Enzyme-Substrate Pair Prediction
This webserver implements a tool for the prediction of enzyme-substrate pairs. Our model takes the amino acid sequence of an enzyme and a representations of the metabolite (InChI string, KEGG Compound ID, or SMILES string) as its input. The output of the model is a score between 0 and 1, where a score close to 1 means that the metabolite is likely to be a substrate for the given enzyme, and a score close to 0 means that the metabolite is unlikely to be a substrate. Scores between 0.4 and 0.6 indicate that the model cannot make a reliable prediction.
This online tool can be used to make predictions for single enzyme-metabolite pairs or for multiple enzyme-metabolite pairs. Please note that the prediction accuracy is only high for metabolites that were present in our training set (you can download a full list here). For more details on how to use this tool, see our Help page.
For details about the prediction model, see our paper A general model to predict small molecule substrates of enzymes based on machine and deep learning.