The effective design of combinatorial libraries to balance fitness and diversity facilitates the engineering of useful enzyme functions, particularly those that are poorly characterized or unknown in biology. We introduce MODIFY, a machine learning (ML) algorithm that learns from natural protein sequences to infer evolutionarily plausible mutations and predict enzyme fitness. MODIFY co-optimizes predicted fitness and sequence diversity of starting libraries, prioritizing high-fitness variants while ensuring broad sequence coverage. In silico evaluation shows that MODIFY outperforms state-of-the-art unsupervised methods in zero-shot fitness prediction and enables ML-guided directed evolution with enhanced efficiency. Using MODIFY, we engineer generalist biocatalysts derived from a thermostable cytochromecto achieve enantioselective C-B and C-Si bond formation via a new-to-nature carbene transfer mechanism, leading to biocatalysts six mutations away from previously developed enzymes while exhibiting superior or comparable activities. These results demonstrate MODIFY’s potential in solving challenging enzyme engineering problems beyond the reach of classic directed evolution.
SUMMARY: The specificity factor( SC/O ) of ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) is a key kinetic parameter quantifying the ratio of carboxylation to oxygenation reactions. Current methods to measure SC/O are expensive, labor-intensive, require specialized expertise, and/or rely on radiolabeled compounds, hindering Rubisco research. We present RubiSpec-MS (Rubisco Specificity by LC-MS/MS), a method to determine the SC/O of purified Rubiscos using liquid chromatography-coupled mass spectrometry (LC-MS/MS) for separation and sensitive detection of non-radioactive reaction products. We benchmark RubiSpec-MS using multiple model Rubisco enzymes. We find that the assay accurately reproduces corresponding literature values for each enzyme and performs similarly in a side-by-side comparison using an established radiometric method. RubiSpec-MS provides a faster, more accessible, and highly parallelizable method to measure Rubisco SC/O using equipment common in academic core facilities and industrial laboratories.