External References

Discovery of Novel Gain-of-Function Mutations Guided by Structure-Based Deep

• TEM-1 β-lactamase
• Blue Fluorescent Protein
• Phosphomannose isomerase

• 7/10 increased antibiotic degradation (β- lactamase)
• 4/10 improved fluorescence (BFP)
• 6/10 improved solubility (PMI)

• 5 fold improvement in soluble expression (β- lactamase)
• 5 fold greater flurosecence (BFP)
• 5 fold greater solubility (PMI)

Machine learning-aided engineering of hydrolases for PET depolymerization

• Up to 3.8 fold improvement in the protein yield of 27 different variant suggested by the algorithm compared to wild type
• Improvement of 10°C in melting temperature

Biosensor and machine learning-aided engineering of an amaryllidaceae enzyme

• 60% improvement in product titer
• 2 fold higher catalytic activity

Improved Bst DNA Polymerase Variants Derived via a Machine Learning Approach

• 3 fold improved yield, robust to purification
• Greater thermal tolerance and thermostability (73°C)

Enhancing PET Degrading Enzymes: A Combinatory Approach

• Used AI with rational design and ancestral sequence design for a 58 fold improvement in the yield

Deep learning guided enzyme engineering of Thermobifida fusca cutinase for increased PET depolymerization

• 5.3 fold improved hydrolysis compared to wild type
• Up to 5.7°C higher half inactivation temperature