Phantom data is mixed with human data with poor documentation

[jcohenadad]

I had no idea sub4 was spinoza, because it is written nowhere in the README (the README is empty), and the description.tsv file says what sub4 is "homo sapiens":

So, two options:

• the README and description.tsv is fixed
• the spinoza data is uploaded to another dataset

The latter is "cleaner" from a data analysis standpoint (because no need for an "if case") and data curation standpoint (more homogeneous, which is what we expected from a BIDS dataset).

Additional context:

• https://github.com/spinal-cord-7t/coil-qc-code/issues/26#issuecomment-2057883541
• https://github.com/spinal-cord-7t/coil-qc-code/issues/20#issuecomment-2057883222

[Kyota-exe]

@jcohenadad Do you think we can move forward with moving the Spinoza data to another dataset? For all it's worth, I agree that this would be the cleaner solution.

[jcohenadad]

@jcohenadad Do you think we can move forward with moving the Spinoza data to another dataset? For all it's worth, I agree that this would be the cleaner solution.

@evaalonsoortiz is leading this project and should have the final word

[evaalonsoortiz]

Is the proposal to separate the datasets but keep the same data analysis script? The data analysis is not the same for human and phantom data (maybe 40% of it is common to both). If we separate the datasets, then I think I'll separate the scripts as well.

[jcohenadad]

the script would indeed need to be separate

[evaalonsoortiz]

Ok @Kyota-exe let's go ahead and separate both the datasets and scripts

[Kyota-exe]

I have removed the phantom scans from ds005025 (new version v2.2.0) and moved them to ds005090. @jcohenadad and @evaalonsoortiz have been added as admins to the new phantom database.

[evaalonsoortiz]

We now have 2 separate data repos and 2 separate analysis scripts, therefore I will close this issue.