The catalytic activity of enzymes is intricately determined by their amino acid sequences and assay conditions, particularly temperature. Navigating the complex interplay among sequence, temperature, and catalytic function is crucial for unlocking a multitude of enzyme applications. Machine learning (ML) has recently emerged as a tool for the quantitative prediction of enzyme activity from protein sequences. Unfortunately, ML models designed to predict the comprehensive enzyme activity parameter, kcat/Km, from protein sequences are rare. Combining both protein sequence and temperature as input features further challenges predictions. In this presentation, we will first discuss our experimental combinatorial approach to engineer a hyperthermophilic β-glucosidase variant with enhanced activity at low temperatures. Our study revealed that these hyperthermophilic enzymes utilize amino acids that are relatively underrepresented across various β-glucosidases. We will then discuss the development of our multi-module ML framework that predicts β‑glucosidase kcat/Km values based on protein sequence and temperature. Each module is designed to capture a distinct aspect of the interplay among protein sequence, temperature, and kcat/Km for β‑glucosidase activity; when integrated, they form an ML framework that maps the sequence and temperature spaces associated with β‑glucosidase kcat/Km. This modular approach allowed for optimizations of ML models within each module, collectively achieving notable generalization performance when predicting temperature-dependent kcat/Km values for protein sequences not encountered during training. Our results highlight that our unique multi-module ML framework may serve as a valuable architecture for predicting highly complex catalytic variables across protein sequence and temperature spaces.
