Katalyst: Knowledge-Guided Semantic Alignment via Contrastive Learning for Enzyme Turnover Prediction.

Katalyst achieves the most accurate enzyme turnover prediction with a 3.01% reduction in RMSE and significant improvements in R2 and SRCC over state-of-the-art methods.

• Why it matters: Accurate enzyme turnover numbers are vital for understanding enzyme kinetics, but current experimental and computational methods are limited by high costs and low throughput, hindering progress in enzyme research.
• What they did: The approach integrates enzyme sequences, textual descriptions, substrates, and products into a shared latent space using multi-channel encoders and contrastive learning, along with an interaction-aware fusion module, evaluated across multiple datasets.
• The result: Katalyst demonstrates superior performance with an RMSE of 0.9880, R2 of 0.5829, and SRCC of 0.7603, enabling more reliable enzyme kinetics predictions and advancing computational enzyme analysis.