Hi Shige, those are fine, but there is a warning here about performing the mri_surf2surf step. You need to make sure that you are doing this properly because it will run and complete even if you are not doing it properly:).

to map the left hemi, run something like: mri_surf2surf --srcsubject subject --srchemi lh --srcsurfreg fsaverage_sym.sphere.reg --trgsubject fsaverage_sym --trghemi lh --trgsurfreg sphere.reg --tval lh.subject.yourdata.mgh --sval subject/surf/lh.yourdata.mgh

to map the right hemi: mri_surf2surf --srcsubject subject/xhemi --srchemi lh --srcsurfreg fsaverage_sym.sphere.reg --trgsubject fsaverage_sym --trghemi lh --trgsurfreg sphere.reg --tval rh.subject.yourdata.mgh --sval subject/surf/rh.yourdata.mgh

Note that in the 2nd call, you specify "subject/xhemi" as the subject and still continue to use "lh" eventhough you are mapping the rh. Like I said, it is tricky!

doug

On 12/06/2012 01:42 PM, Shigetoshi Takaya wrote:

Thanks for your reply, Doug

After inter-hemispheric registration onto fsaverage_sym using surfreg, I did not use mris_preproc in my analysis because of the following two reasons: (i) I thought I can't use "--meas" because the input is an arbitrary surface data. (ii) I don't want to use "--paired-diff" because I don't want the input "pruned" before taking the difference.

Instead, I took the following steps: (1) resample the data onto fsaverage_sym using mri_surf2surf (2) take the difference in each subject using fscalc $fscalc lh.lh_subj1.mgz -sub lh.rh_subj1.mgz --o lh.lh-rh_subj1.mgz (3) concatenate the subjects together $mri_concat lh-rh_subj1 lh-rh_subj2 .... --o (4) smooth (mris_fwhm) and analyze (mri_glmfit).

Are these steps not good for mri_glmfit-sim?

Best, Shige

-----Original Message----- From: Douglas N Greve [mailto:greve@nmr.mgh.harvard.edu] Sent: Thursday, December 06, 2012 11:59 AM To: Shigetoshi Takaya Cc: freesurfer@nmr.mgh.harvard.edu Subject: Re: mri_glmfit-sim for interhemispheric comparison

Shige, it is looking for tables of cluster p-values (the csd files). When you followed the wiki instructions, did you tell it to use fsaverage_sym as the target subject? You must have used a subject called "average".

doug

On 12/06/2012 11:53 AM, Shigetoshi Takaya wrote:

...

Hi Doug,

I did surface-based inter-hemispheric comparison.

I used an arbitrary surface data (.mgz) resampled from a volume data (.nii), otherwise I follow the instruction.

http://surfer.nmr.mgh.harvard.edu/fswiki/Xhemi

I got a good result for mri_glmfit. But when I run mri_glmfit-sim, the following error messeage popped up.

ERROR: cannot find

$FRESURFER_HOME/average/mult-comp-cor/fsaverage_sym/lh/cortex/fwhm00/a bs/th20/mc-z.csd

I did not find "fwhm00"anywhere in the downloaded file "mult-comp-cor", although I found fwhm01-30 directories. Am I missing somethine?

Thank you very much in advance,

Shige

-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu Phone Number: 617-724-2358 Fax: 617-726-7422

Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html Outgoing: ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/

Hi Inge, below is a way to do this. But you should only use this for overlay and not try to do statistics on the results because the results will be biased because you've done something to the LH that you have not done to the RH. You will need two new programs (mris_left_right_register and mris_apply_reg) which I have put here:

ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/mris_left_right_register.centos5 ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/mris_apply_reg.centos5

Those are versions for cenos5; let me know if you need a different platorm. mris_left_right_register may take an hour or so to run.

doug

cd $SUBJECTS_DIR/fsaverage/surf mris_curvature -thresh .999 -n -a 5 -w -distances 10 10 lh.inflated mris_curvature -thresh .999 -n -a 5 -w -distances 10 10 rh.inflated

mris_left_right_register lh.sphere rh.sphere lh.sphere.left_right rh.sphere.left_right mris_apply_reg --src lh.yourfile.mgh --trg rh.lh.yourfile.mgh --stfreg lh.sphere.left_right rh.sphere.left_right

On 12/11/2012 05:39 AM, Inge Kasbohm Amlien wrote:

Hi list, I have a rh overlay (from caret) registered to fsaverage rh. I need it on fsaverage left hemi. Is there an easy way to flip an overlay?

Cheers Inge