Hi John
Unfortunately, you cannot use "~" in the command line call for probtrackx, as it does not get interpreted before being passed to probtrackx.
Instead, you need to use the full path to any of your files starting from "/"
Cheers,
Saad
On 19 Jul 2011, at 22:50, John Sheppard wrote:
> Greetings everyone,
>
> I have DTI data that are already fully processed from a collaborator (through bedpostX) and I am now trying to call probtrackx to perform probabilistic tractography on these data, using as seeds and targets 68 cortical regions defined automatically in Freesurfer space as part of its standard "recon-all" pipeline (the 68 regions are the areas of the aparc atlas).
>
> I have tried following this FSL tutorial to interface probtrackx with Freesurfer label files: http://www.fmrib.ox.ac.uk/fsl/fdt/fdt_surface.html. I was able to follow this pipeline successfully step-by-step up to the invocation of probtrackx, where I encounter error messages.
>
> My command line right now is as follows:
>
> probtrackx -s ~/wong_lab/MDTI1/E07144/SessionDTI_rotate.bedpostX/merged -x ~/wong_lab/network_analysis/MDTI1/DTI/preprocessing/E07144/labelfiles/lh_aparc_seed-000.label --mesh=~/wong_lab/MRI_data/MDTI1/freesurfer/E07144/surf/lh.white.asc --dir=~/wong_lab/network_analysis/MDTI1/DTI/preprocessing/E07144/fdt_paths --xfm=~/wong_lab/network_analysis/MDTI1/DTI/preprocessing/E07144/freesurfer2fa_warp --invxfm=~/wong_lab/network_analysis/MDTI1/DTI/preprocessing/E07144/fa2freesurfer_warp -P 5000 --forcedir --seedref=~/wong_lab/MRI_data/MDTI1/freesurfer/E07144/mri/nifti/brain --mode=seedmask --s2tastext -l -c 0.2 -S 2000 --steplength=0.5 --opd -m ~/wong_lab/MDTI1/E07144/SessionDTI_rotate.bedpostX/nodif_brain_mask
>
> Error returned by probtracx:
>
> Log directory is: ~/wong_lab/network_analysis/MDTI1/DTI/preprocessing/E07144/fdt_paths
> loading mesh files
> error opening file
>
> I've also tried using linear transforms (FLIRT) instead of nonlinear, which results in the same error message:
> probtrackx -x ~/wong_lab/network_analysis/MDTI1/DTI/preprocessing/E07144/labelfiles/lh_aparc_seed-000.label --mesh=~/wong_lab/MRI_data/MDTI1/freesurfer/E07144/surf/lh.white.asc -s ~/wong_lab/MDTI1/E07144/SessionDTI_rotate.bedpostX/merged --dir=~/wong_lab/network_analysis/MDTI1/DTI/preprocessing/E07144/fdt_paths --xfm=~/wong_lab/network_analysis/MDTI1/DTI/preprocessing/E07144/freesurfer2fa.mat -P 5000 --forcedir --seedref=~/wong_lab/MRI_data/MDTI1/freesurfer/E07144/mri/nifti/brain --mode=seedmask --s2tastext -l -c 0.2 -S 2000 --steplength=0.5 --opd -m ~/wong_lab/MDTI1/E07144/SessionDTI_rotate.bedpostX/nodif_brain_mask
>
> Log directory is: ~/wong_lab/network_analysis/MDTI1/DTI/preprocessing/E07144/fdt_paths
> loading mesh files
> error opening file
>
> Regarding the "error opening file" message, I am not sure which file is referred to here. However, I verified the paths and integrity of all the files called in the command line. That is, nii.gz files can be read with mri_info, and the Freesurfer surface and label files can be loaded using Freesurfer.
>
> In case these details are important, I created one FreeSurfer label file for every aparc region defined by Freesurfer (e.g., lh_aparc_seed-000.label, rh_aparc_seed-035.label) using 'mri_annotation2label' (a Freesurfer utility?) to convert "?h.aparc.annot" annotation files from Freesurfer to a set of label files for each subject. Eventually I want to use all of these label files as targets, but I have removed that option from the command line to keep things simple for debugging. Also, since the DTI data from earlier stages of FSL are from an analysis that's already been published, my intuition is that I must be calling the function incorrectly rather than dealing with bad data.
>
> Any advice or suggestions would be greatly appreciated. In case I'm omitting any critical information, please let me know any other details I should provide.
>
> Thank you for your time!
>
> John Sheppard
>
> --
> John Sheppard
> PhD Student
> Watson School of Biological Sciences
> Cold Spring Harbor Laboratory
> [log in to unmask]
> +1 (630) 957-8321 (US)
>
--
Saad Jbabdi
University of Oxford, FMRIB Centre
JR Hospital, Headington, OX3 9DU, UK
(+44)1865-222466 (fax 717)
www.fmrib.ox.ac.uk/~saad
|