I am currently trying to compute a connectome using tractography data from MRtrix3 (10M streamlines) and AAL3 atlas parcellations. I registered the AAL3 atlas template into diffusion space, and everything seems ok (registration between 5tt image, T1 image, AAL3 parcellation, and diffusion image). However, for some subjects, I get the following error: “tck2connectome: [WARNING] The following nodes do not have any streamlines assigned:
tck2connectome: [WARNING] 162
tck2connectome: [WARNING] (This may indicate a poor registration)”, and the nodes with missing streamlines are not always the same. In some cases, I verified if registration was indeed poor, and it didn’t seem so. Is there anything else that could cause this error?
This is just a
WARNING message, not an
ERROR message. I don’t think there is anything to worry about since it’s OK to have some regions with no streamline.
To my knowledge, there’s many reasons will reduce the amount and the length of streamlines, which may cause this problem, e.g., cropping at the GM-WM interface, the setting of the maximum angle in degrees between successive steps and the setting of the maximum length of any track.
Thank you for your response!
So would you say that, for example, modifying the parameters of tckgen (e.g. maxlength and cutoff) could possibly solve this warning? Or should I just take note of this warning and proceed with my connectivity analysis?
I would recommend to take a note of it and move on.