[pymvpa] pymvpa2-prep-afni-surf wrapper?

[Nick Oosterhof]

> On 12 Feb 2016, at 14:21, Maria Hakonen <maria.hakonen at gmail.com> wrote:
> 
> I would like to try surface searchlight, and for that I need to resample the surfaces to standard topologies (with different resolutions) using MapIcosehedron, align surfaces to a reference functional volume, and merge left and right hemispheres into single surface files.
> However, for me it is unclear what file should I use as a reference functional volume.

It should be a representative volume that is coregistered to the other functional volumes.

If you want to use a functional volume, this is usually an EPI BOLD image. It should have sufficient contrast for coregistration with a T1 anatomical volume. In a typical use case, this is the volume used as a reference volume in the motion correction step, i.e. the volume all other functional volumes are aligned to. 

It should *not* be a normalized volume (containing, for each voxel, percent signal change or z-scores) , because such volumes tend to have very little anatomical information and are not suitable for coregistration. 

Alternatively you can use a T1 anatomical volume that is aligned to the functional data (with the '-a' option).