Installing Canopy

Back to Canopy front page

Canopy is written in Python and integrates several existing tools for sequence alignment and phylogenetic inference. Canopy is provided as a downloadable package (works on Linux and MacOSX systems) and as a Docker image (works on all major operating systems).

Install Canopy from the Docker image.

  1. Download (and rename) the image:
    docker pull veidenberg/canopy
    docker tag veidenberg/canopy canopy
    • Alternatively, you can build your own image from the Dockerfile in the Canopy source code package. For that, download/unpack the tarball (see step 1 in the guide below) and run the build command:
      docker build -t canopy .
    • The image contains preinstalled Canopy with all the dependencies.
    • Runs on all major operating systems, but requires Docker.
  2. Run the image:
    docker run --rm -it -v "($pwd)/data:/results" canopy
    • The “-v” flag will provide a data folder for Canopy input/output ($pwd = current working directory).
    • Continue with the usage instructions. The example dataset is included with the image (or add your own to the specified data folder). Type “exit” when done.

Install Canopy from the source code.

  • Canopy source code is available form here.
  • Prerequisites:
    • Python version 2.7.x
    • Python library dependencies:
      • Numpy (ver. 1.14.0)
      • Biopython (ver. 1.58)
      • Dendropy (ver. 3.10.0)
    • You can install these libraries manually (using pip) or use the Canopy installer script.

Install Canopy with the installer script.

  • This will install the Canopy module, the main application script and all the library dependencies.
  • You need to have pip (preinstalled in Python 2.7.9+) and setuptools installed first.
    On Ubuntu Linux, these can be installed with:
    sudo apt-get install python-pip
    pip install setuptools
  1. Download Canopy, unpack it, and go to its directory:
    curl -O http://wasabiapp.org/download/canopy/canopy-0.1.5.tar.gz
    tar pxf canopy-0.1.5.tar.gz
    cd canopy-0.1.5/
    

  2. Install Canopy, either globally:
    sudo python setup.py install

    or for a single user:
    python setup.py install --user

  3. Test the new Canopy installation:
    canopy test

    • You should see the following confirmation text:
      Succeed!

      or, if you installed to a Python virtual environment:
      Virtual environment is ready to launch by "source env/bin/activate"

    • If some of the dependencies failed to install, you will be presented with the following options:
      Biopython(==1.58) is not installed
      Dendropy(==3.10.0) is not installed
      Please choose:
        1. exit (Install the requirement libraries manually and then run the script again.)
        2. setup python virtual environment.
      Your choice: [1/2]:
      

      Here, you can select to install the listed libraries manually (option 1), or retry the automatic installation in Python virtual environment (option 2):
      Your choice: [1/2]:2
      Creating virtual environment
      virtualenv env
      Activating...
      source env/bin/activate
      Installing python packages...
      pip install numpy
      pip install biopython==1.58
      pip install dendropy==3.10.0
      

      After this, a virtual Python environment has been set up in ./env directory.
      Enter the environment to use Canopy:
      source env/bin/activate

      To exit the environment:
      deactivate

  4. The installation is now complete. You are ready to run Canopy.