%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% B-factor and local structure quality estimation %
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1. How was the local quality estimated?
The local quality was defined as the distance deviation (in Angstrom) between residue positions
in the model and the native structure. It was estimated using support vector regression that makes
use of the local structural between the model and (1) threading templates, (2) structure alignment
templates, (3) reference decoys, and (4) sequence-based secondary structure and solvent accessibility
predictions. Large-scale benchmark tests show that the estimated local quality has an average error
of 1.4 Angstrom and AUC score of 0.89 for well-predicted I-TASSER models.
Based on these tests, the local quality estimations tend to be more accurate for residues:
1) that have higher threading alignment coverage
2) that are located at alpha-helix and beta-strand regions
3) that are buried (at 25% threshold)
The estimated local quality for the model is available at the columns 68-74 in the model's PDB file
and also at the bottom of this page (the column with label RSQ_3).
2. What is B-factor?
The B-factors can be taken as indicating the stability of different parts of the structure.
Atoms with low B-factors are well-ordered and those with high B-factors are very flexible.
Normalized B-factor for a target protein is defined as z-score-based normalization of the
raw B-factor values. The normalized B-factor (called B-factor profile, BFP) is predicted using a
combination of both template-based assignment and profile-based prediction. Based on the distributions
and predictions of the BFP, residues with BFP values higher than 0 are less stable in experimental
structures. The BFP is converted back to B-factor by inverse z-score-based transformation.
The B-factor for the model is available at the columns 61-66 in the model's PDB file. Both normalized at
and raw B-factors are available at the bottom of this page (the columns with labels nBF and rBF).
3. How were the TM-score and RMSD estimated?
TM-score and RMSD are known standards for measuring the structural similarity between two structures which
are usually used to measure the accuracy of structure modeling when the native structure is known. In case
where the native structure is unknown, it becomes necessary to predict the quality of the modeling prediction,
i.e. what is the distance between the model and its native structure? To answer this question, we tried to
