Artificial Intelligence (AI), Synthetic Biology, and Robotics Combine to Improve Enzymes

Researchers at the Center for Advanced Bioenergy and Bioproducts Innovation (CABBI) have developed a fast, automated platform that integrates artificial intelligence, synthetic biology, and robotics to optimize industrial enzymes. By utilizing a biofoundry to automate the design-build-test-learn cycle, the team successfully addressed common bottlenecks in protein engineering such as low efficiency and high costs. This advancement marks a significant milestone for the AI sector by demonstrating how machine learning can drive a fully self-driving laboratory to accelerate the global bioeconomy.

The research team at the Carl R. Woese Institute for Genomic Biology (IGB) utilized the iBioFAB biofoundry to create a user-friendly process for improving industrial enzymes. Traditionally, engineering proteins to enhance efficiency or specific functions is a labor-intensive and expensive endeavor requiring specialized expertise. By combining AI with robotics and informatics, the CABBI team minimized the need for human intervention, allowing for a more versatile approach that can be applied across various industrial sectors to support agriculture and domestic energy production.

The workflow functions as a closed-loop system where an AI tool first analyzes datasets of known enzyme structures to suggest specific sequence changes. These designs are then physically produced by automated protein-building machines within the biofoundry and rapidly tested to characterize their performance. The resulting data is fed back into a secondary AI model, which refines the next iteration of protein designs, moving the facility closer to becoming a completely self-driving laboratory.

In a practical application of this technology, the researchers targeted two industrially relevant enzymes, achieving performance increases of 16 and 26 times their original activity levels. This case study demonstrated that the AI-powered platform could achieve these results with greater speed and less manual labor than traditional methods. Supported by the U.S. Department of Energy, this research highlights the transformative potential of AI in automating complex biological engineering tasks and transitioning toward a bio-based economy.