Skip to main content

Table 10 Coverage evaluation. Average short, medium (Med), and long range coverage values at δ = 0 resolution for the top L/10 contact predictions from various predictors (with our methods listed in bold font) evaluated on CASP11 targets and sorted by short range scores

From: Predicting protein residue-residue contacts using random forests and deep networks

Method Short Med Long
CONSIP2 (94) 22.54% 15.66% 4.07%
RaptorX-Contact (94) 17.27% 12.03% 2.21%
rf_full (78) 14.45% 7.62% 1.68%
MULTICOM-NOVEL (94) 13.37% 5.66% 0.44%
rf_select (78) 13.18% 6.75% 1.51%
sda_Ensemble (78) 11.09% 5.81% 1.05%
sda_balanced (78) 10.20% 4.47% 0.58%
svm (78) 9.53% 4.33% 1.18%
PLCT (91) 7.25% 3.59% 0.36%
Pcons-net (87) 7.00% 7.50% 2.73%
raghavagps-paaint (76) 5.05% 3.60% 0.65%
DCA (88) 5.02% 6.74% 2.20%
IASL-COPE (91) 4.41% 3.93% 1.24%
sda_unbalanced (78) 3.87% 3.74% 0.91%
FoDTcm (32) 2.06% 0.98% 0.15%
MLiD (94) N/A 9.69% 1.81%