Open oesteban opened 3 months ago
Hi @oesteban: Yes- I was shying away from providing too many different segmentation atlases in templateflow, as we are planning to maintain them in the cerebellar_atlases repro (in different atlas spaces). I assume there will be an increasing number of different maps over the next years. We are of course inspired by the template-flow naming convention to make the repro intuitive. It's your call: We can either port those atlases over, or just provide a link to the cerebellar_atlases repro in the readme file.
It's nice not to have to maintain two different places where the files are stored, so maybe we should only port the most-used atlases (Buckner, Lobular, KIng, Nettekoven32) as deterministic segmentations into template flow and link to our repro for more detailed files?
Joern
https://github.com/DiedrichsenLab/cerebellar_atlases