use case 2.1 - computing the mean age

SIESTA-eu / wp15

work package 15, use case 2

0 stars 2 forks source link

use case 2.1 - computing the mean age #6

Closed robertoostenveld closed 6 months ago

robertoostenveld commented 7 months ago

This is the first very simple use case. It should be easy to reproduce, and the structure shoudl serve as example for the ones that are going to follow.

robertoostenveld commented 7 months ago

@schoffelen I have just updated the README a bit. Can you document the following extensions to the iteration process as identified by @jdelacueva?

Check the input dataset provenance
Check if the input data has a license. If so, include the license and consider how that affects the output data (for example, the same license applies to output as input data).
Check if the data contains personal data and document this
Consider licenses for the software that the pipeline depends on (for example, FSL, FreeSurfer, MATLAB)

I am not yet sure how the following is to be implemented.

Apply legal license to the output data. Include CreDIT contributors.

Should the pipeline write a LICENSE file to the output directory? That could be implemented in the README document at the moment that the output directory is created.

robertoostenveld commented 7 months ago

@marcelzwiers there are also these two steps in the iteration process:

Design and implement a way to generate pseudodata, specific for the relevant data features that are needed by the analysis pipeline in the previous step and that need pseudonymisation
Convert original data into pseudodata

Can you add a section "Pseudo data" to the README that documents how the pseudo version of the data is to be constructed using the code you shared in https://github.com/SIESTA-eu/wp15/tree/main/generative-BIDS? I suggest to add this section between the "Input data" and "Output data" headings.

robertoostenveld commented 7 months ago

Can you add a section "Pseudo data" ...

Marcel and I discussed this in person; it has now been added.

schoffelen commented 7 months ago

Check the input dataset provenance

I referenced the associated publication that describes the dataset.

Check if the input data has a license. If so, include the license and consider how that affects the output data (for example, the same license applies to output as input data).

The data has been shared under the CC0 license.

Check if the data contains personal data and document this

The dataset contains indirect personal identifiers: age and height, in the participants.tsv file

Consider licenses for the software that the pipeline depends on (for example, FSL, FreeSurfer, MATLAB)

The software has been released under the GPL2|GPL3 licenses (this is how it is mentioned on the R-package (and associated packages) website.

robertoostenveld commented 6 months ago

This seems to be complete, so let's close