Unofficial PyMOL(TM) Python wheel files

Supported Platforms

This repository offers unofficial binary wheels for the open-source version of PyMOL(TM) for all major operating systems. These wheel files are intended for use in a Python project. If you are searching for PyMOL as ready-to-use application go to this repository: pymol-open-source-setup.

Contents of this document

• About PyMOL
• Contents of this repository
  • Scripts
  • Pre-built
  • Vcpkg
• Build wheel file
• From source
  • Prerequisites
  • Step-by-step guide
• License
• Acknowledgements

About PyMOL

PyMOL™ is a powerful visualization software for rendering and animating 3D molecular structures. PyMOL is a trademark of Schrödinger, LLC.

Please note that the files provided here are unofficial. They are informal, unrecognized, and unsupported, offered for testing and evaluation purposes only. No warranty or liability is provided, and the software is made available "as-is."

Contents of this repository

Scripts

There is one Python script that is used for building PyMOL specific shader files.

Pre-built

To enable the direct build of a ready-to-use wheel file, a pre-built _cmd module is already compiled and will be used in the wheel build process if no self built one can be found. Furthermore, the libGL library is also provided to make the build process easier and less error-prone.

Vcpkg

To make the build process of the _cmd module as easy and convenient as possible, the vcpkg package manager is used. The package manager should handle all build dependencies PyMOL needs.

Install

To install PyMOL run:

pip install pymol-open-source-whl

Build wheel file

To build the wheel file (based on the provided pre-built _cmd module), follow these steps (the working directory is the repository root directory):

1. Install uv and use it to create the locked project environment.

2. Sync the development dependencies:

   uv sync --locked
   

3. Set up the build environment by running:

   ./pymake.bat setup_dev_env  # Windows
   ./pymake.sh setup_dev_env   # macOS/Linux
   

4. Build the wheel file using the pre-built _cmd module

   ./pymake.bat build_wheel     # Windows
   ./pymake.sh build_wheel      # macOS/Linux
   

   After a successful build the wheel file can be found under the dist/ folder.

For a platform-specific native wheel build, use cibuildwheel through PyMake:

./pymake.bat build_wheels platform=windows archs=AMD64  # Windows
./pymake.sh build_wheels                            # macOS/Linux

Each cibuildwheel build installs the resulting wheel in an isolated environment and runs the configured PyMOL test suite before producing the artifact.

To run the packaged test suite locally, use:

./pymake.bat test  # Windows
./pymake.sh test   # macOS/Linux

On a fresh checkout this generates sources, prepares vcpkg, installs the native extension editable, and then runs the tests under tests/.

From source

The following information is about building only the _cmd module from source.

Prerequisites

• CMake
  • Be aware: Add the cmake to your PATH variable. Check by running cmake --version
• Clang & Clang++:
  • Xcode developer tools

Step-by-step guide

1. Load the CMake project (e.g. in PyCharm right-click on the CMakeLists.txt file and select Load CMake Project)
2. Create a Release CMake build profile (e.g. in PyCharm click on the CMake icon in the bottom left panel and click on the Gear icon > CMake Settings)
3. Add the following line to your CMake build profile:

   -DCMAKE_TOOLCHAIN_FILE=$CMakeProjectDir$/vendor/vcpkg/scripts/buildsystems/vcpkg.cmake
   

4. Build the project.

License

Copyright (c) Schrodinger, LLC

Published under a BSD-like license, see LICENSE.

Acknowledgements

Schrödinger for being the driving force behind the continued development of PyMOL after Warren's passing, ensuring that the open-source version remained alive and well.

NOTE: the following list has not been updated since Fall 2003.
Since then, the PyMOL effort has grown to such an extent that it is no longer practical to recognize everyone individually. Fortunately, a public record of participation exists and can be appreciated on the internet, and especially via the PyMOL mailing list archives. Suffice it to say that the PyMOL user community now numbers well into the thousands and includes scientists, students, and educators worldwide, spread throughout academia and the biotechnology and pharmaceutical industries. Though DeLano Scientific LLC specifically supports and maintains the PyMOL code base, the project can only continue to succeed through the sponsorship and participation of the broader community.

Founder and Principal Author:

  Warren L. DeLano 

Major Authors (5000+ lines of code):

  Ralf W. Grosse-Kunstleve (SGLite Module)

Minor Authors (500+ lines of code):

  Scott Dixon (Metaphorics CEX support)
  Filipe Maia (Slice Objects)

Other Contributors: These are the people who have gone out of their way to help the project with their ideas, actions, advice, hardware donations, testing, information, sponsorship, peer support, or code snippets.

