Additional file 1 of DGCPPISP: a PPI site prediction model based on dynamic graph convolutional network and two-stage transfer learning

Additional file 1: Table S1. Summary of datasets. Table S2. Statistics of lengths of all sequences in the experiment. Table S3. Classification of amino acids based on electrostatic and hydrophobicity. Table S4. Performance comparison with different sizes of convolution kernels. Fig. S1. Impact of different neighborhood ranges on model performance. Fig. S2. Visualization comparison of DGCPPISP and EGRET on proteins from Dset_186_72_PDB164. Fig. S3. Three examples of PPI site prediction on dataset Dset_331.