Academic, Industry Leaders Form OpenFold AI Research Consortium to Develop Open Source Software Tools To Understand Biological Systems and Discover New Medicines

Founders of the non-profit consortium also announce release of its first AI model to predict protein structures

A set of leading academic and industry partners are announcing the formation of OpenFold, a non-profit artificial intelligence (AI) research consortium of organizations whose goal is to develop free and open source software tools for biology and drug discovery. OpenFold is a project of the Open Molecular Software Foundation (OMSF), a non-profit organization advancing molecular sciences by building communities for open source research software development.

Latest Aithority Insights: Why Contextual Targeting Deserves Another Look with Artificial Intelligence (AI)

“OpenFold is many things to us, a code, a forum, a set of great minds to discuss our favorite topics! It has been a wonderful experience so far, and we are really excited to build out the next stages of the roadmap!”

OpenFold’s founding members are the Columbia University Laboratory of Mohammed AlQuraishi, Ph.D., Arzeda, Cyrus Biotechnology, Genentech’s Prescient Design, and Outpace Bio. The consortium, whose membership is open to other organizations, is hosted by OMSF and supported by Amazon Web Services (AWS) as part of the AWS Open Data Sponsorship Program. OMSF also hosts OpenFreeEnergy and OpenForceField.

Brian Weitzner, Ph.D. Associate Director of Computational and Structural Biology at Outpace and a co-founder of OpenFold said, “In biology, structure and function are inextricably linked, so a deep understanding of structure is required to elucidate molecular mechanisms and engineer biological systems. We believe that open collaboration and access to powerful AI-powered structural biology tools will transform biotechnology and biosciences by empowering researchers and educators spanning life science companies, tech companies and academia with free access to use and extend these tools to accelerate discovery and develop life-changing technologies.”

The first major research area for the consortium is to create state-of-the-art AI-based protein modeling tools which can predict molecular structures with atomic accuracy. The OpenFold consortium is modeled after pre-competitive technology industry open source consortia such as Linux and OpenAI.

AI and ML News: Why SMBs Shouldn’t Be Afraid of Artificial Intelligence (AI)

First consortium-released AI model to predict protein structure yielding impressive results

The OpenFold founders also officially announced today the full release of its first protein structure prediction AI model developed in Dr. AlQuraishi’s laboratory, first publicly acknowledged on Twitter on June 22, 2022. The model is based on groundbreaking work at Google DeepMind and the University of Washington’s Institute for Protein Design. The software is available under a free and open source license from The Apache Software Foundation at https://github.com/aqlaboratory/openfold. Training data can be found on the Registry of Open Data on AWS. A formal preprint and publication will be forthcoming.

Yih-En Andrew Ban, Ph.D., VP Computing at Arzeda and co-founder of OpenFold said, “This first OpenFold AI model is already producing highly accurate predictions of protein crystal structures as benchmarked on the Continuous Automated Model EvaluatiOn (CAMEO), and has yielded on-average higher accuracy and faster runtimes than DeepMind’s AlphaFold2.” An example output from OpenFold, with comparison to experimental data, is included in the figure.

CAMEO is a project developed by the protein structure prediction scientific community to evaluate the accuracy and reliability of predictions.

Lucas Nivon, Ph.D., CEO at Cyrus and co-founder of OpenFold, said, “The first release of the OpenFold software includes not just inference code and model parameters but full training code, a complete package that has not been released by another entity in the space. It will allow a full set of derivative models to be trained for specialized uses in drug discovery of biologics, small molecules, and other modalities.”

Researchers around the world will be able to use, improve, and contribute to what the consortium founders describe as their “predictive molecular microscope.” Current and future work will extend these derivative models to integrate with other software in the field and to be more useful for protein design and biologics drug discovery specifically.

Richard Bonneau, Ph.D., Executive Director at Genentech’s Prescient Design said, “OpenFold is many things to us, a code, a forum, a set of great minds to discuss our favorite topics! It has been a wonderful experience so far, and we are really excited to build out the next stages of the roadmap!”

Multiple other corporate and non-profit organizations are currently joining the OpenFold consortium as full members, and the founders invite biotech, pharma, technology and other research organizations to join. The consortium is currently evaluating proposals for new AI protein projects from academic groups around the world.

Top Artificial Intelligence Insights: Determining the Potential of Your AI Algorithm Starts with Measurement

[To share your insights with us, please write to sghosh@martechseries.com]