Collaborative Research: ACED: Accelerating Protein Engineering with Evolution-Guided Generative AI and a Self-Driving Biofoundry

Proteins play a central role in many processes essential to life and have wide-ranging applications in medicine, energy, agriculture, and biotechnology. However, natural proteins are often not ideal for these practical uses. Protein engineering, a field that aims to design proteins with improved or novel functions, has transformed industries by creating tailored proteins. While traditional approaches, such as the Nobel Prize-recognized directed evolution method, have been remarkably successful in numerous protein engineering applications, they are typically slow, costly, and resource-intensive. This project seeks to advance protein engineering by combining cutting-edge artificial intelligence (AI) methods with advanced laboratory automation. By harnessing the power of AI to predict and design protein sequences and integrating it with an automated experimental platform, this research aims to greatly accelerate the discovery of new proteins, offering immense potential across multiple scientific domains with significant commercial and societal impact on medicine, biotechnology, energy, agriculture, chemical manufacturing, consumer products, and more.


This project introduces a novel interdisciplinary approach leveraging recent AI breakthroughs in large language models and generative models, to guide protein function analysis and protein engineering, unlocking an unparalleled efficiency for functional protein discovery. The research focuses on developing new AI techniques tailored to the unique challenges of protein engineering, such as sparse data and the need to balance multiple complex protein properties. By leveraging protein evolution insights and generative modeling, the AI system will guide the design of functional proteins with enhanced properties. An integrated automated Biofoundry will design, create and test AI-designed proteins, validating and then refining the design as needed, enabling a high-throughput, closed-loop discovery process. Beyond advancing the field of protein engineering, the project's algorithmic innovations will contribute to foundational research in AI and computing, with the potential for broad applications in other scientific and technological domains.

This award is co-funded by the Directorate for Computer and Information Science and Engineering and by the Directorate for Biological Sciences.

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.

Award Number: 2435755
Principal Investigator: Huimin Zhao
Funds Obligated: $166,668
State: IL