  Daan van Aalten
  Paul Adams 
  Stephen Adler
  Jun Aishima 
  Dennis Allison
  Ricardo Aparicio
  Daniel Appelman   
  Diosdado "Rey" Banatao
  Michael Banck
  Ulrich Baumann
  Joseph Becker
  Balaji Bhyravbhatla
  Jeff Bizzaro
  Jeff Blaney 
  Juergen Bosch 
  Michael Bower
  Sarina Bromberg
  Axel Brunger
  Robert Campbell
  Bronwyn Carlisle 
  Duilio Cascio
  Julien Chiron 
  Shawn Christensen
  Scott Classen
  David Cooper
  Larry Coopet
  Jacob Corn
  Ben Cornett
  Andrew Dalke 
  Koen van der Drift 
  Harry Dailey
  Byron DeLaBarre
  Bill DeGrado
  Thomas Earnest
  Nathaniel Echols
  John Eksterowicz
  Erik Evensen
  David Fahrney
  Tim Fenn
  Thierry Fischmann
  Michael Ford
  Esben Peter Friis
  Kevin Gardner
  R. Michael Garavito
  John Gerig
  Jonathan Greene
  Michael Goodman  
  Joel Harp
  Reece Hart
  Richard Hart
  Peter Haebel
  Matt Henderson
  Douglas Henry 
  Possu Huang 
  Uwe Hoffmann
  Jenny Hinshaw
  Carly Huitema
  Bjorn Kauppi
  Greg Landrum
  Robert Lawrence Kehrer 
  Tom Lee
  Eugen Leitl
  Ken Lind
  Jules Jacobsen
  Luca Jovine
  Andrey Khavryuchenko
  David Konerding
  Greg Landrum
  Michael Love 
  Tadashi Matsushita
  Genevieve Matthews 
  Gerry McDermott 
  Robert McDowell
  Gustavo Mercier      
  Naveen Michaud-Agrawal
  Aaron Miller
  Holly Miller
  Tim Moore
  Kelley Moremen
  Hideaki Moriyama
  Nigel Moriarty 
  Geoffrey Mueller
  Cameron Mura
  Florian Nachon 
  Hanspeter Niederstrasser 
  Michael Nilges
  Hoa Nguyen
  Shoichiro Ono
  Chris Oubridge
  Andre Padilla
  Jay Pandit
  Ezequiel "Zac" Panepucci
  Robert Phillips
  Hans Purkey
  Rama Ranganathan
  Michael Randal
  Daniel Ricklin 
  Ian Robinson
  Eric Ross
  Kristian Rother
  Marc Saric
  Bill Scott
  Keana Scott
  Denis Shcherbakov
  Goede Schueler
  Paul Sherwood
  Ward Smith
  John Somoza
  David van der Spoel
  Paul Sprengeler
  Matt Stephenson 
  Peter Stogios
  John Stone
  Charlie Strauss
  Michael Summers
  Brian Sutton
  Hanna and Abraham Szoke
  Rod Tweten
  Andras Varadi
  Scott Walsh
  Pat Walters
  Mark White
  Michael Wilson
  Dave Weininger
  Chris Wiesmann
  Charles Wolfus
  Richard Xie

Miscellaneous Code Snippets Lifted From:

  Thomas Malik (fast matrix-multiply code)
  John E. Grayson (Author of "Python and Tkinter")
  Doug Hellmann (Wrote code that JEG later modified.)

Open-Source "Enablers" (essential, but not directly involved):

  Brian Paul (Mesa)
  Mark Kilgard (GLUT)
  Guido van Rossum (Python)
  Linus Torvalds (Linux Kernel)

  Precision Insight (DRI)
  The XFree86 Project (Free Windowing System)
  VA Linux (CVS Hosting)
  Richard Stallman/Free Software Foundation (GNU Suite)
  The unknown authors of EISPACK (Linear Algebra)

Graphics Technology "Enablers" (essential!)

  3dfx (RIP)
  nVidia 
  ATI

Specific Acknowledgments:

• Thanks to Joni W. Lam for making the business work.

• Thanks to John Stone and John Furr for being such excellent colleagues.

• Thanks to Ragu Bharadwaj and Marcin Joachimiak for Java expertise and encouragement.

• Thanks to Apple Computer for continued encouragement, assistance, and HLAs in support of Mac development. Thanks especially to Robert Kehrer for creating so many fun opportunities over the years.

• Thanks to Aaron Miller (GlaxoSmithKline) for a continuous stream of thoughtful opinions and suggestions.

• Thanks to Dave Weininger for suggesting the "roving" feature and for being such an inspirational friend and mentor.

• Thanks to Matt Hahn and Dave Rogers for proving that it can also be done, again.

• Thanks to Mick Savage for providing experienced practical advice on the marketing of scientific software.

• Thanks to Ian Matthew for 3D experience and perspective.

• Thanks for Jeff Blaney for numerous insightful discussions.

• Thanks to Elizabeth Pehrson for making this a team effort.

• Thanks to Erin Bradley for schooling in focus and vision.

• Thanks to Vera Povolona for catalytic clarity and introspection.

• Thanks to Anthony Nichols for proving that it can be done, yet again.

• Thanks to Thompson Doman for timely Open-Source validation.

• Thanks to Manfred Sippl for making it all seem so simple.

• Thanks to Kristian Rother for all his excellent work building on the PyMOL foundation, and in helping others learn to use the software.

• Thanks to Dave Weininger, Scott Dixon, Roger Sayle, Andrew Dalke, Anthony Nichols, Dick Cramer, and David Miller, as well as rest of the Daylight and OpenEye teams for thoughtful discussions on PyMOL and open-source software during my 2002 pilgrimage to Sante Fe, NM.

• Thanks to Ralf Grosse-Kunstleve for his contribution of the "sglite" space group and symmetry handling module.

• Thanks to the scientists and management of Sunesis Pharmaceuticals for supporting PyMOL development since program inception.

• Thanks to the Computational Crystallography Initiative (LBNL) developers for their encouragement, ideas, and support.

• Thanks to Scott Walsh for being the first individual to provide financial support for PyMOL.

• Thanks to the hundreds of individuals, companies, and institutions that have provided financial support for the project.

• Thanks to Brian Paul and the Precision Insight team for development of Mesa/DRI which greatly assisted in the early development of PyMOL.

• Thanks to Michael Love for the first major outside port of PyMOL (to GNU-Darwin/OSX) and for believing in the cause.

• Thanks for Paul Sherwood for making a concerted effort to develop using PyMOL long before the software and vision had matured.

• Thanks to Jay Ponder for thoughtful email discussions on Tinker and the role of open-source scientific software.

• Thanks to hundreds of PyMOL users for the many forms of feedback, bug sightings, and encouragement they've provided.