For each new terminal that you open, you must do:
export SUBJECTS_DIR=$TUTORIAL_DATA/diffusion_recons cd $TUTORIAL_DATA/diffusion_tutorial
If you are at an organized course
Running the commands described below will take more time than you have for this tutorial, so please DO NOT RUN these commands right now. All the data has already been processed for you!
There are three steps involved in processing individual subjects with TRACULA:
- Reconstructing white-matter pathways
The three steps must be run in this order, using the command trac-all.
Martinos Center users: Do not submit trac-all as a job on the cluster using pbsubmit. Instead, run trac-all directly on the command line. If run on the cluster, trac-all will submit each subject listed in the configuration file as a seperate job on the cluster. Adding -jobs <filename> to the trac-all command line will process the configuration file, set up the analysis for all subjects included in it, but will not actually run the analysis. Instead, it will write the commands that need to be run into <filename>. The user is then responsible for submitting those as jobs to be run in parallel on a cluster.
Some pre-processing of the diffusion image data is needed before white-matter pathways can be reconstructed by TRACULA. This pre-processing includes:
- Eddy-current compensation
- Computing measures of head motion during the DWI scan
- Intra-subject registration (individual DWI to individual T1)
- Inter-subject registration (individual T1 to a common template space)
Creation of cortical and white-matter masks from FreeSurfer reconstructions
- Tensor fitting for extraction of tensor-based measures (FA, MD, etc)
- Computing anatomical priors for white-matter pathways from the TRACULA atlas
All of the above pre-processing steps can be run by passing the -prep command-line option to trac-all as follows:
Do not run these commands if you're at an organized course.
It will take a while and has already been done for you.
trac-all -prep -c $TUTORIAL_DATA/diffusion_tutorial/dmrirc.tutorial
The argument of the -c command-line option is the configuration file, where preferences for these pre-processing steps can be specified.
Ball-and-stick model fit
TRACULA uses the ball-and-stick model of diffusion to reconstruct the pathways from the DWI data. FSL's bedpostX fits the ball-and-stick model to the DWI data, estimating probability distributions of the parameters of this model at every voxel. bedpostx should only be run on a computer with high computational power, for more information on bedpostx click here. This command will run bedpostX on the pre-processed data of all the subjects specified in the configuration file:
trac-all -bedp -c $TUTORIAL_DATA/diffusion_tutorial/dmrirc.tutorial
Reconstructing white-matter pathways
The final step is to generate the probability distributions for each of the white-matter bundles you specified in the configuration file. This is done by simultaneously fitting the shape of each pathway to the results of the ball-and-stick model of diffusion from above and to the prior knowledge of the pathway anatomy given by the set of manually labeled training subjects in the TRACULA atlas. The following command will reconstruct the probabilistic distribution of the pathways:
trac-all -path -c $TUTORIAL_DATA/diffusion_tutorial/dmrirc.tutorial
TO CHECK THAT ALL FILES WERE GENERATED
See the trac-all page for a list of correct outputs for the -bedp and -path steps.
Now let's look at the results! Click on Next below.