Skip to main content
. 2013 Sep 17;8(9):e73411. doi: 10.1371/journal.pone.0073411

Table 3. A selected list of 16 binary protein complexes from the “difficult” and “medium difficult” category of the protein docking benchmark dataset version 4.0 [39].

PDB-ID ChainId1:ChainId2 Difficulty class(M = Medium Difficulty,D = Difficult) L-RMSD of best modelwithout XLs Total number of virtualinter-protein XLs Lowest L-RMSD amongthe 10 best modelsusing 7 randominter-protein XLs L-RMSD of best modelusing 7 random inter-protein XLs
1MQ8 A:B M 55.1 23 4.3 4.3
1JK9 A:B D 40.3 31 6.8 7.2
2CFH A:C M 38.2 16 7.1 7.6
2NZ8 A:B M 24.1 35 7.9 7.9
1ATN A:D D 44.7 9 8.0 8.0
1FQ1 A:B D 31.5 28 8.0 8.0
1I2M A:B M 30.9 25 8.4 8.4
2Z0E A:B M 25.0 12 7.5 8.5
1F6M A:C D 52.1 20 6.0 8.6
1BKD R:S D 34.5 43 9.0 9.0
2J7P A:D M 39.8 31 9.8 9.8
1HE8 B:A M 90.8 38 10.7 10.7
1IBR A:B D 47.3 49 13.0 13.0
1Y64 A:B D 55.8 65 19.7 24.9
1IRA Y:X D 35.7 26 34.4 34.4
1H1V A:G D 83.0 20 42.4 42.4

Each complex has more than 7 predicted (virtual) inter-protein cross-links and was employed to test the impact of cross-links on protein docking calculations. Best models correspond to the models with the shortest mean SAS distance for all 7 cross-links. L-RMSD corresponds to the RMSD value among the smaller protein partners also known as ligands.