Hi Catherine, this usually means that something funny is going on with the contrast matrix. Eg, you made the contrast under windows? If so, then make it again under linux. If not , then send the contrast file doug

On 06/24/2013 09:52 AM, Catherine Bois wrote:

Dear Freesurfer experts, I am currently trying to do a cortical thickness group analysis with one factor, three levels . I have run mri_preproc and mri_surf2surf, and am now trying to run

mri_glmfit --y lh.group.thickness.10B.mgh --fsgd easy2.file --C contrast1.mtx --surf fsaverage lh --cortex --glmdir lh.group

However, I keep getting this error;

INFO: gd2mtx_method is doss Saving design matrix to lh.group/Xg.dat Normalized matrix condition is 1 Matrix condition is 7 Found 149955 points in label. Pruning voxels by thr: 0.000000 Found 149953 voxels in mask Saving mask to lh.group/mask.mgh Reshaping mriglm->mask... search space = 74612.446501 MatrixReadTxT: could not scan value [2][1]

The format of both the fsgd file and the contrast file is ASCII text

My fsgd file is formatted as follows:

GroupDescriptorFile 1 Title Group analysis Class HR Class Con Class ILL Input EHRS_XXXX_2 HR Input EHRS_XXXX_2 HR Input EHRS_XXXX_2 HR Input EHRS_XXXX_2 Con Input EHRS_XXXX_2 Con Input EHRS_XXXX_2 ILL

and my contrast file is;

1 -1 0

I do not understand why I keep getting this error. When I include the flag -no-contrasts-is-ok, the mri_glmfit does work. There are no commas in either file. This leads me to believe something is wrong with my contrast file? However I am not certain what this could be. Help would be greatly appreciated.

Thank you very much,

Cathy