DSST internal Neurobagel node and query

[surchs]

Context

The https://cmn.nimh.nih.gov/dsst group wants to explore having an internal Neurobagel endpoint for some of the intramural data, so that local users can search across datasets for participants that fit certain cohort criteria.
This is similar to the https://github.com/neurobagel/project/issues/105 milestone because it is an internal deployment.
Our main goal at this point is making sure our docs and tools meet the needs of an internal deployment - so feedback.
In the mid-term, DSST may be interested in

• read-only federation with our public endpoints
• spinning up a public endpoint for their own data

Why

• Follows Annotation tool: feature complete #35
• Follows Deploy on Release #64
  • This will be the first truly user-run endpoint. So we will have to adopt a clear deployment system that is triggered by some human action (not just push "latest" from main for every PR)
• Part of Grow the amount and usefulness of data #11 and Grow userbase #49
• To get feedback from our target user group, particularly BIDS experts
• So in the future we might add open NIMH DSST data to the neurobagel federation network and make them findable

Outcomes

• DSST has a running internal neurobagel instance (on local hardware)
• Two local (not necessarily local) datasets are annotated and ingested
• intramural query is running and provides access info to local filesystem
• We have feedback for the deployment docs and the query

What

• Check again that the annotation tool creates directly useable data dictionaries for the CLI to use
• Add minimal explanation to the download page annotation_tool#615
• Add logo to all tools #61
• The assessment tool workflow has a short explanation text annotation_tool#613
• add_data_to_graph.sh errors out for older versions of curl api#213
• Rename to digest digest#77

[surchs]

Eric is currently occupied with other projects, we'll wait until the new year to follow up with the implementation steps.

[rmanaem]

Since we decided to stick with the version 2+ of docker compose we need to reach out to Adam or Eric to see if they can update their server on their end to be able to deploy our stack.