Data science is a team-driven endeavor. We are pleased to work with industry, academic, and government partners—bringing our multidisciplinary expertise, complex datasets, and HPC resources to bear on technological advancements outside the Lab.
—James Brase, LLNL Deputy Associate Director for Computing Programs
The DSI facilitates vital partnerships between academia, industry, and other U.S. government agencies—thus evolving LLNL’s strategic data science vision and guiding future.
For decades, LLNL has been partnering with other organizations to solve complex problems for our nation, and we are always looking for new opportunities for engagement. In our multidisciplinary applied science laboratory, scientists and engineers with extensive experience developing solutions to a broad set of technical challenges.
With industry collaborators we're developing next-generation compute architectures and algorithms optimized for large-scale machine learning applications. We're expanding our relationships with top-tier academic institutions to broaden awareness of LLNL’s unique research and applications in data science. Through these engagements, we make LLNL datasets more widely available and our high-performance computing systems more widely accessible.
Accelerating Therapeutics for Opportunities in Medicine
The Accelerating Therapeutics for Opportunities in Medicine (ATOM) Consortium explores data-driven pharmaceutical models for safe, effective drug development.
ATOM leverages specialized HPC hardware and software to accelerate molecular analysis and scale predictive machine learning (ML) algorithms. Among other activities, the team is building a library of ML models for automated comparison of hundreds of drug properties known to be important for drug design.
"We’re developing new computational tools to evaluate properties of different molecular combinations, looking for new compounds that show promise before laboratory experimentation needs to take place," says Jonathan Allen, LLNL’s R&D manager for the ATOM partnership.
Opportunities with the University of California System
The University of California (UC) system has 10 campuses across the state and more than a quarter of a million students. The DSI is actively focused on continuing existing partnerships and building close collaborations with many UC campuses with a goal of reaching all of them in the coming years.
For instance, during the annual Data Science Challenge, LLNL mentors guide UC Merced students through solving a unique problem in a scientific discipline. The Challenge was expanded in 2021 to include UC Riverside.
"My hope is that this is a really good experience for them and that they go back and tell their friends what a cool place Livermore is to work," explains computer scientist Brian Gallagher, who served as a mentor for the 2020 Challenge and co-organized the 2021 Challenge.