[DAT]: UCLA

[kaurao]

Which dataset is it?

UCLA, already preprocessed and covers multiple diagnosis

Implementation

Did not implement anything

Your implementation

No response

Dataset access restrictions

• [X] Public and open access (no registration)
• [ ] Public and open access (registration required)
• [ ] Restricted access (needs approuval)
• [ ] Available only in specific locations (Juseless, Jureca)

Anything else to say?

https://openneuro.org/datasets/ds000030/versions/1.0.0

[fraimondo]

Will be implemented for release

[LeSasse]

I would be happy to look into it, if no one else is workimg on it yet?

[fraimondo]

I would be happy to look into it, if no one else is workimg on it yet?

@LeSasse all yours!

[LeSasse]

Data available on openneuro (https://openneuro.org/datasets/ds000030/versions/1.0.0) is unfortunately version 1.0.0 which has no fmriprep derivatives that I would ideally like to use, so I found the old openfmri dataset hosted via datalad here: http://datasets.datalad.org/?dir=/openfmri/ds000030 Unfortunately, I think this one is version 1.0.3 which still does not contain the fmriprep derivatives. From what I can tell from the openneuro website, the fmriprep derivatives are contained in there from version 1.0.4 onwards and are available in an S3 bucket here: (https://legacy.openfmri.org/s3-browser/?prefix=ds000030/ds000030_R1.0.5/uncompressed/). I think with these data we can not use datalad.

Few options:

1. Somehow get an updated datalad dataset (remote or local)
2. make a class that can get S3 data
3. fmriprep data ourselves
4. leave the data out for now

[LeSasse]

@fraimondo @kaurao During the datalad office hour I was told that the openneuro people are working on fmriprepping all the different datasets again, so Adina will let us know once our particular UCLA dataset is done. I suggest that we set up the datagrabber using the local data that @dlarabi has available in the meantime, and once the openneuro dataset is available we can integrate it using the remote dataset.