Hi, Emily-

>> ++ Environment variable = already set to 'SUMA_SUMA_TESSCON_AutoScale'. Value of 'NO' from /home/AD/ebelleau/.sumarc is ignored.
>> To kill such warnings Set AFNI_ENVIRON_WARNINGS to NO

I believe that this is a non-lethal message-- as the second line says, you could turn off these warnings (and it is just a warning). I believe you might just be setting some internal AFNI variable twice within your setup on that computer. However, I'm not entirely sure. That might be annoying, but non-prohibitive to tracking.

>> -- Error SUMA_LoadDsetOntoSO_eng (SUMA_Color.c:10118):
>> Could not find network_file "o.PR_000.niml.tract" on disk for dset /media/larson_buffalo/MRI7/KAward/Conditioning_From_Cluster/K_dti/sub238/dmri/DTI/o.PR_000.niml.dset

>> This is the code that I used:
>> $suma -vol /media/larson_buffalo/MRI7/KAward/Conditioning_From_Cluster/K_dti/sub238/dmri/DTI/o.PR/NET_000_ROI_001_002+orig -vol /media/larson_buffalo/MRI7/KAward/Conditioning_From_Cluster/K_dti/sub238/dmri/DTI/o.PR_000_PAIRMAP+orig -gdset /media/larson_buffalo/MRI7/KAward/Conditioning_From_Cluster/K_dti/sub238/dmri/DTI/o.PR_000.niml.dset -vol /media/larson_buffalo/MRI7/KAward/Conditioning_From_Cluster/K_dti/sub238/DTI/DT_FA+orig

Gotcha. You aren't doing anything wrong at all (assuming that those pathways are correct). The error message is because SUMA is looking for a "*.tract" file-- now, those are only made during DET and MINIP tracking, and you did PROB tracking. Also, you are not even trying to load a "*.tract" file. So why the message???
---> the answer is a diabolical oversight on my part. The "*.niml.dset" file contains all the matrix information calculated by 3dTrackID-- e.g., between what targets tracts were found, and then the properties in those connections (number of tracts, mean FA, stdev of FA, mean MD, etc.). That information can be viewed as a pseudo-correlation matrix in SUMA, which is presumably why you might be loading it in here-- you can toggle between brain views and the matrix view.
----->*However*, you can also view the matrix info as a network of nodes (i.e., centers of your targets) and edges (connections between them), with the properties dispayed as colors on the edges.
------->*In further fact*, Ziad has kindly programmed things so that you can display the edge properties either as straight lines, *or* as tractography tracts that were found. That is useful if you've done DET or MINIP tracking where there *are* linear tracts output in a *.tract file. And when the *.gdset file is written out, it automatically has a little string at the bottom to look for these accompanying tract files.
----------> The error *you're* getting is because there aren't those "*.tract" files output in PROB mode, so it's looking in vain for them. And it's not your fault, it's just the way things are currently-- it shouldn't be looking for these files, since they're not created in PROB mode.

Now, when I load a volume and gdset output by "3dTrackID -mode PROB ...", I *do* get that error message, but I also get SUMA loading up fine and showing me everything-- that is, it's also a non-lethal message. Is that the case for you? This should be the kind of error/warning that is ignorable for the time-being (while also being something I should amend internally in the code).

--pt