ml4h

ml4h is a toolkit for machine learning on clinical data of all kinds including genetics, labs, imaging, clinical notes, and more. The diverse data modalities of biomedicine offer different perspectives on the underlying challenge of understanding human health. For this reason, ml4h is built on a foundation of multimodal multitask modeling, hoping to leverage all available data to help power research and inform clinical care. Our tools help apply clinical research standards to ML models by carefully considering bias and longitudinal outcomes. Our project grew out of efforts at the Broad Institute to make it easy to work with the UK Biobank on the Google Cloud Platform and has since expanded to include proprietary data from academic medical centers. To put cutting-edge AI and ML to use making the world healthier, we're fostering interdisciplinary collaborations across industry and academia. We'd love to work with you too!

ml4h is best described with Five Verbs: Ingest, Tensorize, TensorMap, Model, Evaluate

  • Ingest: collect files onto one system
  • Tensorize: write raw files (XML, DICOM, NIFTI, PNG) into HD5 files
  • TensorMap: tag data (typically from an HD5) with an interpretation and a method for generation
  • ModelFactory: connect TensorMaps with a trainable architectures
  • Evaluate: generate plots that enable domain-driven inspection of models and results

Getting Started

Advanced Topics:

  • Tensorizing Data (going from raw data to arrays suitable for modeling, in ml4h/tensorize/README.md, TENSORIZE.md )

Setting up your local environment

Clone the repo

git clone [email protected]:broadinstitute/ml.git

Setting up your cloud environment (optional; currently only GCP is supported)

Make sure you have installed the Google Cloud SDK (gcloud). With Homebrew, you can use

brew cask install google-cloud-sdk

gcloud config set project your-gcp-project

Conda (Python package manager)

  • Download onto your laptop the Miniconda bash or .pkg installer for Python 3.7 and Mac OS X from here, and run it. If you installed Python via a package manager such as Homebrew, you may want to uninstall that first, to avoid potential conflicts.

  • On your laptop, at the root directory of your ml4h GitHub clone, load the ml4h environment via

    conda env create -f env/ml4h_osx64.yml

    If you get an error, try updating your Conda via

    sudo conda update -n base -c defaults conda

    If you have get an error while installing gmpy, try installing gmp:

    brew install gmp

    The version used at the time of this writing was 4.6.1.

    If you plan to run jupyter locally, you should also (after you have conda activate ml4h, run pip install ~/ml (or wherever you have stored the repo)

  • Activate the environment:

    source activate ml4h

You may now run code on your Terminal, like so

python recipes.py --mode ...

Note that _recipe_s require having the right input files in place and running them without proper inputs will not yield meaningful results.

PyCharm (Python IDE if interested)

  • Install PyCharm either directly from here, or download the Toolbox App and have the app install PyCharm. The latter makes PyCharm upgrades easier. It also allows you to manage your JetBrains IDEs from a single place if you have multiple (e.g. IntelliJ for Java/Scala).
  • Launch PyCharm.
  • (Optional) Import the custom settings as described here.
  • Open the project on PyCharm from the File menu by pointing to where you have your GitHub repo.
  • Next, configure your Python interpreter to use the Conda environment you set up previously:
    • Open Preferences from PyCharm -> Preferences....
    • On the upcoming Preferences window's left-hand side, expand Project: ml4h if it isn't already.
    • Highlight Project Interpreter.
    • On the right-hand side of the window, where it says Project Interpreter, find and select your python binary installed by Conda. It should be a path like ~/conda/miniconda3/envs/ml4h/bin/python where conda is the directory you may have selected when installing Conda.
    • For a test run:
      • Open recipes.py (shortcut Shift+Cmd+N if you imported the custom settings).
      • Right-click on if __name__=='__main__' and select Run recipes.
      • You can specify input arguments by expanding the Parameters text box on the window that can be opened using the menu Run -> Edit Configurations....

Setting up a remote VM

To create a VM without a GPU run:

./scripts/vm_launch/launch_instance.sh ${USER}-cpu

With GPU (not recommended unless you need something beefy and expensive)

./scripts/vm_launch/launch_dl_instance.sh ${USER}-gpu

This will take a few moments to run, after which you will have a VM in the cloud. Remember to shut it off from the command line or console when you are not using it!

Now ssh onto your instance (replace with proper machine name, note that you can also use regular old ssh if you have the external IP provided by the script or if you login from the GCP console)

gcloud --project your-gcp-project compute ssh ${USER}-gpu --zone us-central1-a

Next, clone this repo on your instance (you should copy your github key over to the VM, and/or if you have Two-Factor authentication setup you need to generate an SSH key on your VM and add it to your github settings as described here):

git clone [email protected]:broadinstitute/ml.git

Because we don't know everyone's username, you need to run one more script to make sure that you are added as a docker user and that you have permission to pull down our docker instances from GCP's gcr.io. Run this while you're logged into your VM:

./ml/scripts/vm_launch/run_once.sh

Note that you may see warnings like below, but these are expected:

WARNING: Unable to execute `docker version`: exit status 1
This is expected if `docker` is not installed, or if `dockerd` cannot be reached...
Configuring docker-credential-gcr as a registry-specific credential helper. This is only supported by Docker client versions 1.13+
/home/username/.docker/config.json configured to use this credential helper for GCR registries

You need to log out after that (exit) then ssh back in so everything takes effect.

Finish setting up docker, test out a jupyter notebook

Now let's run a Jupyter notebook. On your VM run:

${HOME}/ml/scripts/jupyter.sh -p 8889

Add a -c if you want a CPU version.

This will start a notebook server on your VM. If you a Docker error like

docker: Error response from daemon: driver failed programming external connectivity on endpoint agitated_joliot (1fa914cb1fe9530f6599092c655b7036c2f9c5b362aa0438711cb2c405f3f354): Bind for 0.0.0.0:8888 failed: port is already allocated.

overwrite the default port (8888) like so

${HOME}/ml/scripts/dl-jupyter.sh 8889

The command also outputs two command lines in red. Copy the line that looks like this:

ssh -i ~/.ssh/google_compute_engine -nNT -L 8888:localhost:8888 <YOUR VM's IP ADDRESS>

Open a terminal on your local machine and paste that command.

If you get a public key error run: gcloud compute config-ssh

Now open a browser on your laptop and go to the URL http://localhost:8888

GitHub

https://github.com/broadinstitute/ml4h