Schaefer2018 parcellation in individual surface space : no sphere.reg file

[vferat]

I'm trying to follow this example in order to parcellate an individual brain. However whenI try to run mri_surf2, it raises an error saying it does not find the lh.sphere.reg ( not provided in the repo) Do I have to manually create this file ? If yes can you explain me the different steps to achieve that ?

Actual situation

mri_surf2surf: could not read surface .../fsaverage6/surf/lh.sphere.reg No such file or directory Return code: 255

Steps to reproduce the issue (optional)

Run the following command:

mri_surf2surf 
 --hemi lh
 --tval **...**/PT02/lh.Schaefer2018_1000Parcels_17Networks_order.annot
--sval-annot **...**/fsaverage6/label/lh.Schaefer2018_1000Parcels_17Networks_order.annot --srcsubject fsaverage6
--trgsubject PT02

Thanks for all yur work, Victor

[rubykong]

@vferat This happens if your fsaverage6 and PT02 are not in the $SUBJECTS_DIR. You need to first create a symbolic link for fsaverage6 in your $SUBJECTS_DIR (it should be the folder contain PT02)

ln -s your_freesurfer_dir/freesurfer/5.3.0/subjects/fsaverage6 

[vferat]

Hey @rubykong , thanks for you answer Both subject (PT02 and fsaverage) are in the same folder. I only downloaded the fsaverage6 folder available in this repo, but there is no lh.sphere or lh.sphere.reg here. Do I have to download the full fsaverage6 folder ( this one ? ) and overwrite files with the ones of this repo ?

[rubykong]

@vferat You might need to replace this fsaverage6 folder with the one from freesurfer: freesurfer/5.3.0/subjects/fsaverage6

[vferat]

Perfect @rubykong it's working ! Thanks a lot. 👍

Last question : To use thesurf2surf with the fsaverage6provided , does my subject need to be registered to the average.curvature.filled.buckner40.tif default template ?

[rubykong]

@vferat I am actually not super sure. We normally just run the standard recon-all pipeline. I think you can check with the freesurfer mailing list. You can also load the individual surface mesh (e.g. PT02/surf/lh.inflated) and the mri_surf2surf output (i.e. PT02/lh.Schaefer2018_1000Parcels_17Networks_order.annot) in freeview to visually check if it looks okay.

[vferat]

I ran parcellation using both mris_ca_label(with the .gcs files) and surf2surf for the100 labels 7 networks and fond a 2% difference in voxel labels between both files, so I assume the method works correctly. If I found more information, I'll keep you inform. Thanks again for your help