MINDS is a framework designed to integrate multimodal oncology data. It queries and integrates data from multiple sources, including clinical data, genomic data, and imaging data from the NIH NCI CRDC and IDC (Imaging Data Commons) portals.

Installation

Currently the cloud version of MINDS is in closed beta, but, you can still recreate the MINDS database locally. To get the local version of the MINDS database running, you will need to setup a PostgreSQL database and populate it with the MINDS schema. This can be easily done using a docker container. First, you will need to install docker. You can find the installation instructions for your operating system here. Next, you will need to pull the PostgreSQL docker image and run a container with the following command.

docker run -d --name minds -e POSTGRES_PASSWORD=my-secret-pw -e POSTGRES_DB=minds -p port:5432 postgres

Finally, to install the MINDS python package use the following pip command:

pip install med-minds

After installing the package, please create a .env file in the root directory of the project with the following variables:

HOST=127.0.0.1
PORT=5432
DB_USER=postgres
PASSWORD=my-secret-pw
DATABASE=minds   

PostgreSQL Migration (v0.0.6)

Version 0.0.6 introduces a migration from MySQL to PostgreSQL as the database backend, offering:

• Better performance for complex queries
• Advanced data types and indexing options
• More robust transaction support
• Better standards compliance

If you're upgrading from a previous version that used MySQL, please ensure your database environment is updated to use PostgreSQL 12 or later.

Usage

Initial setup and automated updates

If you have locally setup the MINDS database, then you will need to populate it with data. To do this, or to update the database with the latest data, you can use the following command:

# Import the med_minds package
import med_minds

# Update the database with the latest data
med_minds.update()

Querying the MINDS database

The MINDS python package provides a python interface to the MINDS database. You can use this interface to query the database and return the results as a pandas dataframe.

import med_minds

# get a list of all the tables in the database
tables = med_minds.get_tables()

# get a list of all the columns in a table
columns = med_minds.get_columns("clinical")

# Query the database directly
query = "SELECT * FROM clinical WHERE project_project_id = 'TCGA-LUAD' LIMIT 10"
df = med_minds.query(query)

Building the cohort and downloading the data

# Generate a cohort to download from query
query_cohort = med_minds.build_cohort(query=query, output_dir="./data")

# or you can now directly supply a cohort from GDC
gdc_cohort = med_minds.build_cohort(gdc_cohort="cohort_Unsaved_Cohort.2024-02-12.tsv", output_dir="./data")

# to get the cohort details
gdc_cohort.stats()

# to download the data from the cohort to the output directory specified
# you can also specify the number of threads to use and the modalities to exclude or include
gdc_cohort.download(threads=12, exclude=["Slide Image"])

Please cite our work

@Article{s24051634,
    AUTHOR = {Tripathi, Aakash and Waqas, Asim and Venkatesan, Kavya and Yilmaz, Yasin and Rasool, Ghulam},
    TITLE = {Building Flexible, Scalable, and Machine Learning-Ready Multimodal Oncology Datasets},
    JOURNAL = {Sensors},
    VOLUME = {24},
    YEAR = {2024},
    NUMBER = {5},
    ARTICLE-NUMBER = {1634},
    URL = {https://www.mdpi.com/1424-8220/24/5/1634},
    ISSN = {1424-8220},
    DOI = {10.3390/s24051634}
}

Contributing

We welcome contributions from the community. If you would like to contribute to the MINDS project, please read our contributing guidelines.

License

This project is licensed under the MIT License - see the LICENSE file for details.