The National Science Foundation (NSF) has awarded Northwestern University $20 million over four years to establish the nation's first publicly accessible, AI-powered cloud laboratory for protein engineering.
Called the AI-Driven, Rapid, Experimental Automation Machine (DREAM) Cloud Lab, this facility aims to become a national resource to dramatically accelerate the development of new medicines, sustainable materials, agricultural technologies and other biotechnology innovations.
The facility is among 20 inaugural NSF-funded Programmable Cloud Laboratory (PCL) test beds, established to create a nationwide network of AI-enabled, remotely accessible laboratories. Rather than requiring researchers to purchase and maintain expensive, sophisticated scientific equipment for their own labs, the network will allow scientists to remotely program experiments, access cutting-edge instruments and accelerate discovery through shared data and AI tools.
Positioned within Northwestern's Center for Synthetic Biology (CSB), DREAM will serve as the protein-engineering hub within the broader PCL network, sharing data, AI models and best practices with other cloud laboratory nodes while enabling collaborations across scientific disciplines.
"Northwestern is honored to be selected by the NSF to host one of the inaugural Programmable Cloud Laboratory test beds in the country," said President Mung Chiang. "DREAM is an important initiative that will create the platform for our Center for Synthetic Biology to collaborate with researchers across the nation, advancing important breakthroughs in protein engineering and broadly in biotechnology. This award is more than financial but also strategic, reflecting again the caliber and ambition of Northwestern research leaders."
"This funding recognizes Northwestern's longstanding leadership at the intersection of engineering, AI and biology," said Chris Schuh, dean of the McCormick School of Engineering. "This investment is not only a milestone for the University but also for the rapidly growing biotechnology ecosystem across Chicago and Illinois."
DREAM will be physically housed within the CSB's Synthetic Biology Foundry, a shared research facility equipped with advanced robotics and protein-engineering tools. The Synthetic Biology Foundry is co-located with the Querrey InQbation Lab, Northwestern's startup incubator hub, allowing research breakthroughs to quickly transition from the lab and into new companies and products.
"DREAM will positively transform our society, spinning out new companies and creating new jobs by accelerating new technologies for manufacturing, critical mineral mining, healthy aging, agriculture and advanced materials," said Northwestern's Julius B. Lucks, the principal investigator (PI) of DREAM. "Building DREAM within the Northwestern Center for Synthetic Biology allows us to integrate our approaches for responsible innovation, boosting our regional biotech ecosystem and serving the entire nation as a hub for transformative science and engineering that benefits everyone."
Lucks is the Margery Claire Carlson Professor of Chemical and Biological Engineering at McCormick and co-director of the Center for Synthetic Biology. DREAM's co-PIs are Danielle Tullman-Ercek, the James N. and Nancy J. Farley Professor in Manufacturing and Entrepreneurship, professor of chemical and biological engineering at McCormick and co-director of the Center for Synthetic Biology; Han Liu, the Orrington Lunt Professor of Computer Science at McCormick and professor of statistics at the Weinberg College of Arts and Sciences; and Michael Jewett, a professor of bioengineering and chemical engineering at Stanford University. Key personnel include Joshua Leonard, professor of chemical and biological engineering at McCormick, and Michelle Hoffmann, executive director of the Chicago Biomedical Consortium.
Power of protein engineering
Proteins perform nearly every essential function of life. Increasingly, scientists seek to engineer proteins for a wide range of applications, including developing new therapeutics, recycling plastics, detecting environmental contaminants and recovering critical minerals. Yet designing proteins that can perform the exact desired function remains slow, expensive and accessible to relatively few laboratories.
Although recent advances in AI have made it possible for scientists to use computation to design promising proteins, they still face a critical bottleneck. Scientists need to build and test proteins to generate the data needed to improve AI models. DREAM is designed to remove that bottleneck.
Through a cloud-based interface, users will specify the desired functions they want a protein to perform. AI models then will propose promising designs and - after being evaluated and cleared through a rigorous biosafety and biosecurity review process - robotic systems will automatically build and test those proteins. The resulting experimental data will continually improve the AI models. The test bed will integrate perspectives from an Ethics and Responsible Innovation Board to ensure that the entire system will create a design-build-test-learn cycle that rapidly refines proteins to meet users' objectives safely and securely.
"Building and testing the proteins are historically tedious, slow and costly steps," Tullman-Ercek said. "To overcome this issue, the DREAM facility leverages cutting-edge developments in the synthesis of proteins using cell-free technologies that were pioneered at Northwestern. It is exciting to see how our past successes pave the way for new developments that will solve important global challenges."
More than 300,000 proteins, 30 million data points
During the four-year award, DREAM's investigators expect to synthesize and characterize more than 300,000 proteins, generating up to 30 million data points - creating one of the largest publicly available datasets for AI-guided protein engineering. The team plans to release the data and AI models through an open-access framework, establishing a national knowledge base that scientists can build upon for years to come.
By providing researchers, startups and industry partners with world-class AI-driven protein-engineering capabilities as well as open-source models, DREAM will become a resource not just for academic investigators, but for startups and industry who are pushing the frontier of AI for biotechnology.
The DREAM node will sit at the heart of the Midwest bioeconomy and, given its national connectedness to the national test bed of other AI "self-driving" labs, will be a foundational resource for the expanding Chicago and Illinois biotechnology sector. Along with Northwestern's Innovation and New Ventures Office, the Chicago Biomedical Consortium will provide expertise for DREAM's translational processes, ensuring that research transfers to applications and products that serve the American public. As a part of this goal, DREAM also will serve as a training ground for students and researchers in AI, laboratory automation and synthetic biology - helping prepare the future workforce for one of the fastest-growing sectors of the U.S. economy.
"The Northwestern DREAM Cloud Lab will be a game changer for the State of Illinois," said Christy George, president and chief executive officer, Illinois Economic Development Corporation. "As the only AI-protein engineering hub in the country, it will add another critical core facility to our strong life sciences ecosystem and offer new opportunities for companies looking to partner and grow in Illinois."
More Northwestern connections
NSF's new PCL Test Bed program is aligned with the U.S. government's Genesis Mission, a national effort that aims to harness AI for scientific discovery. The NSF will invest $380 million across all 20 selected teams.
Northwestern engineers also will contribute to two other PCL test beds. Wei Chen, chair and professor of mechanical engineering, Wilson-Cook Professor in Engineering Design and professor of industrial engineering and management sciences, will serve as co-PI of the Artificial Intelligence for Materials Design Network, led by Johns Hopkins University. Chris Wolverton, the Frank C. Engelhart Professor of Materials Science and Engineering, will serve as co-PI of ATHENA: Advanced Testbed for High-throughput Experimentation in Nano- and Atomic Science, led by the University of Tennessee, Knoxville.
"The NSF Programmable Cloud Laboratories initiative will unlock new discoveries in a wide range of fields and while also advancing AI-enabled technologies that form the backbone of the automated science revolution," said Erwin Gianchandani, NSF assistant director for Technology, Innovation and Partnerships. "These awards represent a significant step toward achieving a virtuous cycle of automated hypothesis generation, autonomous experimentation and generation and interpretation of large volumes of high-quality experimental data to drive the next set of hypotheses, all with researchers and students alongside the automated systems at every step of the way."
