This is a very rough guide to running the scripts in the tutorial dataset directory. This needs some TLC. Needs to be updated to refer specifically to the tutorial dataset and to include instructions on how to copy the necessary scripts from the LATEST_SCRIPTS directory.
1. Create the necessary experiment directory structure for storing your experimental data.
2. Copy the preprocessing scripts you plan on using.
N.B. THIS HASN'T BEEN UDPATED TO REFLECT THE ACTUAL TUTORIAL DATASET OR THE LATEST SCRIPTS DIRECTORY. I JUST WANTED TO COPY THIS HERE FROM ANOTHER LOCATION THAT WAS BEING DELETED. IGNORE FOR NOW.
Examples are given for the Object_Localizers
1. Create directory for new subject
2. Copy and organize dicom folders with raw data (e.g.,
012_objects_BOLD2X, etc.)
3. Make a COPY of the scripts_defaults dir in your subject's directory
and rename it scripts.
4. Update global_vars.sh to reflect the specific variables for your
subject.
5. From your subject's top-level directory (e.g.,
.../Object_Localizers/$subj/), run prep0_to3d.sh
./scripts/prep0_to3d.sh
This will convert the dicoms to BRIKs, create some directories and
tar-zip the dicom data.
6. From your subject's preprocessing directory, run the remaining
preprocessing scripts:
../scripts/prep1_tscvr.sh (check motion correction and motion
correction parameters plot)
../scripts/prep3_anat.sh (check aligment between anat and epi)
7. ...And the analysis scripts:
../scripts/surf4_glmprep.sh
6a. When you are satisfied that you are done with preprocessing (this
might be after you've looked at the GLM results), tar-gzip the
preprocessing directory to save space:
tar czf preprocessing.tgz preprocessing
You can always unpack it to (re)do more preprocessing.
tar xzf preprocessing.tgz
7. From your subject's analysis directory, run the glm analysis
../scripts/surf5_glmsurf.sh
8. Look at the results and draw your ROIs, etc.