predict the TM-score and RMSD of the model relative to its native structure based on the predicted local
distances, with the corresponding TM-score and RMSD formula.
For more information about the local quality and B-factor predictions, please refer to the
following article:
Jianyi Yang and Yang Zhang, ResQ: A unified algorithm for estimating B-factor and residue-specific quality
of protein structure prediction, submitted, (2014).
Estimated TM-score: 0.91
Estimated RMSD: 1.58 Angstrom
Estimated local quality and B-factor:
#RES SS SA COV nBF rBF RSQ_1 RSQ_2 RSQ_3 RSQ_4
1 C E 0.25 2.00 40.83 14.76 10.24 7.31 6.92
2 C E 0.29 1.42 36.05 9.76 13.60 4.91 4.65
3 C E 0.30 0.91 31.93 5.38 11.48 3.54 3.50
4 C B 0.33 0.47 28.33 2.34 2.06 1.81 1.74
5 C E 0.35 0.64 29.73 3.53 6.03 2.51 2.43
6 C E 0.42 0.91 31.91 9.06 9.59 2.91 2.57
7 C E 0.46 0.78 30.86 9.77 9.88 3.26 2.80
8 C E 0.45 0.33 27.16 11.52 9.95 3.11 2.58
9 C B 0.51 0.25 26.53 9.92 9.48 2.44 1.87
10 C E 0.85 0.52 28.68 5.16 4.46 1.49 1.19
11 C E 0.96 0.50 28.53 2.47 2.55 1.68 1.50
12 C E 0.96 0.12 25.49 0.98 1.55 1.39 1.23
13 S B 1.00 -0.17 23.11 0.67 0.65 0.92 0.92
14 S E 1.00 -0.37 21.44 0.84 0.81 0.74 0.74
15 S B 1.00 -0.52 20.26 0.59 0.70 0.70 0.73
16 S B 1.00 -0.57 19.81 0.41 0.45 0.61 0.63
17 S B 1.00 -0.56 19.87 0.54 0.59 0.93 0.97
18 C B 1.00 -0.40 21.21 0.51 0.47 0.73 0.64
19 C B 1.00 -0.35 21.65 0.63 0.57 0.76 0.65
20 C B 1.00 -0.19 22.93 0.90 0.86 0.77 0.56
21 C B 1.00 -0.01 24.38 0.72 0.83 0.82 0.67
22 C E 0.99 0.42 27.94 1.94 1.93 1.40 1.24
23 C E 0.99 0.74 30.50 2.39 2.47 1.91 1.76
24 C E 1.00 0.49 28.44 4.26 3.36 1.68 1.35
25 C E 1.00 0.48 28.37 3.29 3.07 1.23 0.97
26 C E 1.00 0.43 27.96 0.92 0.72 0.91 0.83
27 C E 1.00 0.22 26.25 0.73 0.65 0.90 0.76
28 C B 1.00 -0.07 23.92 0.97 0.87 1.12 1.01
29 S B 0.99 -0.40 21.24 1.19 1.16 1.03 1.01
30 S B 0.99 -0.55 20.02 0.94 0.90 0.85 0.87
31 S B 0.99 -0.60 19.59 1.11 0.98 0.65 0.63
32 S B 0.99 -0.62 19.45 0.70 0.56 0.85 0.86
33 S B 0.99 -0.54 20.05 0.79 0.78 0.78 0.80
34 S B 0.99 -0.49 20.48 1.27 1.01 0.61 0.58
35 S B 0.99 -0.25 22.43 1.47 1.69 0.98 0.76
36 S E 0.99 0.16 25.81 1.89 3.02 1.32 1.17
37 C E 0.99 0.41 27.84 6.13 2.68 2.02 1.68
38 C E 0.99 0.33 27.20 2.58 2.42 1.77 1.66
39 S E 0.99 0.03 24.74 1.41 1.56 1.33 1.27
40 S E 0.99 -0.02 24.29 1.66 1.36 1.67 1.58
41 S E 0.99 -0.02 24.32 1.32 1.67 1.06 1.04
42 S E 0.99 -0.05 24.07 1.66 1.48 1.43 1.34
43 C E 0.99 0.13 25.52 1.13 1.07 1.08 1.01
44 C E 0.99 0.12 25.43 0.89 1.16 1.11 1.01
45 C E 0.99 0.41 27.86 1.33 1.67 1.11 1.02
46 C E 0.99 0.58 29.21 2.25 2.34 2.34 2.29
47 C B 1.00 0.15 25.70 1.54 1.56 1.33 1.20
