The training30_DNA, training30_RNA and training30_protein datasets include proteins that were used to train hybridNAP for the prediction on the DNA_T, RNA_T and protein_T datasets, respectively.

Each file includes list of proteins (identified using their UniProt ID and the corresponding PDB chains) for which their PDB chains have < 30% sequence similarity with the proteins (PDB chains) in the DNA_T, RNA_T and protein_T datasets, respectively.
Each line includes three parts separated by '_':
- The first part is the UniProt ID
- The second part is the corresponding PDB ID
- The third part is the binding type