Active Grants on OMSF

Alchemistry

Mon, 01 Jan 0001 00:00:00 +0000

Alchemistry

Alchemistry is a collaborative, ever-changing community dedicated to advancing the science and practice of molecular free energy calculations. Its members represent a diverse cross-section of academia and industry, united by a shared goal: to introduce the latest and greatest in computational free energy methods.

Each year, Alchemistry organizes the Workshop on Free Energy Methods in Drug Design, a meeting that convenes experts from across academia and industry to discuss advances in theory, computation, and application. The workshop provides a forum for exploring algorithmic innovations, benchmarking efforts, and real-world case studies that demonstrate how free energy methods can accelerate pharmaceutical research. Through talks, tutorials, and community discussions, participants help shape the next generation of molecular simulation tools and standards.

Materials from past workshops are freely available through the Alchemistry website, offering a lasting educational resource for the global molecular simulation community. This open archive extends the reach of the event, ensuring that its insights continue to inform and inspire researchers long after each meeting concludes.

Registration for the Free Energy Workshop can also be found at the website! Reach out to alchemistry@omsf.io for any inquiries.

Website: alchemistry.org
Event Information: 2026 Free Energy Workshop Schedule

BPDMC

Mon, 01 Jan 0001 00:00:00 +0000

Boston Protein Design and Modeling Club

The Boston Protein Design and Modeling Club is a collaborative community of researchers, students, and professionals discussing advances at the frontier of computational protein science. Based in one of the world’s leading hubs for biotechnology and academia, the club provides a forum for sharing ideas, presenting research, and fostering cross-institutional collaboration across the broad landscape of protein design, modeling, and simulation.

At its core, the BPDMC is dedicated to the principle that open exchange accelerates discovery. By bringing together experts from academia and industry, the club creates a space for constructive dialogue on the latest innovations in computational biology - from algorithmic breakthroughs and biophysical modeling to machine learning–driven design and experimental validation, all against the backdrop of free pizza.

And if you can’t make the meetings, BPDMC maintains an active YouTube channel, featuring recordings of all past presentations. Each talk is freely available to the public, providing a comprehensive archive of research and discussion across computational and experimental protein science. The channel serves as a resource for students, researchers, and practitioners interested in the latest advances in protein design and modeling.

Website: https://www.bpdmc.org/
YouTube: https://www.youtube.com/c/bostonproteindesignandmodelingclub

OMSF POSE

Mon, 01 Jan 0001 00:00:00 +0000

Building open source ecosystems in molecular sciences through collaboration and technology

The NSF POSE Phase II project, awarded to OMSF in September 2023, focuses on enhancing the open-source ecosystem in molecular sciences through strategic collaboration and technological innovation. It aims to leverage the expertise of OMSF’s in-house software engineers to build a shared infrastructure across OMSF’s hosted projects. These contributions are expected to advance the development of programs and foster a collaborative environment for scientific software development, aligning with the project’s goals to bolster molecular modeling and open science practices.

For the entire proposal, visit this Zenodo page. For the latest updates regarding this initiative, visit omsf.substack.com.

Program Outputs:

Contact: Karmen Condic-Jurkic

Open Force Field

Mon, 01 Jan 0001 00:00:00 +0000

The Open Force Field Initiative represents a collaborative effort between academic and industry researchers, united in their goal to revolutionize the science and infrastructure necessary for developing advanced small molecule and biomolecular force fields. This initiative is driven by the need for more accurate molecular mechanics force fields, which are essential for enhancing the predictive power and utility of molecular modeling in pharmaceutical discovery. The current force fields, largely based on modeling work from the 1980s and 1990s, have limitations in accuracy that constrain their effectiveness in guiding pharmaceutical discovery and design.

To address this, the Initiative is focused on building iteratively more accurate force fields, leveraging improved theoretical methods, modern software infrastructure, and extensive machine-readable data. The aim is to produce new, more accurate, and extensible force fields that can be steadily improved to meet the demands of modern pharmaceutical R&D. These force fields are designed to improve predictions in a variety of applications, including binding affinity, selectivity, drug resistance, partitioning, solubility, kinetics, and other properties.

Central to this effort is the development of a modern, open, sustainable, and extensible framework for automated force field improvement and application. The Initiative has released rapid iteratively improved versions of AMBER-compatible small molecule force fields, and plans to develop comprehensive force fields that break free from legacy accuracy limitations while maintaining compatibility with existing simulation software. This approach aligns closely with industry needs, ensuring relevance and applicability in R&D contexts.

