Skip to main content

Table 5 Similarities between the optimal and native sequences (intrastructure) and similarities between the optimal sequences for a pair of proteins in the same HOMSTRAD family (interstructure)

From: Orientation-dependent backbone-only residue pair scoring functions for fixed backbone protein design

 

All residues

Core residues only

 

All structures

RMSD < 2.5 Ã…

RMSD ≥ 2.5 Å

All structures

RMSD < 2.5 Ã…

RMSD ≥ 2.5 Å

BP median %ID to

13.4%

13.5%

13.2%

17.6%

17.6%

17.8%

native

      

ROSETTA median

25.9%

26.5%

23.6%

35.6%

36.2%

33.2%

%ID to native

      

BP median

22.4%

24.0%

17.7%

29.4%

30.8%

25.3%

interstructure %ID

      

ROSETTA median

22.8%

25.4%

17.2%

29.8%

33.6%

21.4%

interstructure %ID

      

BP interstructure

183 (45%)

118 (41%)

65 (56%)

182 (45%)

116 (40%)

66 (57%)

%ID > ROSETTA

      

interstructure %ID

      

BP interstructure

222 (55%)

171 (59%)

51 (44%)

203 (50%)

161 (55%)

42 (36%)

%ID < ROSETTA

      

interstructure %ID

      

BP interstructure

2 (0.49%)

2 (0.70%)

0 (0%)

22 (5.4%)

14 (4.8%)

8 (6.9%)

%ID = ROSETTA

      

interstructure %ID

      
  1. The similarities were calculated as percent sequence identity (%ID). The optimal sequences were calculated using the 6D residue pair scoring functions with Belief Propagation (BP) and using the ROSETTA program. The last three rows give the number of structure pairs for which the BP sequence similarity was less than, greater than, or equal to the similarity calculated using ROSETTA. The calculations were performed for a total of 407 structure pairs, each in a different HOMSTRAD protein family.
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us