Hi Elisa, can you send the command line and full terminal output? doug
On 7/4/13 11:50 AM, Elisa Scariati wrote:
Dear freesurfers,
I'm currently making a per-vertex analysis using the --pvr flag of mri_glmfit.
I already made within-group correlations between thickness and lGI correcting for age and gender that worked perfectly.
However, when I tried to make a between groups comparison, I ran into an error and I'm wondering if I did it correctly.
I want to compare the slope of the pvr between my groups, so I will use a DODS design. In my fsgd file I have the following columns: group, age, gender. So, counting my pvr, I should have in my contrast matrix 2*(3+1)=8 columns.
I made my contrast this way : 0 0 0 0 0 0 1 -1
However, when running mri_glmfit, I get the following error:
ERROR: dimension mismatch between X and contrast C22qVScon.txt X has 7 cols, C has 8 cols
As my matrix file has 6 columns corresponding to what would be expected from my fsgd file (two for group, gender and age, that contain the correct values), it seems that there is only one column added for the pvr instead of two.
Do anyone know if I made a mistake or is it not possible to compare pvr between groups this way?
Thank you in advance for your answer, Best regards Elisa
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer