Hi Batool - Yes, it would be a problem. It'd be similar to having some scans with one spatial resolution and some with another, or administering one questionnaire to some of your subjects and a different one to the others and then trying to compare the answers. If the 16 directions are a subset of the 49 directions, then you can discard the remaining 33 directions from the subjects that have 49.

Best,
a.y

From: freesurfer-bounces@nmr.mgh.harvard.edu [freesurfer-bounces@nmr.mgh.harvard.edu] on behalf of Rizvi, Batool [br2499@cumc.columbia.edu]
Sent: Wednesday, March 29, 2017 8:37 AM
To: freesurfer@nmr.mgh.harvard.edu
Subject: [Freesurfer] TRACULA - subjects with different number of directions

Hi,

We were wondering if it would be an issue if we used a dataset in which we included 5 subjects that have a different number of directions (49 directions) than the rest of the subjects with a standard number (16 directions). Even if the preprocessing steps work, will it be an issue comparing the subjects statistically?

If so, how can we treat the 5 subjects differently, but still include them in our study?

Thank you!

Batool