On 10/21/19 10:24 AM, miracle ozzoude wrote:

External Email - Use Caution

Hello Experts,

I do have couple questions about using mri_glmfit and mri_glmfit-sim for this analysis based on the tutorial page. https://surfer.nmr.mgh.harvard.edu/fswiki/PetSurfer  1) Do i have to add any other flags when using mri_glmfit for this analysis? mri_glmfit --y ${results_dir}/all.mgx.subctxgm.mni305.sm05.nii.gz --fsgd $pet_fsgd --C $matrix1 \ --mask ${SUBJECTS_DIR}/fsaverage/mri.2mm/subcort.mask.mgz --glmdir ${results_dir}/sub.pet.B6.glmdir

No, that is fine

2. When correcting for multiple comparisons, should i use --2spaces or

--3spaces?

It depends on what you are doing. If only subortical, then don't include a spaces flag. If you are eventually going to be reporting sub cort, left hemi, and righ hemi, then use --3spaces

3. I ran a paired ttest analysis using age as regressor of no interest

for amyloid uptake. My contrast is  "1 0" based on the tutorial https://surfer.nmr.mgh.harvard.edu/fswiki/PairedAnalysis. How do i interpret the direction of contrast if I get a result from mri_glmfit-sim (see below)? This is because it doesn't have the Max column for surface based analysis # Cluster   Size(n)   Size(mm^3)     MNIX   MNIY    MNIZ        Max    CWP    CWPLow    CWPHi   1        12709      101672.0      26.00  -69.00  -41.00       5.46844  0.01037  0.00798  0.01316  Right-Cerebellum-White-Matter

You get the sign from the input. If you computed timepoint1-timepoint2, then positive means timepoint1>timepoint2

Thank you. best, Paul

---

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