See if this tutorial will help https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/MultipleComparisonsV6.0...
On 2/14/2020 3:28 PM, Kim, Gwang-Won wrote:
Dear Douglas N. Greve,
Thank you very much for your email.
I ran mri_glmfit without --wis. Then I tried to ran mri_glmfit-sim as following: mri_glmfit-sim –glmdir g2v3.wls –2spaces –perm 1000 1.3 abs.
But there is an error message “ERROR: design matrix is not orthogonal, cannot be use with permutation. If this something you really want to do, run with –perm-force”.
How can I do if I want to solve this problems?
Thanks,
Gwang-Won
Message: 6
Date: Fri, 14 Feb 2020 09:42:28 -0500
From: "Douglas N. Greve" dgreve@mgh.harvard.edu
Subject: Re: [Freesurfer] Cluster-wise correction for multiple
comparisons
To: freesurfer@nmr.mgh.harvard.edu
Message-ID: fd8cd06d-6bcf-a108-4fbe-52080d93cf83@mgh.harvard.edu
Content-Type: text/plain; charset="utf-8"
You don't need to specify a mask when you run mri-glmfigt-sim. When you
ran mri_glmfit, you already specified the mask and that gets stored in
the glmdir and then read by mri_glmfit-sim. Your mri_glmfit-sim command
line look right (but you don't need the --bg 1 unless you really want to
background the process, but using only one process will not speed it up).
the other thing I want to point out is that (pseudo) weighted-least
squares will not be taken into account by the permutation. I would rerun
mri_glmfit without --wls and then run mri_glmfit-sim
On 2/7/2020 3:01 PM, Kim, Gwang-Won wrote:
Hi there,
First, I ran "mri_glmfit" using mask including 4 ROIs(file name:
> 4_ROI.label) as follows: mri_glmfit --glmdir g2v3.wls --y ces.nii.gz
> --wis cesvar.nii.gz --fsgd 2group.fsgd --C group.diff.mtx --lable
> 4_ROI.label? --surface fsaverage lh - --fwhm 5 ?eres-save.
Second, I'm going to process "mri_glmfit-sim" to perform a
> cluster-wise correction for multiple comparisons.
Please let me know how to process "mri_glmfit-sim" using masks (file
> name: 4_ROI.label) .
Is it right? ?mri_glmfit-sim --glmdir g2v3.wls --perm 1000 3 abs
> --2spaces --bg 1?
Thank you,
Gwang-Won
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer