torchmed / oncotator

Oncotator

License

Oncotator is free for non-profit users. Please see the LICENSE file here for more information.

Package Overview

The name of the directory, oncotator, is also the name of the distribution. This distribution contains the oncotator package.

For more information: http://www.broadinstitute.org/cancer/cga/oncotator

This distribution is the standalone version of Oncotator. If you wish to use the web interface: http://www.broadinstitute.org/oncotator

Please note that the web interface uses an older version of Oncotator and older datasources.

All documentation can be found in the Oncotator forums: http://gatkforums.broadinstitute.org/categories/oncotator

Installation

Currently, Windows is unsupported, though this is due to a dependency, pysam, being unsupported in Windows.

IMPORTANT: You will need root access to your python interpreter or a python virtual environment. More information about virtual environments can be found on the following site: https://pypi.python.org/pypi/virtualenv

As a reminder, virtualenv.py can be run as a standalone script, thereby bypassing superuser requirements. Please see the above link for more details.

Before installing, we recommend installing pyvcf and numpy manually, before attempting the Oncotator install. You may need to prepend each of the following commands with sudo:

$ pip install numpy
$ pip install pyvcf

This distribution is installable through the standard setup.py method. Note that Distribute will be installed as part of the setup process if it isn't already:

$ python setup.py install

Because the setup.py specifies an entry point as a console script, oncotator and initializeDatasource will be installed into your Python's bin/ directory

Unit Test Setup

NOTE: Unit tests require a minimum of 4GB to run.

Before running the unit tests for the first time, please perform the following steps:

1. Execute the following three lines in the same directory as setup.py:

   $ mkdir -p out
   $ ln -s test/configs configs
   $ ln -s test/testdata testdata
   

2. Many unit tests rely on having the standard set of hg19 datasources, which are in a separate download. To point the unit testing framework to your datasources, you must create a personal test config:

   $ cp configs/personal-test.config.template configs/personal-test.config
   In configs/personal-test.config, replace ```dbDir=MY_DB_DIR/``` with ```dbDir=``` the appropriate path to you oncotator datasource directory.
   

Running the Automated Unit Tests (with Virtual Env Creation)

The automated unit tests (run_ci_tests.sh) require 6 GB to run. This can take a fair amount of time (~20 minutes), since a full install into a new virtual environment is performed.

Execute the following line in the same directory as setup.py (provide the appropriate path to the db dir with your datasources):

$ bash run_ci_tests.sh <DB_DIR>

Running the Automated Unit Tests (without Virtual Env Creation)

You can simply run the unit tests in the currently active python environment, which takes a lot less time (< 6 minutes), but requires all dependencies to be installed. However, you must follow the instructions for Unit Test Setup above (Steps 1 and 2), if not already performed. Then run (in the same directory as setup.py):

$ nosetests --all-modules --exe -w test -v --processes=4 --process-timeout=480  --process-restartworker

Please note that there is a known bug with --processes and output to XML. If you alter the above nosetests command to include junit xml (--with-xunit), remove the last three options (`--processes=4 --process-timeout=480 --process-restartworker`). This will cause tests to only run on one core.

Creating a Virtual Environment for Running Oncotator

Follow these steps from the same directory as setup.py. The first command will take several minutes:

bash scripts/create_oncotator_venv.sh <venv_location>
source <venv_location>/bin/activate
python setup.py install

Version Information

Once Oncotator is installed, run it with the -V flag to get version information:

$ Oncotator -V

Git Process Starting with v1.0.0.0 (Developers)

For an overview on the oncotator process for adding features, bugfixes, and general day-to-day branching, please see:: http://nvie.com/posts/a-successful-git-branching-model/

Help

Please post questions, issues, and feature requests in the forum at http://gatkforums.broadinstitute.org/categories/oncotator