For one of my participants, whenever I try to compute their MNI transformation, it messes up their fiducials. The AC and PC get moved down several inches. Then when I try to import an atlas that is in MNI space, it fails miserably. Any suggestions would be greatly appreciated.
Computing the MNI transformation using the SPM function may fail in some rare cases.
I don't have any easy replacement option to recommend from Brainstorm.
You could maybe register with an MNI template and reslice this volume directly with SPM12, and then import it in Brainstorm.
Or process it with FreeSurfer, CAT12, BrainSuite or BrainVISA. These tools would rewrite a clean volume that could possibly work.
Refer to the Brainstorm tutorial page for help with these tools.
It turned out to be something I messed up! I was using an mri from our old study. When I used the new sequence, it worked fine. Sorry for the confusion.