I haven't worked with such files before, but they sound like tractography
parcellated connectomes (parcels to parcels).  If you take the ROIs used
and make them into a .dlabel.nii file, you can fake a template parcels by
parcels .pconn.nii file, and then replace its data with the tractography
data.

To convert the ROIs file(s) to labels, use -*-reduce commands with
INDEXMAX, and use -*-math to zero out places where there are no ROIs
(-*-reduce with MAX would work to find these).  You can then use
-*-label-import commands, and -cifti-create-label to make the .dlabel.nii
file (make sure your right and left labels aren't identical strings, add
suffixes like V1_L or such).

Faking a pconn will look a little silly, the way I'd do it is use
-cifti-parcellate to parcellate the dlabel with itself (making a
.plabel.nii), then -cifti-merge to repeat it to make a 2-column version,
then feed that to -cifti-correlation to make the .pconn.nii (you'll get
NaNs, but that isn't important, we just want the parcels by parcels
mappings so we can import the matrix data into cifti).

Finally, you'll need to get the data out of the .asc file into some format
workbench understands.  If it is simple enough, and it has the full square
matrix in it, you may be able to use the -from-text option of
-cifti-convert directly.  If not, you'll probably need to use
matlab/octave/etc to make the data into a simple square matrix and write it
back out as gifti or simple text, and then use the relevant -cifti-convert
option.

HCP-Users is the place to ask questions like this (and has the most
subscribers), the other lists are for other purposes.

Tim


On Thu, Jun 23, 2016 at 9:31 AM, Longchuan Li <[email protected]> wrote:

>
>
> Hi, HCP experts
>
> I searched this topic in the mailing list for a while and seem not have
> found the answer: after I performed seed-to-target analyses in probtrackx2
> using surface-based ROIs, how can I convert the results
> "seed_to_target_ROIs.asc" to a format that is supported by Workbench and
> could be loaded and visualized in WB?
>
> Thank you in advance!
> Longchuan
>
>
>
> _______________________________________________
> HCP-Users mailing list
> [email protected]
> http://lists.humanconnectome.org/mailman/listinfo/hcp-users
>

_______________________________________________
HCP-Users mailing list
[email protected]
http://lists.humanconnectome.org/mailman/listinfo/hcp-users

Reply via email to