On 9/17/17 5:26 PM, miracle ozzoude wrote:

Sent from my BlackBerry 10 smartphone. *From: *miracle ozzoude miracooloz@gmail.com *Sent: *Sunday, September 17, 2017 11:01 AM *To: *Freesurfer support list *Subject: *Multiple comparison, color maps and contrast

Hello,

I want to run a whole brain cortical thickness analysis and I will like to confirm some steps. a) for correcting for multiple comparison, I want to use both Monte Carlo and permutations in order to compare the difference. Are these the right commands? Monte Carlo : mri_glmfit-sim --glmdir lh.name.glmdir --sim mc-z 5000 2 mc-z.abs.2 --sm-sign abs --cwp 0.05 --overwrite

That looks correct, but you don't need to run it unless you are not doing a whole hemisphere correction.

Permutation: mri_glmfit-sim --glmdir lh.name.glmdir --sim perm 5000 2 permcsd --sim-sign abs --cwp 0.05 --perm-resid --overwrite

That is correct, though you may want to name the csd something more meaningful like perm.abs.2

b) I have 2 groups and a contrast of (1 -1 0 0) representing where group 1> group 2 removing the effect of age. If I correct for multiple comparisons using abs and while visualizing my result the color map for the surviving clusters is blue , this means areas where group 1 < group 2. If the color map for the surviving clusters is red/yellow, this means areas where group 1> group 2.

correct

Thank you. Best, Paul

Sent from my BlackBerry 10 smartphone.

---

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