For a new drug to reach market, it is often an arduous and expensive process. Years of complicated safety and efficacy checks, extensive laboratory testing, clinical trials, and strict federal oversight all ensure that the medicines reaching American consumers are safe to use.
However, by the time all the regulatory red tape has been cut, it can take more than a decade and more than $1 billion to get a single drug to become commercially available. Only a fraction of drug discoveries make it this far. With a shared goal of developing a faster solution, without sacrificing safety, a team of researchers from Frederick National Laboratory (FNL), Lawrence Livermore National Laboratory (LLNL), and the University of California, San Francisco (UCSF) co-founded the Accelerating Therapeutics for Opportunities in Medicine (ATOM) project.
In June 2021, the Department of Energy (DOE) and National Cancer Institute (NCI) signed a five-year Memorandum of Understanding to improve the drug discovery process, shorten the timeline, and ultimately benefit the general population. They took a broad approach to cancer research, working to improve efficiency and effectiveness of predictive oncology using generative AI software, models, and educational resources.
Developing ATOM required specialties: advanced machine learning tools and collaborative cancer research. The NCI, which oversees FNL, is the nation’s primary cancer research organization and the world’s largest funder of cancer research. DOE laboratories like LLNL offer elite computing, modeling, simulation, machine learning, and AI expertise.
Ultimately, six national laboratories from DOE and NCI contributed to the ATOM project. The team partnered with industry partners and two elite research institutions — UCSF and Texas A&M University — to create open-access and FAIR (Findable, Accessible, Interoperable, Reusable) software that safely advances and speeds up the drug discovery process.
All ATOM creations are available at computational.cancer.gov, including the ATOM Modeling PipeLine (AMPL), which is an open-source software for building and sharing models for drug discovery.
Since ATOM’s inception, researchers have widely adopted its tools for biomedical research. For instance, Oak Ridge National Laboratory’s Frontier supercomputer uses ATOM software to train models for international collaborations. At LLNL, AMPL is being used to build prediction models to screen small-molecule compounds for properties suitable for therapeutic use. FNL is using AMPL for its RAS Initiative, which is aimed at combating cancers with certain genetic mutations.
The impacts of ATOM are also being felt across higher education and in international computing centers. UCSF developed a unique student training program designed to equip the next generation of scientists with a multi-disciplinary background in traditional and cutting-edge approaches to drug discovery. ATOM tools have been installed in the United States, the United Kingdom, Germany, and India, including at the Food and Drug Administration, Zuse Institute Berlin HPC Center, and Microsoft Azure.
As of October 2024, ATOM software has an average of 2,000 viewers and 200 clones per month. ATOM has released nine core tutorials to support new users, released a new version of the software, shared real-world use cases, and hosted an ATOM hackathon at the University of Delaware. More than 200 college students have been trained on ATOM technology.
The path to clinical approval for new drugs is still, rightfully, a long and pricey proposition. But thanks to a massive interagency collaboration of national laboratories, we have ATOM — and we have significant progress on hastening drug discovery and development today and into the future.
This technology received the 2025 Interagency Partnership Award. Learn more here and discover more awardees in our Awards Gallery.
Click on any images below to view larger versions and photo captions.