Help for Brain/MINDS' GIN server

This GIN🔗 custom server goal is to provide a unified access to all Brain/MINDS🔗 datasets, and it will be enriched whenever new datasets becomes publicly open.

The website allows quick and intuitive datasets browsing/searching and picking individual files for download.

Datasets are optimally accessed through DataLad🔗 command-line interface (cli).

Example of Dataset retrieval (using DataLad cli)

This procedure implies typing the indicated commands at the shell prompt within a terminal emulator.

Prerequisite

You will need DataLad cli installed on your machine (see https://handbook.datalad.org/en/latest/intro/installation.html 🔗)

Step by step procedure

(1) Clone an existing repository

When visiting the dataset's landing page on Brain/MINDS dataportal🔗, you can easily get the DataLad command syntax by a simple click within the "link & command" bar:

Click on the dedicated button (red arrow 1) to copy command in the clippoard, then paste it in your terminal, and execute.

Alternatively, you'll need to type the following command in your terminal (including the dataset's https URL):

⚠️ Important: Do NOT include the .git suffix in the URL, otherwise large files' content can not be downloaded!

datalad install -r https://cau-gin.brainminds.riken.jp/brainminds/BMA-2019

The above command has created a new sub-directory for the dataset.

At this point, the actual content of the dataset's files isn't downloaded yet (hence you didn't have to wait long for the command to finish), but nevertheless it is already possible to list all files contained in the dataset.

As a matter of fact, all heavy files are, for now, replaced by symbolic links (or "pointer files") that serve as placeholders. The actual content can be downloaded at will in a further step, as described below.

(2) Retrieve file(s) content

If you want to retrieve only a specific file, use the command get and specify the path to the desired file:

cd BMA-2019
datalad get bma_2019_exvivo/base_data/sp2_avg_mri_exvivo_t2wi_v1.0.0.nii.gz

The same command get can be used to download the whole content of a directory recursively:

datalad get bma_sp2_cortical_flatmaps/

Hence, retrieving the whole repository content is simple as the following:

datalad get .

(3) Drop whole dataset content

As datasets tend to use a lot of disk space, you may wish to release the used space at some point, without having to completely delete the dataset definition (i.e. go back to the situation when the datalad install was completed)

cd BMA-2019
datalad drop .

FAQ

How do I get the dataset's URL?

The https URL of the dataset is displayed on the main page of the dataset. Click on HTTPS (red arrow 1) to display the https URL. Then, simply click the button right to the URL field (red arrow 2) to copy the URL in the clipboard.

⚠️ Important: Be sure to remove .git suffix from the URL when using it with DataLad cli, otherwise annexed content won't be available (see https://handbook.datalad.org/en/latest/basics/101-139-gin.html#sharing-and-accessing-the-dataset 🔗).

References

DataLad cli commands🔗