An essential component of the Initiative is its commitment to open source, open data, and open science. All software, code, data, and force fields are made freely available under permissive Open Source Initiative and Creative Commons approved licenses. This openness not only provides a foundation for further scientific exploration but also enables industry partners to extend the force fields with proprietary data or develop their own workflows using the open tools from this initiative. This approach not only fosters innovation but also contributes to the sustainability of the project.

To guide its strategic direction, the Initiative has established a scientific advisory board comprising academic experts in force field development and molecular modeling. This board, along with regular virtual and in-person meetings, provides valuable insights, ensuring that the Initiative stays aligned with the most relevant engineering and scientific efforts in the field of biomolecular simulation and force field development. In addition, this board welcomes industry partners from organizations that fund this effort.

Website: https://openforcefield.org/
Github: https://github.com/openforcefield
Contact: info@openforcefield.org

Open Free Energy

Mon, 01 Jan 0001 00:00:00 +0000

Open Free Energy is dedicated to developing software that is not only stable and reliable for industrial needs but also fosters the development of new methods destined to become industry standards. The project aims to build bridges between academic and industrial users, allowing them to benefit from the latest advancements in free energy methods. Part of their work includes enhancing the interoperability of existing open source software and assuming maintenance responsibilities for valuable existing packages that lack active maintainers.

The Consortium’s primary goals are threefold: first, to accelerate the development of robust, scalable, and flexible open-source GPU-accelerated alchemical free energy packages; second, to create infrastructure for interoperable workflow components; and third, to establish an open component registry. These aims are geared towards enabling portable, interoperable, and robust free energy calculation workflows, addressing the challenges of the diverse and rapidly evolving free energy methods and the significant investment required by individual pharmaceutical companies to adapt these methods in-house.

The need for such an initiative became evident following the 2018 Alchemical Free Energy Methods in Drug Discovery meeting in Göttingen. Representatives from numerous pharmaceutical companies recognized the need for alchemical free energy methods in drug discovery in biomolecular modeling. However, the ecosystem of alchemical free energy calculation packages lacked interoperability, robustness, and rapid development, hindering its ability to meet the burgeoning needs of the pharmaceutical industry and the increasing availability of inexpensive GPU computing resources.

Check the links below to find out more about Open Free Energy!

Website: https://openfree.energy/
Github: https://github.com/openfreeenergy
Contact: OpenFreeEnergy@omsf.io

Open Rosetta by Rosetta Commons

Mon, 01 Jan 0001 00:00:00 +0000

Open Rosetta is a project aimed at transitioning the Rosetta Commons, a global community of over 100 labs, to an open source model. Rosetta Commons has been at the forefront of protein and RNA modeling and design for over two decades - The Rosetta software suite, renowned for its versatility in modeling a wide range of biomolecules and molecular interactions, is currently available under a custom license for non-commercial use through the University of Washington.

The overarching goal of Rosetta Commons is to continue advancing state-of-the-art tools in macromolecular modeling and design, particularly by incorporating deep learning technology. The move to an open source model aligns with the community’s values and is seen as the most efficient way to harness the power of the Rosetta software suite. This collaborative approach involves academia, government labs, research centers, and corporate partners, aiming to tackle the newest challenges in biomolecular modeling.

This change is not without challenges. It involves a thorough reexamination and adaptation of the organizational structure of Rosetta Commons. This initiative aims to lay the groundwork for a new funding model, establish a technical roadmap, and create a structure that invites contributions from scientists at all career stages across various employment sectors. This transformation is designed to honor the community’s technical achievements, commitment to education and diversity, and its culture of collaborative innovation, embodying Rosetta’s principle of working “better together.”

Website https://www.rosettacommons.org/

OpenADMET

Mon, 01 Jan 0001 00:00:00 +0000

OpenADMET

OpenADMET is an open effort to build predictive models of ADMET properties and understand the mechanisms by which they arise. This means systematically characterizing the proteins and mechanisms that give rise to these properties (protein structures and scaled functional and genetic assays), understanding how small molecules interact with these mechanisms (through high-throughput nanoscale chemistry to explore/exploit chemical space), and integrative computational models (AI/ML ligand/structural, mechanistic, and physiological modeling).

We see our role as a guide to the community by developing open datasets and computational models. One way to ignite innovation in ADMET modeling is through community blind challenges. Blind challenges can provide accurate benchmarks of current performance and help us understand how much we have left to achieve. The paragon example is the CASP challenge that set up conditions for the “AlphaFold” breakthrough in protein structure prediction. For ADMET challenges, we plan on using both our generated data on anti-targets of broad interest and ADMET data donated from the community, as we did in our first challenge with the ASAP AViDD center.

