The Neurobagel Query Tool

Neurobagel's query tool is a web interface for searching across a Neurobagel graph based on various subject clinical-demographic and imaging parameters.

The query tool is a React application, developed in TypeScript using a variety of tools including Vite, Cypress, and MUI.

Quickstart

The query tool is hosted at https://query.neurobagel.org/ and interfaces with Neurobagel federation API.

Local Installation

To run the query tool locally, you have two options:

1. Use our docker image
2. Do a manual install from the cloned git repo.

but before proceeding with either you need to set the environment variables.

Mandatory configuration

Environment variable	Type	Required	Default value if not set	Example
NB_API_QUERY_URL	string	Yes	-	https://federate.neurobagel.org/
NB_IS_FEDERATION_API	boolean	No	true	true

NB_API_QUERY_URL

You'll need to set the NB_API_QUERY_URL environment variable required to run the query tool. NB_API_QUERY_URL is the Neurobagel API URL that the query tool uses to send requests to for results.

NB_IS_FEDERATION_API

If the API you'd like to send queries to is not a federation api, you need to set the NB_IS_FEDERATION_API to false as it is true by default.

Set the environment variables

To set environment variables, create a .env file in the root directory and add the environment variables there. If you're running a neurobagel node-API locally on your machine (following the instructions here), your .env file would look something like this:

NB_API_QUERY_URL=http://localhost:8000/
NB_IS_FEDERATION_API=false # For node API

if you're using the remote (in this example federation) api, your .env file would look something like this:

NB_API_QUERY_URL=https://federate.neurobagel.org/

The protocol matters here. If you wish to use the Neurobagel remote API, ensure your NB_API_QUERY_URL uses https instead of http.

Docker installation

To obtain the query tool docker image, simply run the following command in your terminal:

docker pull neurobagel/query_tool:latest

This Docker image includes the latest release of the query tool and a minimal http server to serve the static tool.

To launch the query tool Docker container and pass in the .env file you have created, simply run

docker run -p 5173:5173 --env-file=.env neurobagel/query_tool:latest

Then you can access the query tool at http://localhost:5173

Note: the query tool is listening on port 5173 inside the docker container, replace port 5173 by the port you would like to expose to the host. For example if you'd like to run the tool on port 8000 of your machine you can run the following command:

docker run -p 8000:5173 --env-file=.env neurobagel/query_tool:latest

Manual installation

To install the query tool directly, you'll need node package manager (npm) and Node.js. You can find the instructions on installing npm and node in the official documentation.

Once you have npm and node installed, you'll need to install the dependencies outlined in the package.json file. You can do so by running the following command:

npm install

To launch the tool in developer mode run the following command:

npm run dev

You can also build and then run the tool from (production) build of the application by running the following command:

npm run build && npm run preview

You can verify the tool is running by watching for the` info messages from Vite regarding environment, rendering, and what port the tool is running on in your terminal.

Developer setup

Having installed the dependencies, run the following command to enable husky pre-commit and post-merge hooks:

npx husky install

Usage

To define a cohort, set your inclusion criteria using the following:

• Age: Minimum and/or maximum age (in years) of participant that should be included in the results.
• Sex: Sex of participant that should be included in the results.
• Diagnosis: Diagnosis of participant that should be included in the results
• Healthy control: Whether healthy participants should be included in the results. Once healthy control checkbox is selected, diagnosis field will be disabled since a participant cannot be both a healthy control and have a diagnosis.
• Minimum number of sessions: Minimum number of imaging sessions that participant should have to be included in the results.
• Assessment tool: Non-imaging assessment completed by participant that should be included in the results.
• Modality: Imaging modality of participant scans that should be included in the results.

Once you've defined your criteria, submit them as a query and the query tool will display the results.\ The query tool offers two different TSV files for results:

• Dataset-level results TSV contains: dataset id, dataset name, dataset portal uri, number of matching subjects, and available imaging modalities
• Participant-level results TSV contains: dataset id, subject id, age, sex, diagnosis, assessment, session id, session file path, number of sessions, and imaging modality

The output files can be joined using DatasetID as key.

