Table 5 Medium range accuracy evaluation 2. Average medium range accuracy values at δ = 1 resolution for the top 10, top L/10, and top L/5 contact predictions from various predictors (with our methods listed in bold font) evaluated on CASP11 targets and sorted by Top 10 accuracy scores
From: Predicting protein residue-residue contacts using random forests and deep networks
Method | Top 10 | L/10 | L/5 |
|---|---|---|---|
CONSIP2 (94) | 80.74% | 77.50% | 73.01% |
RaptorX-Contact (94) | 77.77% | 73.34% | 68.28% |
MLiD (94) | 66.49% | 63.34% | 58.86% |
rf_full (78) | 61.15% | 60.06% | 56.63% |
rf_select (78) | 56.41% | 53.23% | 49.81% |
MULTICOM-NOVEL (94) | 54.84% | 53.89% | 52.80% |
PLCT (91) | 51.93% | 53.02% | 53.36% |
Pcons-net (87) | 49.87% | 48.60% | 48.02% |
sda_Ensemble (78) | 48.08% | 46.78% | 43.54% |
svm (78) | 47.18% | 43.87% | 42.41% |
IASL-COPE (91) | 45.78% | 42.75% | 42.55% |
DCA (88) | 42.32% | 36.06% | 31.16% |
sda_balanced (78) | 40.90% | 39.06% | 37.39% |
sda_unbalanced (78) | 33.08% | 33.39% | 34.02% |
raghavagps-paaint (76) | 32.63% | 32.22% | 32.48% |
FoDTcm (32) | 11.25% | 17.87% | 18.87% |