OpenADMET is a nascent coalition of aligned efforts funded by different organizations. Our work currently involves personnel at OMSF, UCSF, Octant, and MSKCC. Our initial funding is through an ARPA-H grant: “AVOID-OME”. Since then, we’ve additionally been funded by the Gates Foundation & Schrodinger, to expand into toxicity and fundamental molecular properties, and by the Astera Institute, to expand our metabolism dataset coverage. We are hard at work on these efforts, and please stay tuned as we begin to launch some new competitions and ways to get involved over the next year.

Website: openadmet.org
Github: github.com/OpenADMET

OpenFold

Mon, 01 Jan 0001 00:00:00 +0000

OpenFold

OpenFold is a non-profit AI research and development consortium dedicated to democratizing access to the most advanced AI systems, capable of engineering life’s molecules, for a diverse array of users including academics, biotech and pharmaceutical companies, and students. These tools are designed to accelerate fundamental biological research and facilitate the development of new cures, previously unimaginable without the aid of AI.

Central to OpenFold’s vision is the recognition of the vital link between structure and function in biology. Understanding the intricacies of biological systems, their engineering, and how to influence them is deeply intertwined with knowledge of their structure. To this end, OpenFold’s AI-based protein modeling tools are built to predict molecular structures with atomic accuracy, a feat that is being made accessible in open source for the first time. This advancement is envisaged as a “predictive molecular microscope,” which researchers worldwide can use, enhance, and contribute to, thereby propelling further discoveries and innovations.

Additionally, OpenFold is focused on optimizing the performance of their model for use on state-of-the-art, widely available GPUs. This hopes to not only foster an environment of collaboration and open-source development, but also ensure that the benefits of these advanced AI tools remain accessible to a broad range of users, catalyzing advancements in biology and drug discovery.

Website: https://openfold.io/
Github: https://github.com/aqlaboratory/openfold

Rosetta POSE

Mon, 01 Jan 0001 00:00:00 +0000

Transitioning Rosetta Commons to a self-sustainable Open Source Ecosystem

The NSF POSE Phase II grant supports the transition of Rosetta Commons toward a fully open-source ecosystem, modernizing governance, licensing, and operational structures to enable broader collaboration across the Rosetta ecosystem. Building on earlier strategic work, the initiative focuses on developing shared standards, improving interoperability with emerging tools, and establishing sustainable infrastructure for community-driven software development.

The project aims to strengthen long-term sustainability of the Rosetta software suite, expand contributor participation in the developement of Rosetta software, and advance open scientific infrastructure standards.

Contact: Julia Koehler-Leman

VWSCC

Mon, 01 Jan 0001 00:00:00 +0000

Virtual Winter School on Computational Chemistry

Not everyone can travel to scientific conferences due to financial constraints, health concerns, family responsibilities, or teaching commitments. These barriers often disproportionately impact women with young children, but many others also benefit from a virtual format that removes these obstacles.

The Virtual Winter School on Computational Chemistry are committed to offering an alternative to these classical models. Offering a week long course in the winter, VWSCC gives its participants a comprehensive walkthrough through the basics and best practices in computational chemistry.

Virtual lectures offer a sustainable alternative, allowing participants to engage from home without the need for travel. Recorded sessions also provide flexibility, enabling attendees to watch at their convenience. To ensure accessibility, the Virtual Winter School offers free registration, making scientific content available to researchers worldwide, regardless of financial, health, or family circumstances.

Additionally, the online format enhances global collaboration, fostering connections among peers from different regions. Our goal is to create a dynamic platform that supports international dialogue in computational chemistry and interdisciplinary exchange. To make the content accessible to a broad audience, we encourage speakers to include introductory material in their presentations.

Website: <winterschool.cc>
Current Chair: Cate Anstöter

WESTPA

Mon, 01 Jan 0001 00:00:00 +0000

WESTPA, short for the Weighted Ensemble Simulation Toolkit with Parallelization and Analysis, is an advanced Python-based framework designed for executing the weighted ensemble (WE) path sampling method. The WESTPA software encompasses a range of features, including binned and binless WE strategies, options for reweighting to steady states for rate and state population estimations from shorter trajectories, and the capability for advanced customization through plugins and extensions.

Notable attributes of WESTPA include:

Scalability, with the ability to operate efficiently on thousands of CPU cores or GPUs.
Interoperability, offering seamless integration with various stochastic dynamics engines, such as molecular dynamics or Monte Carlo simulations.
Extensibility, due to a modular design that simplifies the development of new plugins.
Portability, ensuring compatibility with various Unix operating systems, including Linux and OS X, on standard clusters and supercomputers.
Open-source availability, with all source code accessible under the MIT license.

Website https://westpa.github.io/westpa/
Github https://github.com/westpa/westpa/wiki