This page contains information on how to install ConKit on your local machine or server. A number of different options exist; however, note that the installation via the Python Package Index remains to be the recommended and simplest.

Python Package Index

The easiest way to install ConKit is via easy_install or pip. To do so, simply run the following command on your command line and you are ready to go.

To install using easy_install:

$> easy_install conkit

To install using pip:

$> pip install conkit


The executable scripts are automatically installed in your default bin directory. There is nothing more that you need to do.

Source Code

If you would like to install ConKit from source, download the latest version from the GitHub repository. Then, use the following commands to install ConKit.

$> git clone
$> cd conkit

Once downloaded, you might want to check that all functions are available on your system. Run the following command:

$> python test

If this has completed successfully, you are good to go and you can now install ConKit.

$> python build install

ConKit is now successfully installed in your system’s default Python.


Similarly to the Python Package Index install, the executable scripts are automatically installed.

External software


If you install ConKit via the Python Package Index, the dependencies are automatically installed for you!

Required dependencies

Python 2.7, 3.4, 3.5, or 3.6
Download Python
Numba 0.36.2 (or later)
Download Numba
NumPy 1.8.2 (or later)
Download NumPy
BioPython 1.64 (or later)
Download BioPython
Matplotlib 1.3.1 (or later)
Download matplotlib

Optional dependencies

scikit-learn 0.18 (or later)
Download scikit-learn
Download HHblits
HHblits Database
Download HHblits Database
Download CCMpred