Supplementary supplementary file contains additional text and tables

Program PB-RMSD The program performs protein chain assignment by Protein Block according to RMSD & RMSDA metrics. Input - standard PDB file.

List of the PDB identifies of the learning sample - List contains PDB ID and chain ID

Protein blocks dihedral angles - Dihedral angles for 16 protein blocks taken from original paper De Brevern et. al[1]

Protein blocks cartesian coordinates - cartesian coordinates for 16 protein blocks in free format

Protein block in PDB format - The archive contains PDB files for each of 16 protein blocks

Learning sample PDB ID and chain ID - ZIP archive (231 M). Archive contain structural data on all pentapeptides of the training set. Sequence, position in the chain, structural class and RMSD to the corresponding protein block, dihedral angles. Presented clusters of protein blocks. Pentapeptides are sorted by RMSD to the nearest PB. See head of file for clarity.

Reduced aminoacid alphabets - In the files of this directory, all reduced alphabets that were used for model building. See read.me file for clarity.

Predction_benchmarking - prediction quality (Q16) for learning sample chain set

Detailed_prediction - ZIP archive (251 M). Archive contains detailed information for prediction protocol learning sample chains. See read.me for clarity.

[1] De Brevern A, Etchebest C, Hazout S. Bayesian probabilistic approach for 275 predicting backbone structures in terms of protein blocks. Proteins: Structure, 276 Function, and Bioinformatics. 2000;41(3):271–287