Managing Python Packages in SAMSON: A Practical Guide for Molecular Modelers

Many computational modelers rely on custom or third-party Python packages for tasks ranging from data analysis to deep learning. But integrating these packages into modeling platforms can be frustrating—especially when the environment doesn’t support pip or editing installations.

If you’re working in SAMSON, you have a convenient and integrated way to manage your Python environment directly within the platform. Whether you need to install widely-used packages like numpy or rdkit, uninstall what you no longer need, or even develop and test your own local packages, SAMSON’s built-in Python Package Manager gives you the tools to do it—all without leaving the application.

Where to Find the Package Manager

You can manage your Python packages by opening the Python Console and going to Edit > Manage packages….

Installing Packages from PyPI

The interface allows you to provide the name of the package you want to install, similar to how you’d do it with pip in a terminal:

• scikit-learn – for machine learning
• biopython – for bioinformatics workflows
• matplotlib – for customized data visualization

Simply type the package name and click “Install.” You can also specify versions or add pip flags, for example:

Installing Local Python Packages

If you’re building your own tools or scripts, you can install local packages by providing the path to the package folder. The only requirement is that it should be a valid Python package (i.e., it contains a setup.py file).

You also have the option to install local packages in editable (aka develop) mode, which is extremely useful if you’re making frequent changes. Editable mode allows the installed package to dynamically reflect updates to the source code without needing a reinstall.

Note: On macOS, editable-mode packages might require additional setup if they don’t load correctly at restart.

Uninstalling Packages

Need to clean up your environment? Just type the name of the package and hit “Uninstall.” That’s it. No command-line gymnastics required.

Why This Matters

If you’re a molecular modeler juggling multiple scripts and custom components, the ability to manage packages *within* the same environment where you visualize, simulate, and analyze can reduce errors, streamline your workflow, and save time.

Moreover, you don’t need system administrator rights, which makes life easier in shared or institutional computing environments.

To learn more, visit the full SAMSON scripting documentation.

SAMSON and all SAMSON Extensions are free for non-commercial use. You can download SAMSON at https://www.samson-connect.net.