Supplementary Materials

1. PDB IDs of the 933 training datase: download

2. Sequences of the 933 training datase: download

3. Classification of the 933 training dataset:

Protein classes PDB ID, Swiss-prot ID, protein name Sequences
All Alpha (83) download download
All Beta (89) download download
Unstructured (1) download download
Alpha/Beta (750) download download
Peptide (10) download download

All Alpha: α-helix > 30% & β-sheet < 6%;
All Beta: α-helix < 6% & β-sheet > 30%;
Unstructured: α-helix < 20% & β-sheet < 10%;
Alpha/Beta: other secondary structure content;
Peptide: # residue < 41.

4. Calculated contact order database for TREMBL dataset. Each record contains "ID, Abs_CO, CO, Length, 2-state/Multi-state Description" (15,999,775 sequences. 500 MB): download

 

Access count: (Since Oct. 1, 2007)