The Deep Review

Manuscript description

This repository is home to the Deep Review, a review article on deep learning in precision medicine. The Deep Review is collaboratively written on GitHub using a tool called Manubot (see below). The project operates on an open contribution model, welcoming contributions from anyone (see CONTRIBUTING.md or an existing example for more info). To see what's incoming, check the open pull requests. For project discussion and planning see the Issues.

The original version of the Deep Review was published in 2018 and should be cited as:

Ching T, Himmelstein DS, Beaulieu-Jones BK, Kalinin AA, Do BT, Way GP, Ferrero E, Agapow P-M, Zietz M, Hoffman MM, Xie W, Rosen GL, Lengerich BJ, Israeli J, Lanchantin J, Woloszynek S, Carpenter AE, Shrikumar A, Xu J, Cofer EM, Lavender CA, Turaga SC, Alexandari AM, Lu Z, Harris DJ, DeCaprio D, Qi Y, Kundaje A, Peng Y, Wiley LK, Segler MHS, Boca SM, Swamidass SJ, Huang A, Gitter A, and Greene CS. 2018. Opportunities and obstacles for deep learning in biology and medicine. Journal of The Royal Society Interface 15(141):20170387. doi:10.1098/rsif.2017.0387

Current stage: planning Deep Review 2019

As of writing, we are aiming to publish an update of the deep review each year, with the next such release occurring in June 2019. We will continue to make project preprints available on bioRxiv, and aim to continue publishing the finished reviews in a peer-reviewed venue as well. Like the initial release, we are planning for an open and collaborative effort. Please see issue #810 to contribute to the discussion of future plans, and help decide how to best continue this project.

Manubot updates: We recently updated this repository to use the latest Manubot version. Citations must now be semicolon separated like [@doi:10.1002/minf.201501008; @doi:10.1002/jcc.24764] and citation tags are required when the identifier contains forbidden characters. Previously, multiple citations were just separated by whitespace. In addition, we're switching from wrapping text at a character cutoff to "one sentence per line" as described in USAGE.md. Please make sure you base your pull requests off of the latest version of the greenelab:master branch. Keep your fork synced by setting its upstream remote to greenelab and running:

# If your branch only has commits from greenelab:master but is outdated
git pull --ff-only upstream master

# If your branch is outdated and has diverged from greenelab:master
git pull --rebase upstream master

Headline review format

The initial manuscript was a headline review for Journal of the Royal Society Interface on a topic overlapping the computer and life sciences in the area of systems pharmacology. The headline review solicitation states:

A Headline Review is one in a short, targeted series of high-level reviews within a particular topic of a burgeoning research area. We encourage authors to write in a style that opens the door to a broad range of readers working at the physical sciences - life sciences interface. We intend the reviews to address critical developments in an area of cross-disciplinary research and, when possible, to place such research in a broader context. This is not a place for comprehensive literature surveys.

We do encourage you to speculate in an informed way, and to be topical and provocative about the subject without worrying unduly about space, (the provisional target length is 8-12,000 words). Please think of this as an article which will be a landmark in your area, and will come to be considered as a classic paper of the literature.

Inspiration

On August 2, 2016, project maintainer Casey Greene introduced the project and its motivations:

I was recently inspired by Harold Pimentel's crowd-sourced collection of deep learning papers. Instead of having one individual write this, I thought that this invitation provided a wonderful opportunity to take advantage of the wisdom of crowds to bring a team together around this topic.

This repository provides a home for the paper. We'll operate on a pull request model. Anyone whose contributions meet the ICJME standards of authorship will be included as an author on the manuscript. I can't guarantee that it will be accepted, but I look forward to trying this approach out.

On August 5, Deep Review was announced with a tweet.

Manubot

Manubot is a system for writing scholarly manuscripts via GitHub. Manubot automates citations and references, versions manuscripts using git, and enables collaborative writing via GitHub. The Manubot Rootstock repository is a general purpose template for creating new Manubot instances. See USAGE.md for documentation how to write a manuscript.

Please open an issue for questions related to Manubot usage, bug reports, or general inquiries.

Repository directories & files

The directories are as follows:

• content contains the manuscript source, which includes markdown files as well as inputs for citations and references. See USAGE.md for more information.
• output contains the outputs (generated files) from the manubot including the resulting manuscripts. You should not edit these files manually, because they will get overwritten.
• webpage is a directory meant to be rendered as a static webpage for viewing the HTML manuscript.
• build contains commands and tools for building the manuscript.
• ci contains files necessary for deployment via continuous integration. For the CI configuration, see .travis.yml.

Local execution

To run the Manubot locally, install the conda environment as described in build. Then, you can build the manuscript on POSIX systems by running the following commands.

# Activate the manubot conda environment (assumes conda version >= 4.4)
conda activate manubot

# Build the manuscript, saving outputs to the output directory
sh build/build.sh

# At this point, the HTML & PDF outputs will have been created. The remaining
# commands are for serving the webpage to view the HTML manuscript locally.

# Configure the webpage directory
python build/webpage.py

# View the manuscript locally at http://localhost:8000/
cd webpage
python -m http.server

Sometimes it's helpful to monitor the content directory and automatically rebuild the manuscript when a change is detected. The following command, while running, will trigger both the build.sh and webpage.py scripts upon content changes:

sh build/autobuild.sh

Continuous Integration

Whenever a pull request is opened, Travis CI will test whether the changes break the build process to generate a formatted manuscript. The build process aims to detect common errors, such as invalid citations. If your pull request build fails, see the Travis CI logs for the cause of failure and revise your pull request accordingly.

When a commit to the master branch occurs (for example, when a pull request is merged), Travis CI builds the manuscript and writes the results to the gh-pages and output branches. The gh-pages branch uses GitHub Pages to host the following URLs:

• HTML manuscript at https://greenelab.github.io/deep-review/
• PDF manuscript at https://greenelab.github.io/deep-review/manuscript.pdf

For continuous integration configuration details, see .travis.yml.

License

Except when noted otherwise, the entirety of this repository is licensed under a CC BY 4.0 License (LICENSE.md), which allows reuse with attribution. Please attribute by linking to https://github.com/greenelab/deep-review.

Since CC BY is not ideal for code and data, certain repository components are also released under the CC0 1.0 public domain dedication (LICENSE-CC0.md). All files matched by the following glob patterns are dual licensed under CC BY 4.0 and CC0 1.0:

• *.sh
• *.py
• *.yml / *.yaml
• *.json
• *.bib
• *.tsv
• .gitignore

All other files are only available under CC BY 4.0, including:

• *.md
• *.html
• *.pdf
• *.docx

Except for the following files with different licenses:

• build/assets/anchors.js which is released under an MIT License

Please open an issue for any question related to licensing.