Skip to main content

Table 1 Improvement of protein secondary structure prediction with the addition of a "residue-specific" RSA threshold using leave-one-out cross-validation, compared with this improvement using a fixed 16% RSA threshold.

From: Impact of residue accessible surface area on the prediction of protein secondary structures

  Applied Threshold
  Fixed (16%) Mean Median
A 0.67 3.93 2.25
C -5.02 -0.05 -0.42
D -1.5 -1.26 -0.94
E -2.77 1.33 -3.71
F 0.18 5.99 7.15
G 0.90 5.98 5.53
H 0.04 -4.23 -4.80
I -19.00 -16.19 -16.17
K 9.53 9.91 11.63
L 0.54 4.20 2.24
M -6.82 -7.71 -8.29
N 1.21 1.71 1.63
P 1.15 1.70 1.47
Q -0.61 -1.49 -3.38
R 1.84 0.87 1.15
S 1.29 6.44 4.85
T 0.20 3.12 2.53
V -2.57 -5.80 -8.33
W -6.18 -1.66 -2.22
Y -0.19 10.04 9.78
Total Improvement 3.46 5.79 5.13
  1. The values show percentage of improvement. The bold-underlined values are those values that show improvements when they are compared with the original GOR method. See the text for more details.