This happens because the cached p-values were generated from a simulation that only went up to 30mm FWHM of smoothing. lGI has lots of internal smoothness which often pushes it over this level. You can try not using any smoothing, or, even better, try using permutation instead. https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/MultipleComparisonsV6.0Perm

On 8/3/2021 11:56 AM, Harriet Cornwell wrote:

        External Email - Use Caution        

        External Email - Use Caution        

Hi FreeSurfer team,

 

I am currently trying to run a correlational analysis in FreeSurfer. I am able to run the preproc and mri_glmfit commands successfully, but I am getting error messages when running mri_glmfit-sim for lGI, thickness and area. An example error message is: ERROR: cannot find /usr/local/freesurfer6/freesurfer//average/mult-comp-cor/fsaverage/rh/cortex/fwhm55/abs/th40/mc-z.csd. I have attached the full terminal output so that you can see the variations of this error message that I am getting. Are you able to advise what the issue might be and how I might go about fixing the issue? Just for reference, I am using a smoothing value of 10 for volume, area and thickness, and 5 for lGI. I would also be happy to share my design matrix and scripts if that would make the troubleshooting easier.

 

Thank you in advance for any help you can offer.

 

Best wishes,

 

Harriet Cornwell

PhD Student | Department of Psychology

University of Bath | 10 West 2.08


_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
https://secure-web.cisco.com/1CWMe7DLO904UvEy60601R6EWcFZ0QzAkK7Rp4eKYkRi1qOKUJzgppW4uZ7r-WYuNn9_6Z1anMap3cXuA7w59VvsCr_TPdY8r-o0_ZDn-nHTSExYrPQfNmK9S-fOgVzTaizQ5s8KeMzrt5Ik6Iw2LPObqKVjUbJx7yWUzFybHOGbIcxEi68re3dOhjzvAP5FyN7_PQfmLdwki76zbp0RBUJZI4_2lNhH3IPTzGj8IVfsKGlzoTlNWUIitwWuOcAm6vTlV_Gv_PQHhm4v6Yo3ihw/https%3A%2F%2Fmail.nmr.mgh.harvard.edu%2Fmailman%2Flistinfo%2Ffreesurfer