2607522

NSF PCL-TEST BED: AI-DRIVEN, RAPID, EXPERIMENTAL AUTOMATION MACHINE (DREAM) CLOUD LAB FOR PROTEIN ENGINEERING -NSF PCL-TEST BED: AI-DRIVEN, RAPID, EXPERIMENTAL AUTOMATION MACHINE (DREAM) CLOUD LAB FOR PROTEIN ENGINEERING PROTEINS ARE BIOLOGICAL BUILDING BLOCKS THAT UNDERLIE THE NATURAL STRUCTURES AROUND US AND PERFORM THE VAST CHEMISTRY OF LIFE. THEY ARE INVOLVED IN PRODUCING CHEMICALS, FOODS, AND MATERIALS, DETECTING THREATS, AND PROTECTING OUR HEALTH. WHILE NATURE MAKES A WIDE ARRAY OF PROTEINS, DESIGNING NEW ONES IS CURRENTLY A SLOW AND INEFFICIENT PROCESS. THE DREAM CLOUD LAB FOR PROTEIN ENGINEERING WILL REMOVE THIS BOTTLENECK AND ENABLE THE US TO DOMINATE THE ONE-TRILLION-DOLLAR BIOECONOMY. BY COMBINING ARTIFICIAL INTELLIGENCE (AI) WITH AUTOMATED EXPERIMENTS, DREAM WILL ENABLE USERS ACROSS THE U.S. TO INEXPENSIVELY AND RELIABLY ENGINEER CUSTOM PROTEINS FOR APPLICATIONS RANGING FROM MANUFACTURING TO CRITICAL MINERAL MINING, HEALTHY AGING, AGRICULTURE AND ADVANCED MATERIALS. DREAM WILL SIGNIFICANTLY ADVANCE NATIONAL PRIORITIES BY STRENGTHENING THE U.S. BIOECONOMY, ONSHORING PROTEIN DESIGN AND PRODUCTION, AND HELPING CREATE NEW JOBS AND BOOST ECONOMIC GROWTH. IT WILL TRAIN FUTURE LEADERS AND A SKILLED WORKFORCE FLUENT IN BOTH AI AND BIOLOGY. DREAM WILL DRIVE DOMESTIC MANUFACTURING AND SERVE AS A KEY PILLAR OF NATIONAL SECURITY AND COMPETITIVENESS BY GIVING THE U.S. A NEW COMPETITIVE EDGE IN AI APPROACHES TO ENGINEER BIOTECHNOLOGIES.? DREAM WILL USHER IN A NEW PARADIGM FOR PROTEIN ENGINEERING BY FACILITATING A COMMON AI-DRIVEN DESIGN-BUILD-TEST-LEARN (DBTL) CYCLE THAT CAN BE APPLIED TO ANY PROTEIN ENGINEERING TASK CONSISTING OF: (1) AI-DRIVEN DESIGN, (2) SYNTHESIS (BUILD), (3) HIGH-THROUGHPUT CHARACTERIZATION (TEST) OF PROTEIN LIBRARIES WITH SPECIFIC FUNCTION TARGETS, AND (4) FURTHER TRAINING OF AI MODELS (LEARN) TO ALLOW CONTINUAL IMPROVEMENTS AND GENERALIZATION TO BROAD PROTEIN ENGINEERING TASKS. THIS PROCESS WILL BE ENABLED BY OUR USE OF ?CELL-FREE? PROTEIN SYNTHESIS SYSTEMS WHICH CAN RAPIDLY PRODUCE COMPLEX PROTEINS WITH A SUITE OF AUTOMATED EQUIPMENT.? THE FACILITY WILL ALSO VALIDATE FUNCTION BY CHARACTERIZING THE ENZYME KINETICS OF PROTEINS, PROTEIN AND SMALL MOLECULE INTERACTIONS, ASSEMBLY FORMATION, MECHANICAL PROPERTIES, AND FOLDING AND STABILITY. A BROAD BASE OF USERS WILL BE ABLE TO ACCESS DREAM THROUGH AN AI-ENABLED CLOUD LAB INFRASTRUCTURE THAT ALLOWS USERS TO REQUEST PROTEIN DESIGNS, CHECK PROCESS STATUS, AND RECEIVE GENERATED DATA. DREAM WILL ADAPT TO EVOLVING USER NEEDS THROUGH TARGETED OUTREACH, SYSTEMATIC EVALUATION, AND AN AGILE MANAGEMENT PLAN WHILE INTEGRATING ETHICS AND A MULTI-STAGE BIOSECURITY FRAMEWORK TO ENABLE RESPONSIBLE OPERATION. THE OPEN-ACCESS RELEASE OF DREAM DATA AND MODELS WILL CREATE A NATIONAL PROTEIN FUNCTION KNOWLEDGEBASE ENABLING BREAKTHROUGHS IN BIOLOGICAL DESIGN. DREAM WILL ALSO IMPLEMENT A TRAINING AND EDUCATION PIPELINE WITH STAGES TAILORED TO DIFFERENT LEVELS OF EXPERTISE, FROM COMMUNITY COLLEGE USERS TO PROTEIN ENGINEERS WITH DECADES OF EXPERIENCE. DREAM WILL CONTRIBUTE TO THE WIDER NSF PROGRAMMABLE CLOUD LAB NETWORK BY JOINING OTHER NODES TO SOLVE GRAND CROSS-SECTOR TECHNOLOGICAL CHALLENGES AND SHARE KNOWLEDGE TO STRENGTHEN THE OVERALL NATIONAL NETWORK. DREAM WILL BE A NATIONAL RESOURCE, WORKING ACROSS ACADEMIA, INDUSTRY, AND OTHER CLOUD LAB NODES TO DRIVE THE ECONOMY AND PROVIDE A STRATEGIC U.S. ADVANTAGE IN AI AND BIOTECHNOLOGY. THIS AWARD REFLECTS NSF'S STATUTORY MISSION AND HAS BEEN DEEMED WORTHY OF SUPPORT THROUGH EVALUATION USING THE FOUNDATION'S INTELLECTUAL MERIT AND BROADER IMPACTS REVIEW CRITERIA.- SUBAWARDS ARE PLANNED FOR THIS AWARD.

Northwestern University

THE GOAL OF THIS FUNDING OPPORTUNITY, "TEST BED: TOWARD A NETWORK OF PROGRAMMABLE CLOUD LABORATORIES", IS IDENTIFIED IN THE LINK: HTTPS://WWW.NSF.GOV/PUBLICATIONS/PUB_SUMM.JSP?ODS_KEY=NSF25541

47.084 - NSF Technology, Innovation, and Partnerships

National Science Foundation (NSF)

Evanston, Illinois 60208-0001 United States

Single Zip Code

Test Bed: Toward a Network of Programmable Cloud Laboratories (25-541)