --pca --surf averageja rh --C contrast.mat
sysname Linux
hostname ip45.ucsd.edu
machine x86_64
user     lmr
FixVertexAreaFlag = 1
OneSampleGroupMean 0
y    /home/subjects/rh.IQ.txt.thickness.10.mgh
FSGD IQ.txt
glmdir rh.IQ.txt.glmdir
Creating output directory rh.IQ.txt.glmdir
Loading y from /home/subjects/rh.IQ.txt.thickness.10.mgh
Matrix condition is 1
search space = 94836.7
DOF = 8
Starting fit and test
10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Fit completed in 0.0191 minutes
Computing spatial AR1 on surface
Residual: ar1mn=0.995564, ar1std=0.004225, gstd=6.106113, fwhm=14.378798
Writing results
contrast
    maxvox sig=-3.35122 F=32.6872 at index 22070 0 4    seed=1149137562
Computing PCA (10)
mri_glmfit done

------------
This is my FSGDF:

$ cat IQ_fsgdf.txt
GroupDescriptorFile 1
Title Gruppesammenligning
Class lowIQ
Class highIQ
Variables
Input TOP_161 lowIQ
Input TOP_162 lowIQ
Input TOP_163 lowIQ
Input TOP_164 lowIQ
Input TOP_165 lowIQ
Input TOP_166 highIQ
Input TOP_167 highIQ
Input TOP_175 highIQ
Input TOP_188 highIQ
Input TOP_189 highIQ

--
yours,
LMR

_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer

-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu Phone Number: 617-724-2358 Fax: 617-726-7422 In order to help us help you, please follow the steps in: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting