💿 Installation

Via pip (standard)

Getting the last “stable” version

Strongly recommended; although this project is still in beta state.

pip install moldrug

Linux 🐧

Recent Linux distributions require creation of virtual environment before calling pip.
Additionally the default Python might be too new for Moldrug. Python 3.12 is not yet supported (see pull requests #9 and #10).

For example on Ubuntu 24.04 LTS, one needs to perform following steps

1. Get older Python

   sudo add-apt-repository ppa:deadsnakes/ppa
   sudo apt update
   sudo apt install python3.11-venv
   

2. Create a virtual environment

   python3.11 -m venv .venv
   source .venv/bin/activate
   

3. Only then call pip

   pip install moldrug
   

The other Linux distributions might require installation of older Python and creation of virtual environment as well.

Getting the development version

pip install -U git+https://github.com/ale94mleon/moldrug.git@main

Via conda

moldrug is also available through conda. However, the pip installation is the recommended one.

conda create -n moldrug -c ale94mleon -c conda-forge moldrug

Note

MacOS users may face some problems trying to install because of the AutoDock-Vina dependency. If that is the case, just download the executable as it is explained next.

If some dependencies are missing, please install them through pip. Some of them might be:

pip install meeko crem pyyaml scipy tqdm

AutoDock-Vina dependency and related software

AutoDock-Vina is the only non-pip dependency required for moldrug. AutoDock-Vina is only relevant when the built-in cost functions of moldrug from moldrug.fitness are used. AutoDock-Vina is an ongoing project, and it is advisable to stay up-to-date by regularly checking for the latest release.

As of the creation of this documentation, the most recent version is v1.2.5. We will be using this version as a demonstration, but we strongly recommend using the most recent release for optimal performance and features. For detailed information, please visit the AutoDock-Vina installation instruction.

Linux 🐧

wget https://github.com/ccsb-scripps/AutoDock-Vina/releases/download/v1.2.5/vina_1.2.5_linux_x86_64
chmod a+x vina_1.2.5_linux_x86_64
./vina_1.2.5_linux_x86_64 --help

MacOS 🍏

wget https://github.com/ccsb-scripps/AutoDock-Vina/releases/download/v1.2.5/vina_1.2.5_mac_x86_64
chmod a+x vina_1.2.5_mac_x86_64
./vina_1.2.5_mac_x86_64 --help

Note

MacOS users might need to allow the execution of the application on Privacy & Security depending on the MacOS version.

Windows 🪟

Please, download from release. Conda installation may not work.

Converting pdb to pdbqt for the receptor

This step can be achieved through OpenBabel or ADFR. We recommend ADFR. Depending on the platform, you should be able to access the program prepare_receptor in different ways. In my case, it lies on /Users/$USER/ADFRsuite-1.0/bin/prepare_receptor. Then you can convert your pdb with:

/Users/$USER/ADFRsuite-1.0/bin/prepare_receptor -r your_protein.pdb -o your_protein.pdbqt

Check this tutorial for more information. MacOS users may need to use ADFRsuite inside a docker container to avoid Gatekeeper’s restrictions (see this article).

Getting docking box information

To perform docking you must provide boxcenter (center for AutoDock-Vina) and boxsize (size for AutoDock-Vina) to the cost functions defined in moldrug.fitness. For that, two PyMOl plugins are useful: GetBox and/or autodock. Details of their installation and use are not discussed here, please visit their corresponding repositories for more information.

Working with a docker container

• Use the Docker configuration file on GitHub.

• Visit the moldrug’s Docker container.

Finally, pip install moldrug inside the container.

Running tests

pip install -e .[dev]
pytest tests