Downloading query results

For a given query, the query tool offers two kinds of TSV files for results that users can download. At least one dataset matching the query must be selected in the interface in order to download the query results.

Dataset-level results

The dataset-level results TSV describes the datasets that contain subjects matching the user's query. This TSV contains the following columns:

• DatasetID: unique identifier (UUID) for dataset in the graph. Note that this ID is not guaranteed to be persistent across versions of a graph/across graphs, but will always be identical across a pair of query tool result files. This column can be used as the key to join the dataset-level and participant-level results TSVs for a given query result, if needed.
• DatasetName: human readable name of the dataset
• PortalURI: URL to a website or page about the dataset, if available
• NumMatchingSubjects: (aggregate variable) number of subjects matching the query within the dataset
• AvailableImageModalites: (aggregate variable) list of unique imaging modalities available for the dataset

Example:

DatasetID	DatasetName	PortalURI	NumMatchingSubjects	AvailableImageModalities
http://neurobagel.org/vocab/81e47a09-658d-48ea-b7c5-8d54820d038e	BIDS synthetic	https://github.com/bids-standard/bids-examples	5	http://purl.org/nidash/nidm#T1Weighted, http://purl.org/nidash/nidm#FlowWeighted

Participant-level results

The participant-level results TSV contains the available harmonized participant attributes for subject sessions matching the query in each (selected) matching dataset. Each row in the TSV corresponds to a single matching subject session.

This TSV contains the following columns:

• DatasetID: unique identifier (UUID) for dataset in the graph. Note that this ID is not guaranteed to be persistent across versions of a graph/across graphs, but will always be identical across a pair of query tool result files. This column can be used as the key to join the dataset-level and participant-level results TSVs for a given query result, if needed.
• SubjectID: subject label
• Age: subject age, if available
• Sex: subject sex, if available
• Diagnosis: list of diagnoses of subject, if available
• Assessment : list of assessments completed by subject, if available
• SessionID: session label
• SessionPath: the path of the session directory relative to the dataset root (for datasets available through DataLad) or root of the filesystem where the dataset is stored
• NumSessions: (aggregate variable) total number of available sessions for subject. This number will be the same across rows corresponding to the same subject.
• Modality: imaging modalities acquired in the session, if available

Example:

