using functional ROI masks as seed for tractography

[mm]

Hi,

Thank you for the support. I am pretty new to imaging and have few doubts.

I plan to run mrtrix csd global tractography for single shell (b=700). and then later use the masks derived from fROI for specific brain areas.

I have few doubts here

1) the masks that i have are in a resampled MNI space (2 mm, size is different as well). The diffusion data is in another MNI space (1 mm, size is different as well)

I wanted to know if I can use the masks directly or need to resample one or the other ?

2) I also have the fROI co-ordinates. Is it better to use the co-ordiantes? but even in this case do i have to consider the resampling issue?

3) How can I deproject the fmri space into the diffusion space. I think both are in volume space? I cannot find any scripts that implement that. 

Thank you 

MM