Skip to content
This repository has been archived by the owner on Mar 9, 2024. It is now read-only.

neurobagel/old-query-tool

Repository files navigation

Query Tool


The query tool is a Vue application, developed in JavaScript using a variety of tools including Nuxt, Cypress, and BootstrapVue.

Quickstart | Local Installation | Usage | Testing | License

Please refer to our official documentation for more detailed information on how to use the query tool.

Quickstart

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

NOTE: Currently, to access the query tool, you must be connected to the McGill network.

Local Installation

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

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

Install using Docker

To obtain the query tool docker image, simply run

docker pull neurobagel/query_tool:latest

in your terminal. The Docker image includes the most recent build of the query tool and a minimal http server to serve the static tool.

Manual installation

To run the query tool, 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

use node v16.x LTS!

⚠️ The query tool is built with the Nuxt framework and currently depends on Nuxt2. Nuxt2 does not support node versions beyond the v16 LTS (see e.g. this Github issue). If you want to run the tool locally, make sure you are using node v16.x. A good way to manage different node versions is to use the node version manager tool.

Set the Environment Variables

You'll need to set the API_QUERY_URL environment variable required to run the query tool. API_QUERY_URL is the Neurobagel API URL that the query tool uses to send requests to for results. The query tool utilizes nuxt dotenv module for managing environment variables.

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

API_QUERY_URL=http://localhost:8000/

if you're using the remote api, your .env file would look something like this:

API_QUERY_URL=https://api.neurobagel.org/

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

Launch the Query Tool as a Docker container

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

docker run -p 3000:3000 --env-file=.env neurobagel/query_tool:latest

Then you can access the query tool at http://localhost:3000

Note: the query tool is listening on port 3000 inside the docker container, replace port 3000 by the port you would like to expose to the host.

Launch the Query Tool for a manual 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 the (production) build of the application by running the following command:

npm run build && npm run start

You can verify the tool is running once you receive info messages from Nuxt regarding environment, rendering, and what port the tool is running on in your terminal.

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.

You can refer to the neurobagel documentation to see what the outputs of the query tool look like and how they are structured. You can also download the raw example output files here.

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