This manual is for aa parallel version 2.2 and earlier. You might prefer the new aa wiki for aa version 3.0 and above or information on cloud execution
Contents
File Structure
Raw data
The data from our scanner, via a dicom gateway, get written in the following structure.
/mridata/cbu/CBU[session number]_CBU[session number]/[sessionstarttime]/Series_[number]_[protocol name]
e.g., /mridata/cbu/CBU060500_CBU060500/20060505_153428/Series_003_CBU_EPI_Standard
Users outside Cambridge will probably find their data are written in a different format, and the protocol names are different. This will require a few changes to the aa code - see AutomaticAnalysisInstallation on how to make these
You may change the location of raw data by changing the aap.directory_conventions.rawdatadir
Processed data
While some control is provided over naming conventions, the file structure for processed data is essentially fixed. For each subject, the format is
[aap.acq_details.root]/[subject name]
e.g., /imaging/rhodri/mystudy/CBU060100
within each subject directory you will typically find the following subdirectories
structurals |
stores processed structural (i.e., hires T1 weighted) images |
realtime_tmaps |
stores converted real time t maps (if present) in nifti format |
[named session] |
one directory for each session, with name as provided for sessions in user script. Contains raw and processed EPIs in NIFTI format |
results |
stores first level statistical model (if used) |
within the study directory you will find
rfx |
second level analysis, if used |
location of aa files
Various aa specific files are written into this directory structure. In study root:
aap_parameters.mat |
the aap structure corresponding to the user script last run on this study |
done_[module name] |
diagnostic_[module name] |
Each module specifies the level at which it should be run (e.g., once per study, per subject, or per session). The done_ and diagnostic_ files for a module will then get written to the appropriate study, subject or session directory.