Skip to main content


Table 1 Assessment of DBNI scoring function on the principal interaction model and the full interaction model on Decoy Sets

From: Identification of native protein structures captured by principal interactions

Decoy SetsCASP10–133DRobotTotal/Average
Top1109 (107)177 (179)286 (286)
Z-score1.46 (1.32)4.07 (4.36)2.76 (2.84)
RMSD4.07 (4.26)0.19 (0.19)2.13 (2.22)
PC0.13 (0.12)0.50 (0.58)0.31 (0.35)
F.E.1.30 (1.20)2.95 (3.14)2.12 (2.17)
Z-scoreb0.25 (0.29)1.60 (1.87)0.92 (1.08)
PI0.26 ± 0.050.25 ± 0.050.25 ± 0.05
  1. The Top1 is the number of first-ranked native structures within the decoy sets. The Z-score is the average of Z-score of the native energy, the RMSD represents the average of the RMSD of the minimum score, and the PC is the average of Pearson correlation coefficient between the energy and RMSD from the native structure. The PI represents the average of the percentage of interactions between atoms used for the calculation of energy in the principal interaction model. The F.E. indicates the 20% fraction enrichment, and the Z-scoreb calculates the Z-score with respect to the best decoy. The last row is the number of targets in decoy sets. The values in the columns denote the performance of potential functions in the principal interaction model and the values in parenthesis depict the performance of the potential function in the full interaction model