Hi David - Try opening the merged_*.mgz volume in freeview with the -tv option, i.e., a special "tract volume" instead of a generic volume. There is some extra info in the header of that .mgz file (on the colors and the thresholds for displaying each tract). Freeview won't know how to handle all that if it's told that this is a regular volume. (BTW the stats come from the .trk files, so this merged volume is just for display.)

Hope this helps!

Anastasia.
From: freesurfer-bounces@nmr.mgh.harvard.edu <freesurfer-bounces@nmr.mgh.harvard.edu> on behalf of David Ziegler <dziegler@usfca.edu>
Sent: Friday, March 26, 2021 4:08 PM
To: freesurfer@nmr.mgh.harvard.edu <freesurfer@nmr.mgh.harvard.edu>
Subject: [Freesurfer] TRACULA path error
 

        External Email - Use Caution        

Hey Freesurfer folks,

I'm running into a strange error with the TRACULA longitudinal stream, whereby all of the steps run "without error" but the final merged paths look like a collection of single voxels (see screenshot below). The individual paths typically show up as single or a few voxels as well. This seems to be happening on all/most of my subjects.  I've checked the outputs from trac-all -prep and the FA volumes and tensor maps look ok. Also, the individual tract streamlines (the .trk files) look pretty reasonable (see second screenshot). So, I'm assuming that this is an issue at -path step when converting the .trk files to volumes, but am unsure how to troubleshoot it. I'm assuming that the path stats that are output are derived from the volumes and not the streamlines, so this is probably an important step to correct, right?

I'm running freesurfer-linux-centos7_x86_64-7.1.0-20200511-813297b on Ubuntu 18.04.4 LTS and FSL v4 (FSL v5.0 won't work with the trac-all -prep commands for me) and I run bedpost directly on the dmri folder (I have an SGE cluster). Any chance you've seen this error before?

Thanks in advance for any advice!
David

merged path volumes:


several of the streamline files from the same subject: