THIS REPO HAS BEEN MOVED TO GITLAB https://gitlab.com/cri-iatlas/iatlas-data.git.

GitLab has removed the ability to pull from external (Github) repos for deployment. This repo has been moved to GitLab to mitigate this issue.

The iAtlas app is an interactive web portal that provides multiple analysis modules to visualize and explore immune response characterizations across cancer types. The app is hosted on shinyapps.io at https://isb-cgc.shinyapps.io/shiny-iatlas/ and can also be accessed via the main CRI iAtlas page at http:https://www.cri-iatlas.org/.

This repository contains the source data for building the database iAtlas uses. The code in this repository consumes the source data, partially-verifies it and builds the database.

This app is spit into two repositories:

• iatlas-app - for all the shiny-app R
• iatlas-data - for all the iatlas-related data, DB-creation and DB-populating code (this coderepository)

Follow the instructions below to create and populate your local database. Then, if you want to also run the iatlas-app, clone that repository and follow it's README to get started.

IMPORTANT: For the smoothest installation, install git-lfs and call git lfs install BEFORE cloning this repository.

If you've already cloned, you may be able to do a git pull after installing git-lfs to fetch the large files (unverified).

• git-lfs: https://git-lfs.github.com

  • Some feather files are very large. git-lfs is used to store these files.

  • For installation on the various platforms, please see this git-lfs wiki

• R: https://www.r-project.org/ - v3.6.2

• RStudio: https://rstudio.com/products/rstudio/download

• Docker: https://www.docker.com/products/docker-desktop

  Ensure that the location of the repository is shared via docker:

  • Mac: https://docs.docker.com/docker-for-mac/#file-sharing

  • Windows: https://docs.microsoft.com/en-us/archive/blogs/stevelasker/configuring-docker-for-windows-volumes

• libpq (postgres): https://www.postgresql.org/download/

• lib cairo: https://www.cairographics.org/ (only required for iAtlas client)

• gfortran (libgfortran): usually installed with gcc

• Download the (very large) RNA Seq Expression file.

  • Download EBPlusPlusAdjustPANCAN_IlluminaHiSeq_RNASeqV2.geneExp.feather and put it in the /feather_files folder
  • TODO: Move this file into Synapse. This file currently lives in Shane Brinkman-Davis's Dropbox ([email protected]).
    The original tsv is found at: https://gdc.cancer.gov/node/905/

• STOP your local postgres server, if you have one running. The scripts in this repository will spin up a postgres server in a docker container. Your local postgres server will shadow it, and the app will consequently connect to the wrong server.

Install package manager: HomeBrew (or MacPorts or your package manager of choice)

Then run these in your shell:

• xcode-select --install
• brew install R
• brew install cairo
• brew install git-lfs
• brew install postgres
• download and install RStudio: https://rstudio.com/products/rstudio/download
• download and install Docker: https://www.docker.com/products/docker-desktop

To build the database locally:

1. Open iatlas-data.Rproj in Rstudio.

2. Follow the instructions.

When built, the database will be available on localhost:5432. The database is called iatlas_dev.

The test suite defined in tests/* unit-tests all the support functions and does an integration test across the whole build process using a subset of the real data.

The primary way to run test is:

devtools::test()

You can run only test-files matching a regex expression:

devtools::test(filter = '_db?_')

You can also run the tests interactively - i.e. have the tests auto re-run whenver you make a change:

testthat::auto_test_package()

You can view code-coverage with:

covr::report()

If control_data/ exists in the root of your project, you'll get control-data valiations automatically when you run build_iatlas_db()

If you are altering code but not data, and you don't expect the data output to change, and you want to test it against the real data... this is section is for you.

Before you make your changes, create a folder in the root of the project called "control_data", then run build_iatlas_db() with the known-good code. This will generate copies of all the output data in the control_data (currently about 400 megabytes of feather files). Then you can make your changes and re-run build_iatlas_db(). The second and subsequence passes will validate their output against the existing data in control_data.

Here are the steps:

1. shell> mkdir control_data
2. R> build_iatlas_db() # builds control data first pass
3. Make changes
4. Re-run, build_iatlas_db(), possibly using resume_at or build_only options

If output mismatches the build will abort with a nice message. It will provide a function in the global namespace to overwrite and update the control_data if the new code is considered correct. Otherwise, you can re-run your code once you fix the discrepancy with build_iatlas_db(resume = "auto")

Note: control_data/ is not checked in with the git repo. To reset your control_data, simply delete all the files in the folder. To disable control-data validation, remove the folder entirely.

Information on the data model can be found in the data_model folder which contains this README.md file.

Information on the data structure can be found in the feather_files folder which contains this README.md markdown file.

Input data for the Shiny-iAtlas portal was accessed from multiple remote sources, including Synapse, the ISB Cancer Genomics Cloud, and Google Drive. The feather files derived from this data and used to populate the database are stored in the feather_files folder. Please see the Feather File README.md for more info on these files.