DatasetID	SubjectID	Age	Sex	Diagnosis	Assessment	SessionID	SessionPath	NumSessions	Modality
http://neurobagel.org/vocab/81e47a09-658d-48ea-b7c5-8d54820d038e	sub-01	34.1	http://purl.bioontology.org/ontology/SNOMEDCT/248152002	nan	https://www.cognitiveatlas.org/task/id/4321, https://www.cognitiveatlas.org/task/id/1234	ses-01	/data/neurobagel/bagel-cli/bids-examples/synthetic/sub-01/ses-01	2	http://purl.org/nidash/nidm#FlowWeighted, http://purl.org/nidash/nidm#T1Weighted
http://neurobagel.org/vocab/81e47a09-658d-48ea-b7c5-8d54820d038e	sub-01	34.1	http://purl.bioontology.org/ontology/SNOMEDCT/248152002	nan	https://www.cognitiveatlas.org/task/id/4321, https://www.cognitiveatlas.org/task/id/1234	ses-02	/data/neurobagel/bagel-cli/bids-examples/synthetic/sub-01/ses-02	2	http://purl.org/nidash/nidm#FlowWeighted, http://purl.org/nidash/nidm#T1Weighted
http://neurobagel.org/vocab/81e47a09-658d-48ea-b7c5-8d54820d038e	sub-02	nan	http://purl.bioontology.org/ontology/SNOMEDCT/248153007	http://purl.bioontology.org/ontology/SNOMEDCT/49049000	https://www.cognitiveatlas.org/task/id/4321	ses-01	/data/neurobagel/bagel-cli/bids-examples/synthetic/sub-02/ses-01	2	http://purl.org/nidash/nidm#FlowWeighted, http://purl.org/nidash/nidm#T1Weighted
http://neurobagel.org/vocab/81e47a09-658d-48ea-b7c5-8d54820d038e	sub-02	nan	http://purl.bioontology.org/ontology/SNOMEDCT/248153007	http://purl.bioontology.org/ontology/SNOMEDCT/49049000	https://www.cognitiveatlas.org/task/id/4321	ses-02	/data/neurobagel/bagel-cli/bids-examples/synthetic/sub-02/ses-02	2	http://purl.org/nidash/nidm#FlowWeighted, http://purl.org/nidash/nidm#T1Weighted
http://neurobagel.org/vocab/81e47a09-658d-48ea-b7c5-8d54820d038e	sub-03	22.1	nan	nan	https://www.cognitiveatlas.org/task/id/1234	ses-01	/data/neurobagel/bagel-cli/bids-examples/synthetic/sub-03/ses-01	2	http://purl.org/nidash/nidm#FlowWeighted, http://purl.org/nidash/nidm#T1Weighted
http://neurobagel.org/vocab/81e47a09-658d-48ea-b7c5-8d54820d038e	sub-03	22.1	nan	nan	https://www.cognitiveatlas.org/task/id/1234	ses-02	/data/neurobagel/bagel-cli/bids-examples/synthetic/sub-03/ses-02	2	http://purl.org/nidash/nidm#FlowWeighted, http://purl.org/nidash/nidm#T1Weighted
http://neurobagel.org/vocab/81e47a09-658d-48ea-b7c5-8d54820d038e	sub-04	21.1	http://purl.bioontology.org/ontology/SNOMEDCT/248152002	nan	https://www.cognitiveatlas.org/task/id/4321	ses-01	/data/neurobagel/bagel-cli/bids-examples/synthetic/sub-04/ses-01	2	http://purl.org/nidash/nidm#FlowWeighted, http://purl.org/nidash/nidm#T1Weighted
http://neurobagel.org/vocab/81e47a09-658d-48ea-b7c5-8d54820d038e	sub-04	21.1	http://purl.bioontology.org/ontology/SNOMEDCT/248152002	nan	https://www.cognitiveatlas.org/task/id/4321	ses-02	/data/neurobagel/bagel-cli/bids-examples/synthetic/sub-04/ses-02	2	http://purl.org/nidash/nidm#FlowWeighted, http://purl.org/nidash/nidm#T1Weighted
http://neurobagel.org/vocab/81e47a09-658d-48ea-b7c5-8d54820d038e	sub-05	42.5	http://purl.bioontology.org/ontology/SNOMEDCT/248153007	http://purl.bioontology.org/ontology/SNOMEDCT/49049000	https://www.cognitiveatlas.org/task/id/4321, https://www.cognitiveatlas.org/task/id/1234	ses-01	/data/neurobagel/bagel-cli/bids-examples/synthetic/sub-05/ses-01	2	http://purl.org/nidash/nidm#FlowWeighted, http://purl.org/nidash/nidm#T1Weighted
http://neurobagel.org/vocab/81e47a09-658d-48ea-b7c5-8d54820d038e	sub-05	42.5	http://purl.bioontology.org/ontology/SNOMEDCT/248153007	http://purl.bioontology.org/ontology/SNOMEDCT/49049000	https://www.cognitiveatlas.org/task/id/4321, https://www.cognitiveatlas.org/task/id/1234	ses-02	/data/neurobagel/bagel-cli/bids-examples/synthetic/sub-05/ses-02	2	http://purl.org/nidash/nidm#FlowWeighted, http://purl.org/nidash/nidm#T1Weighted

protected participant-level results in aggregate mode

If the values for all columns except for DatasetID and SessionPath in the participant-level results tsv are set to protected, this indicates the graph being queried has been configured (via its corresponding Neurobagel node API) to return only aggregate information about matches (due to data privacy reasons). This configuration can be modified by setting the NB_RETURN_AGG environment variable to false (the value is by default true). See related section of the documentation here.

Example:

DatasetID	SubjectID	Age	Sex	Diagnosis	Assessment	SessionID	SessionPath	NumSessions	Modality
http://neurobagel.org/vocab/81e47a09-658d-48ea-b7c5-8d54820d038e	protected	protected	protected	protected	protected	protected	protected	protected	protected

Testing

The query tool utilizes Cypress framework for testing.

To run the tests execute the following command:

npx cypress open

License

The query tool is released under the terms of the MIT License