WESTPA (The Weighted Ensemble Simulation Toolkit with Parallelization and Analysis) is a high-performance Python implementation of the weighted ensemble (WE) path sampling strategy. WE enables simulations of rare events that occur on timescales orders of magnitude longer than conventional simulations while maintaining rigorous kinetics.

Key features of WESTPA:

• Highly scalable. Scales out to thousands of CPU cores or GPUs.
  
• Interoperable. Interfaces with any stochastic dynamics engine (e.g., MD, Monte Carlo).
  
• Extensible. Our modular design makes it straightforward to develop plug-ins.
  
• Portable. The software can be used with any Unix operating system.
  
• Free and open source. All source code is available under the MIT license.
  

WESTPA Consortium and Industry Support

The WESTPA Consortium provides provides an opportunity for industry to support the continued development and sustainability of open-source scientific software while engaging with the WESTPA project at a level appropriate to its needs. Contributions are tax-deductible donations made through WESTPA's fiscal sponsor, the Open Molecular Software Foundation (OMSF)[pdf]. There are three levels of Consortium membership:

• Supporting Members ($20,000/year) — Participate in Consortium governance through elected representation on the Governing Board.
• General Members ($50,000/year) — Receive technical support and scientific consultation from the WESTPA team, and participate in Consortium governance through elected representation on the Governing Board.
• Premier Members ($100,000/year) — Receive priority technical support and scientific consultation, including assistance with workflow integration and custom solutions. Premier Members may also appoint a representative to the Governing Board.

If you are interested in Consortium membership or would like to learn more about supporting WESTPA, please contact Lillian Chong using this form.

Please cite us when using WESTPA:

MC Zwier, JL Adelman, JW Kaus, AJ Pratt, KF Wong, NB Rego, E Suárez, S Lettieri, DW Wang, M Grabe, DM Zuckerman, and LT Chong. “WESTPA: An interoperable, highly scalable software package for weighted ensemble simulation and analysis”. J. Chem. Theory Comput., 11: 800-809 (2015).

JD Russo*, S Zhang*, JMG Leung*, AT Bogetti*, JP Thompson, AJ DeGrave, PA Torrillo, AJ Pratt, KF Wong, J Xia, J Copperman, JL Adelman, MC Zwier, DN LeBard, DM Zuckerman, and LT Chong. "WESTPA 2.0: High-Performance Upgrades for Weighted Ensemble Simulations and Analysis of Longer-Timescale Applications". J. Chem. Theory Comput., 18: 638–649 (2022). *equal authorship

To make a feature request, please fill out this form. 

Obtaining and Installing WESTPA

Follow our Getting Started guide or Installing WESTPA page.

Documentation and Tutorials

To leverage the full power of the WE strategy, it is essential to invest the time to learn the theory and best practices for running the simulations. For documentation, please see the WESTPA Wiki (general usage) and Sphinx documentation (code). Our wiki page with WESTPA tutorials and best practices can be found here.

WESTPA User Support

We suggest subscribing to the WESTPA user mailing list. Before posting an issue to the list, please use the most recent version of WESTPA and search the archives.

If you have completed the LiveCoMS WESTPA tutorials and are ready for production simulations of your system, we invite you to join us via zoom for a WE data club where we provide "no strings attached" expert advice on getting your simulations off the ground.

For this data club, please be ready to present preliminary analysis of your WESTPA simulation(s) with w_pdist and plothist to provide plots of the probability distribution as a function of your chosen progress coordinate, time-evolution of this probability distribution, and the average probability distribution as a function of multiple dimensions. More detailed instructions can be found in the LiveCoMS WESTPA tutorials.

Contribute to WESTPA Code

We welcome contributions to WESTPA code and/or documentation! Please see our Governance Document and the WESTPA AI Policy.

Examples of WESTPA plugins/extensions:

• A WE-based string method. [src] [pdf]
• Markovian WE Milestoning (M-WEM). [src] [pdf]
• A hybrid Gaussian-accelerated MD and WE method. [src] [pdf]
• DeepWEST: Deep-learned kinetic modeling for generating initial states. [src] [pdf]
• WE with SPIB-learned progress coordinates. [src] [pdf]
• WE with the CHARMM Drude polarizable force field. [src] [pdf]
• WE rule-based modeling (WEBNG) for systems biology. [src] [pdf]

Funding