Implementation of TASSER on 1489 representative PDB benchmark proteins

Summary of folding results:

• ID --- PDB code of the proteins;
• Native --- C-alpha coordinates of native structures;
• Second --- Secondary structure classification;
  • a --- alpha-protein;
  • b --- beta-protein;
  • ab --- alpha/beta-protein;
  • Ca --- native structures with only C-alpha measured;
  • l --- protein with little secondary structures;
• Length --- Number of residues of the target proteins;
• Template --- Template files (in PDB format) by threading program PROSPECTOR_3;
• Z-score --- Z-score in PROSPECTOR_3 alignment;
• R_ali_temp --- RMSD to native for the best template file over aligned region (in Angstrom);
• Cover_ali --- Coverage of threading alignment for the lowest RMSD template;
• Class --- Target classfication according to PROSPECTOR_3:
  • Easy --- There are at least one siginficant template hit with Z-score>15 or two consensus template hits with Zscore > 8;
  • Medium --- Those sequence that either hit a single template with 7 < Zscore < 15 or hit multiple templates lacking a significant consensus region;
  • Hard --- Those sequence without reliably template hits.
• Models --- Top five models generated through clustering TASSER trajectories by the SPICKER algorithm;
  • model*.pdb: the model combined by SPICKER;
  • close*.pdb: the structure in trajectories that is closest to model*.pdb;
• R_ali_mod --- The lowest RMSD to native in top five models over the threading aligned regions (in Angstrom);
• R_mod --- The lowest RMSD to native in the top five models over entire chain (in Angstrom). The value in parentheses is the rank of the best model;
• R_top1 --- RMSD to native for the model of the highest structural density (in Angstrom).