48 C B 1.00 0.20 26.09 0.67 0.77 1.08 1.01
49 H B 1.00 0.13 25.57 1.05 1.13 1.18 1.09
50 H B 1.00 0.03 24.73 0.86 0.84 0.97 0.90
51 H E 1.00 -0.18 23.03 0.87 0.78 1.05 0.97
52 H E 1.00 -0.34 21.70 0.53 0.52 0.76 0.63
53 H B 1.00 -0.37 21.49 0.56 0.46 0.71 0.62
54 H B 1.00 -0.41 21.14 0.53 0.64 0.74 0.66
55 H E 1.00 -0.40 21.24 0.54 0.61 0.73 0.60
56 H B 1.00 -0.49 20.48 0.70 0.60 0.73 0.65
57 H B 1.00 -0.47 20.62 0.61 0.74 0.79 0.72
58 H B 1.00 -0.41 21.12 0.73 0.79 0.92 0.78
59 H E 1.00 -0.32 21.83 0.74 0.76 0.94 0.84
60 H B 1.00 -0.33 21.82 0.81 0.88 1.09 0.99
61 H B 1.00 -0.23 22.61 2.04 1.98 1.49 1.39
62 H E 1.00 0.14 25.58 2.72 2.83 2.66 2.46
63 C E 1.00 0.37 27.51 2.03 1.93 1.44 1.23
64 C E 1.00 0.54 28.86 3.82 2.10 2.10 1.84
65 C E 1.00 0.46 28.20 2.44 2.29 1.64 1.52
66 C B 1.00 0.14 25.62 2.14 2.03 1.35 1.21
67 S E 1.00 0.05 24.86 1.51 1.27 0.90 0.76
68 S B 1.00 -0.30 22.02 1.00 0.89 0.57 0.56
69 S E 1.00 -0.39 21.25 1.15 1.02 0.90 0.90
70 S B 1.00 -0.50 20.40 0.87 0.83 0.78 0.79
71 S B 1.00 -0.55 19.95 0.81 0.71 0.53 0.52
72 S B 1.00 -0.52 20.20 0.70 0.66 0.74 0.76
73 C B 1.00 -0.26 22.33 0.95 0.83 1.09 0.99
74 H B 1.00 -0.39 21.27 0.96 0.95 1.35 1.24
75 H E 0.99 -0.08 23.80 1.36 1.37 1.15 1.00
76 H B 1.00 -0.19 22.91 1.52 1.69 1.32 1.18
77 H B 1.00 -0.25 22.44 1.32 1.03 1.15 1.07
78 H B 1.00 -0.05 24.10 1.39 1.18 0.84 0.71
79 C E 0.99 0.00 24.49 1.77 2.00 1.21 1.00
80 C B 0.95 -0.07 23.88 1.83 1.78 1.08 0.93
81 C B 0.95 0.05 24.87 1.92 1.70 1.30 1.15
82 C E 0.93 0.22 26.25 2.27 2.19 1.32 1.09
83 C E 0.91 0.34 27.27 2.51 2.10 1.28 1.06
84 C B 0.88 0.28 26.74 3.20 1.73 1.28 0.95
85 C B 0.76 0.02 24.64 2.64 1.82 1.06 0.84
86 H E 0.75 0.28 26.78 2.50 2.11 1.20 1.06
87 H E 0.76 0.32 27.07 3.02 2.01 1.43 1.28
88 H B 0.72 0.12 25.43 3.40 1.52 1.28 1.06
89 H E 0.74 0.50 28.54 4.01 1.71 1.43 1.15
90 C E 0.75 1.00 32.60 4.52 1.73 1.82 1.52
91 C E 0.72 0.65 29.77 3.02 1.60 1.97 1.77
92 C E 0.78 0.45 28.14 2.87 2.09 1.64 1.49
93 C B 0.88 0.27 26.65 1.79 2.03 1.41 1.34
94 C E 0.90 0.18 25.94 4.07 3.79 1.49 1.27
95 C E 0.88 0.23 26.34 1.71 1.91 1.04 0.96
96 C E 0.96 0.42 27.86 2.92 2.42 1.74 1.53
97 C E 0.97 0.19 26.03 1.43 0.94 1.28 1.19
98 H E 0.96 0.17 25.87 0.97 0.97 1.14 1.01
99 H E 1.00 0.25 26.48 0.85 0.78 1.39 1.30
100 H B 1.00 -0.15 23.24 0.76 0.63 0.90 0.82
101 H B 1.00 -0.23 22.56 0.82 0.66 0.89 0.81
102 H E 0.99 -0.06 24.02 0.75 0.66 0.93 0.83
103 H E 1.00 -0.03 24.23 0.80 0.65 1.07 0.99
104 H B 0.89 -0.26 22.36 0.75 0.77 1.07 1.01
105 H B 1.00 -0.17 23.08 0.89 0.99 0.90 0.80
106 H E 0.99 0.13 25.55 1.11 1.02 1.01 0.91
107 H E 1.00 0.01 24.56 1.27 1.04 1.18 1.03
