Table 7 Comparison of F1 scores on the test set for the ESM-2 token classification model and the GNN model, alongside the F1 scores of prosperousplus, obtained by training on our dataset and evaluating on our test set. A Dash (–) indicates that an error occurred during training for the corresponding protease due to the prosperousplus code failing to complete the calculations.

From: Prediction of peptide cleavage sites using protein language models and graph neural networks