The OpenNeuro Derivatives Project is an effort to provide quality control metrics and preprocessed data for all raw datasets on OpenNeuro.
This superdataset is an aggregation of derivative datasets produced by running BIDS Apps on raw datasets from OpenNeuro.
Derivatives currently included, grouped by the tool that produced them:
| Tool | Description |
|---|---|
| MRIQC | Image quality metrics for T1w and BOLD |
| fMRIPrep | Anatomical and functional MRI preprocessing |
| FitLins | Subject- and group-level GLM analyses |
| XCP-D | Post-processing for resting-state fMRI |
The superdataset is intended to be cloned with DataLad:
datalad clone https://github.com/OpenNeuroDerivatives/OpenNeuroDerivatives.git
cd OpenNeuroDerivatives
datalad get ds000001-fmriprep # install and download a full subdataset
datalad get ds000001-fmriprep/sub-01 # fetch one subject's filesEach subdataset has its own README.md describing the tool version, the
methods boilerplate (where applicable), the compute environment, and the
source OpenNeuro dataset.
See each subdataset's README.md, dataset_description.json, and code/ directory
for details on the methods used to generate the derivatives.
Most subdatasets do not have independent DOIs.
To cite them, please cite the source OpenNeuro dataset,
this superdataset, and the BIDS App that produced the derivatives.
Please also consult each subdataset's README and dataset_description.json
for any additional citation instructions.
To cite this aggregation, see CITATION.cff.
This work was funded by the NIH BRAIN Initiative under award R24-MH117179 to Russell A. Poldrack.
Computing for the MRIQC, fMRIPrep, and XCP-D derivatives was performed on the TACC Frontera system under the Pathways allocation. We thank TACC for providing computational resources and support. Please cite Frontera as:
Dan Stanzione, John West, R. Todd Evans, Tommy Minyard, Omar Ghattas, and Dhabaleswar K. Panda. 2020. Frontera: The Evolution of Leadership Computing at the National Science Foundation. In Practice and Experience in Advanced Research Computing (PEARC '20), July 26–30, 2020, Portland, OR, USA. ACM, New York, NY, USA, 11 pages. https://doi.org/10.1145/3311790.3396656
Computing for the FitLins derivatives was performed on the Sherlock cluster. We would like to thank Stanford University and the Stanford Research Computing Center for providing computational resources and support that contributed to these research results.
Released under CC0 1.0 Universal. Subdatasets are also CC0 unless otherwise indicated.
These data are provided on an "as is" basis, and no warranties are expressed or implied pertaining to the quality of the data, or any specific purpose of use. Data consumers are responsible for evaluating the quality of the derivatives for their intended uses.