EnzyCLIP: A Cross-Attention Dual Encoder Framework with Contrastive Learning for Predicting Enzyme Kinetic Constants

TL;DR

EnzyCLIP is a novel dual-encoder framework that uses contrastive learning and cross-attention to accurately predict enzyme kinetic constants from protein and substrate data, advancing computational enzyme analysis.

Contribution

This work introduces EnzyCLIP, combining multimodal embeddings with a CLIP-inspired architecture for joint enzyme kinetic parameter prediction, a novel approach in the field.

Findings

Achieved R2 scores of 0.593 for Kcat and 0.607 for Km prediction.

XGBoost further improved Km prediction to R2 = 0.61.

Demonstrated effective modeling of enzyme-substrate interactions using multimodal contrastive learning.

Abstract

Accurate prediction of enzyme kinetic parameters is crucial for drug discovery, metabolic engineering, and synthetic biology applications. Current computational approaches face limitations in capturing complex enzyme-substrate interactions and often focus on single parameters while neglecting the joint prediction of catalytic turnover numbers (Kcat) and Michaelis-Menten constants (Km). We present EnzyCLIP, a novel dual-encoder framework that leverages contrastive learning and cross-attention mechanisms to predict enzyme kinetic parameters from protein sequences and substrate molecular structures. Our approach integrates ESM-2 protein language model embeddings with ChemBERTa chemical representations through a CLIP-inspired architecture enhanced with bidirectional cross-attention for dynamic enzyme-substrate interaction modeling. EnzyCLIP combines InfoNCE contrastive loss with Huber regression loss to learn aligned multimodal representations while predicting log10-transformed kinetic parameters. The model is trained on the CatPred-DB database containing 23,151 Kcat and 41,174 Km experimentally validated measurements, and achieved competitive performance with R2 scores of 0.593 for Kcat and 0.607 for Km prediction. XGBoost ensemble methods applied to the learned embeddings further improved Km prediction (R2 = 0.61) while maintaining robust Kcat performance.