OPAL: Autonomous BioDesign for Faster Product Development
Four U.S. national laboratories and industry partners are working together to speed up how biology moves from discovery to shipping products. The Orchestrated Platform for Autonomous Laboratories to Accelerate AI-Driven BioDesign (OPAL) uses robotics, AI agents and models, and standardized data to shorten the path from gene to pilot scale for fuels, chemicals, consumer goods, agriculture, and critical mineral recovery.
For product teams, this is about compressing discovery, build, test, and learn into tight loops. Automated labs run more experiments with higher consistency. AI foundation models guide what to try next. Data flows cleanly across tools and sites so you can make decisions earlier with less guesswork.
What's new
- Biology foundation models: OPAL is developing general-purpose models that can be adapted to specific applications and eventually direct AI agents to run investigations autonomously.
- High-quality data at scale: The team will generate the largest, most precise biological datasets to date using DOE user facilities and supercomputers, plus decades of national lab and industry results. As Paramvir Dehal noted, biology has fewer, more variable datasets than text-OPAL addresses that gap.
- Closed-loop experimentation: Models will integrate with robotic platforms to plan, execute, and learn from experiments, shrinking cycles that would take weeks or months down to days.
"AI models are already transforming many fields, but building them for biological research has been slower because there are fewer datasets on genomes, proteins, and metabolic functions of organisms to train them on. And the datasets we do have are highly variable and often organized very differently - unlike the text-based datasets that large language models train on," said Paramvir Dehal, OPAL cross-cut task lead and computational staff scientist in the Lab's Biosciences Area.
"These models advance biological understanding by enabling prediction, model-informed control, and design, with applications from environmental productivity and resilience to biomanufacturing," Dehal added.
Why it matters for product development
- Shorter R&D lead times: Standardized data and AI-guided experiments reduce iteration cycles and cut down on dead-ends.
- Model-informed design: Predict performance upfront, prioritize high-yield variants, and reduce wet-lab burden.
- Smoother scale-up: Automated, consistent data generation improves transfer from bench to pilot and de-risks tech readiness milestones.
- Better team throughput: Robotic execution and AI triage free scientists to focus on higher-leverage decisions.
"OPAL is at the forefront of changing how we do biological research, using advanced AI methods to dramatically improve our understanding of biological systems, but to realize that potential we need to collect significantly more data, and AI can help us do that smartly," said Paul Adams, Associate Laboratory Director for Biosciences and OPAL lead point of contact. "The foundational genomic models from our seed project will have broad application, from critical minerals to high-performance jet fuel precursors, helping to transform biotechnology and industrial processes."
Where OPAL fits in the national AI effort
OPAL is part of the Department of Energy's Genesis Mission, a new initiative to advance AI and speed discovery for science, energy, and national security. A cornerstone of this initiative is the Transformational AI Models Consortium (ModCon), which develops self-improving AI models by leveraging DOE data, facilities, and expertise. OPAL is one of three ModCon projects with Berkeley Lab leadership or major roles.
The project team includes scientists and engineers from Lawrence Berkeley, Oak Ridge, Argonne, and Pacific Northwest national laboratories, with support from DOE's Office of Biological and Environmental Research (BER) and Advanced Scientific Computing Research (ASCR) programs. Learn more about the DOE Office of Science at energy.gov/science and explore DOE user facilities at science.osti.gov/User-Facilities.
Near-term actions for product leaders
- Define high-value use cases: e.g., higher titer/flux for a target molecule, strain robustness, or feedstock flexibility.
- Instrument your data: Adopt consistent schemas, ontologies, and metadata; ensure your LIMS/ELN exposes APIs so models can read/write outcomes.
- Plan for closed-loop work: Stand up small robotic workflows (even partial automation) that can iterate daily with model guidance.
- Make compute accessible: Budget for model training/inference and set up secure paths to share non-sensitive data with partners.
- Build the team: Pair domain scientists with ML engineers; train PMs to scope AI-enabled experiments and interpret model confidence.
- Upskill fast: If your roadmap leans on AI-assisted R&D, equip your team with focused training. See curated programs by role at Complete AI Training.
Where you'll see impact first
- Fuels and chemicals: Faster strain design for higher yields, improved tolerance, and cleaner downstream processing.
- Consumer goods and materials: Precision pathways for specialty ingredients, fragrances, and sustainable polymers.
- Agriculture: Microbial solutions for crop resilience and nutrient efficiency validated with automated experimentation.
- Critical minerals: Biological routes for selective recovery and processing with model-guided optimization.
About Berkeley Lab and the DOE Office of Science
Lawrence Berkeley National Laboratory conducts discovery science and develops solutions for reliable, abundant energy. The lab's strengths span materials, chemistry, physics, biology, earth and environmental science, mathematics, and computing, and its user facilities serve researchers worldwide. Founded in 1931, Berkeley Lab scientists have received 17 Nobel Prizes. DOE's Office of Science is the largest U.S. funder of basic research in the physical sciences. Learn more at energy.gov/science.
Your membership also unlocks:
