Table 1.
The number of proteins in the training, validation and test sets for the PPI and PPI_extendedSFD datasets.
Structural information | PPI dataset | PPI_extendedSFD | ||
---|---|---|---|---|
SF + PPI | Only SF | SF + PPI | Only SF | |
Training set | 2842 | 0 | 2842 | 5961 |
Validation set | 353 | 0 | 353 | 749 |
Test set | 356 | 0 | 356 | 746 |
The PPI dataset only contains proteins for which both structural features and PPI interface annotations are available. The PPI_extendedSFD dataset contains additional proteins for which only structural features are available.