108 H B 1.00 -0.02 24.33 1.37 1.24 1.18 1.09
109 H B 0.93 0.28 26.79 1.44 1.34 1.60 1.47
110 C E 0.85 0.51 28.67 1.92 1.71 1.81 1.69
111 C E 0.88 0.45 28.11 1.62 2.03 2.21 2.17
112 C E 0.96 0.26 26.58 2.15 2.50 1.51 1.31
113 C E 1.00 0.09 25.20 0.80 0.87 0.94 0.82
114 S B 1.00 -0.25 22.42 0.99 1.02 0.72 0.70
115 S E 1.00 -0.22 22.67 1.06 1.09 0.98 0.94
116 S B 1.00 -0.37 21.42 1.10 0.90 0.97 0.94
117 S E 0.99 -0.36 21.51 0.84 0.85 0.88 0.89
118 S B 1.00 -0.31 21.95 0.86 0.94 0.76 0.72
119 S B 1.00 -0.11 23.56 0.86 1.01 0.94 0.81
120 C E 0.91 0.15 25.68 1.92 2.46 1.05 0.90
121 C B 0.95 0.31 27.04 2.20 2.14 1.07 0.89
122 C B 0.93 0.61 29.45 2.28 2.75 1.11 0.88
123 C E 0.90 0.89 31.70 2.99 3.10 1.44 1.19
124 C E 0.99 1.08 33.29 4.34 4.39 1.39 1.14
125 C E 0.99 0.71 30.24 2.25 2.36 1.06 0.91
126 C E 0.99 0.36 27.40 1.09 1.14 1.05 0.91
127 C E 0.99 0.14 25.64 0.93 0.80 1.16 0.99
128 H B 0.99 -0.01 24.40 0.98 0.99 0.77 0.68
129 H E 0.99 0.30 26.94 1.34 1.23 1.08 0.93
130 H E 0.99 0.10 25.28 1.13 0.99 1.07 0.95
131 H B 0.99 -0.24 22.52 0.75 0.78 0.80 0.72
132 H B 0.99 -0.22 22.70 1.10 1.08 1.11 0.94
133 H E 0.99 -0.02 24.30 1.05 0.92 0.96 0.83
134 H B 0.99 -0.28 22.21 0.67 0.66 0.71 0.63
135 H B 0.99 -0.32 21.85 0.70 0.81 0.69 0.61
136 H E 0.97 -0.15 23.29 1.40 1.27 1.04 0.90
137 H E 0.97 -0.00 24.44 1.66 1.25 1.14 0.97
138 H B 0.94 -0.14 23.31 2.59 1.44 1.08 0.86
139 H B 0.93 -0.11 23.55 1.95 1.53 1.11 0.95
140 H E 0.86 0.55 28.97 3.54 2.22 1.47 1.23
141 H E 0.82 0.85 31.41 5.03 1.76 1.56 1.20
142 H E 0.75 1.29 35.00 6.30 2.29 1.59 1.18
143 C E 0.57 1.98 40.66 8.45 5.78 4.99 4.77
RES: Residue number
SS: Predicted secondary structure: C - random coil; H - alpha-helix; S - beta-strand
RSA: Predicted solvent accessibility at 25% cutoff: E - exposed; B - buried
COV: Threading alignment coverage defined as the number of non-gap threading alignments on
the residue divided by the total number of threading templates
nBF: Predicted normalized B-factor (normalize by Z-score transformation)
rBF: Predicted raw B-factor
RSQ_1: Predicted local quality based on LOMETS threading templates
RSQ_2: Predicted local quality based on TM-align structure alignment templates
RSQ_3: Consensus prediction of local quality based on support vector regressions
RSQ_4: Predicted local quality based on submitted decoy structures
You are requested to cite the following article when you use the ResQ server:
Jianyi Yang and Yang zhang, ResQ: A unified algorithm for estimating B-factor and residue-specific quality
of protein structure prediction, submitted, (2014).