You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In #1613, we added a descriptions.tsv file to the Derivatives spec. The file describes different desc-<label> definitions used in a dataset. I believe this general structure can be applied to other entities, especially acq-<label> (acquisitions.tsv), rec-<label> (reconstructions.tsv), proc-<label> (processings.tsv?), and recording-<label> (recordings.tsv).
The text was updated successfully, but these errors were encountered:
I believe this general structure can be applied to other entities
This is in fact, how the idea for descriptions.tsv originated when we discussed in in Copenhagen 2023. There are some minor inconsistencies, like participants.tsv, which should rather have been subjects.tsv (corresponding to the sub entity).
I am generally in favor of adding support like this wherever it is practically meaningful (that is, if we see a concrete use case).
Just an FYI since @tsalo linked me this to read: Our BEP036 is going to propose a change to allow sessions.tsv at the top-level so long as it has the participant_id, the session_id, and a sessions.json describing the "Levels" of session_id. This will allow imaging data sessions and phenotype-only sessions to live together peacefully/in-harmony. 😄
Pinging my fellow BEP-ers here too if they want to read it and/or mention anything else about sessions.tsv.
Your idea
In #1613, we added a
descriptions.tsv
file to the Derivatives spec. The file describes differentdesc-<label>
definitions used in a dataset. I believe this general structure can be applied to other entities, especiallyacq-<label>
(acquisitions.tsv
),rec-<label>
(reconstructions.tsv
),proc-<label>
(processings.tsv
?), andrecording-<label>
(recordings.tsv
).The text was updated successfully, but these errors were encountered: