Installation

We highly recommend the using a prebuilt distribution of cmapPy along with a virtual environment (here we demonstrate how to use it with conda).

Option 1 (recommended): Setup pandasGEXpress in a new conda environment

  • (All operating systems; If you haven’t already) install miniconda
    • Download/follow instructions provided here. Unless you have personal preferences/reasons to do so, we recommend installing Miniconda over Anaconda because it’s more lightweight.
    • On the command line, type conda info to verify that conda has been properly instaled on your system. You should see some information about the “current conda install”; if not, your installation didn’t work.

  • (Linux and Mac) Create & activate your cmapPy environment:

    Note. We currently use Python 2.7.11 for our production code (hence its specification); however, other versions of Python 2 should be stable as well. We do not currently support Python 3.

    Step 1

    Python 2: conda create --name my_cmapPy_env -c bioconda python=2.7.11 numpy=1.11.2 pandas=0.20.3 h5py=2.7.0 requests==2.13.0 cmappy

    • -c bionconda tells conda that it should look for packages in the bioconda channel (that’s where cmapPy lives)

    Step 2

    source activate my_cmapPy_env

  • (Windows) Create & activate your cmapPy environment:

    Step 1

    Python 2: conda create --name my_cmapPy_env python=2.7.11 numpy=1.11.2 pandas=0.20.3 h5py=2.7.0 requests==2.13.0

    Step 2

    pip install cmapPy

    source activate my_cmapPy_env

…and then cmapPy (including command line tools) should be available for use.

To update cmapPy in your conda environment (from activate environment): conda update cmappy

Option 2: Install cmapPy from PyPI

  • pip install cmapPy
  • Note: For use of other virtualenvs, we include a requirements.txt file in the cmapPy package that you can use to install the proper versions of depencies.

Option 3: Install as a development environment

A development environment will allow you to use the cmapPy code as it is in a clone of the repository, allowing you to try out changes and modifications you may wish to make.

Follow the instructions for Option 1 or Option 2 above but change the name of the environment to e.g. my_cmapPy_dev_env and do not include cmappy in the list of packages to install (or do not install it with pip), then activate this environment, i.e.:

Python 2.7: conda create --name my_cmapPy_dev_env python=2.7.11 numpy=1.11.2 pandas=0.20.3 h5py=2.7.0 requests==2.13.0

source activate my_cmapPy_dev_env

Clone the cmapPy github repository, cd into the repo’s top-level directory, and run:

$ python setup.py develop
To test your setup, change into a directory outside the repo, run the python interpreter and try:

cd <ELSEWHERE>

$ python

>> import cmapPy.pandasGEXpress.parse_gct as pg