Hi Federica,

Yes, both of those are perfectly fine. When entering in CONN an ROI that is defined in the volume in subject-space (e.g. a subcortical parcellation file generated by freesurfer) you simply need to provide a version of your functional data that is also defined in the same space (e.g. coregistered to the structural image that you entered into FreeSurfer).

In particular if you have run the default surface-space preprocessing pipeline in CONN, that will have created a "subject-space" version of your functional data that is already coregistered to your T1 image (and it is saved as a secondary dataset labeled 'subject-space') so you can use that functional data to extract the BOLD signal in combination with any volume-level subject-space ROIs. To do so, in the Setup.ROIs tab simply click first on your ROI (e.g. the one containing subject-specific subcortical parcellation FreeSurfer files) and then select the option that reads 'extract from secondary dataset ## (subject-space)'. 

Hope this helps

Alfonso

Originally posted by Federica Toma:

Dear list,

I am performing a surface-based analysis on data previously segmented with FreeSurfer. I have two questions:
1) In surface-based modality, is it possible to do a seed-based analysis using subcortical structures obtained from FreeSurfer?
2) Is it possible to use custom and individual parcellations of thalamic nuclei, obtained from FreeSurfer, to perform a seed-based analysis?

Thank you,

Federica