Answers are grounded in this paper and cite their sources.
Deep learning demonstrates improved accuracy for predicting Km values in enzymes, highlighting its potential for structural analysis.
The Michaelis constant (Km) is a key parameter that describes the binding affinity between an enzyme and its substrate. However, experimentally determining the Kmvalue is both difficult and time-consuming. Therefore, developing a deep learning-based method to predict enzyme Kmvalues is of great significance. In this study, we constructed a deep learning model capable of extracting three-dimensional structural information from protein and substrate structural files. The model leverages a graph neural network to deeply process this information, thereby enabling accurate prediction of enzyme Kmvalues. The model achieved an R value of 0.453 on the SABIO-RK dataset, outperforming the model proposed by Kroll A et al., and demonstrating superior accuracy and generalization capability. This research not only showcases the great potential of deep learning in structure-based prediction of enzyme Kmvalues, but also provides new perspectives and methodologies for advancing this field.
No takes yet. Share an insight, caveat, or question.
Xie et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: