Skip to main content

Table 2 Comparison of different strategies

From: Strategies for measuring evolutionary conservation of RNA secondary structures

   Structural CLUSTAL W
Method Variant Low Medium High Low Medium High
Energy based SCI 0.79 0.32 0.95 0.70 1.00 1.00 0.79 0.31 0.80 0.42 0.90 0.72
  RNAeval 0.82 0.43 0.86 0.45 1.00 0.99 0.82 0.42 0.76 0.32 0.90 0.68
Base-pair distance consensus 0.80 0.28 0.93 0.56 1.00 0.99 0.79 0.27 0.85 0.40 0.92 0.79
  pairwise 0.83 0.28 0.90 0.54 0.99 0.98 0.83 0.27 0.81 0.40 0.90 0.78
Mountain metric consensus 0.78 0.34 0.82 0.38 0.92 0.63 0.78 0.34 0.73 0.29 0.80 0.41
  pairwise 0.79 0.29 0.75 0.33 0.76 0.34 0.79 0.29 0.73 0.30 0.75 0.34
Tree editing consensus, full 0.77 0.32 0.88 0.44 0.99 0.95 0.77 0.32 0.77 0.31 0.86 0.60
  consensus, HIT 0.76 0.30 0.89 0.46 0.99 0.97 0.76 0.28 0.78 0.33 0.87 0.60
  consensus, coarse grained 0.71 0.22 0.81 0.34 0.95 0.73 0.72 0.21 0.74 0.26 0.83 0.45
  consensus, w. coarse grained 0.74 0.26 0.84 0.36 0.98 0.88 0.74 0.25 0.73 0.28 0.82 0.46
  pairwise, full 0.78 0.31 0.77 0.36 0.88 0.63 0.78 0.31 0.75 0.34 0.87 0.56
  pairwise, HIT 0.77 0.27 0.77 0.36 0.90 0.66 0.76 0.26 0.76 0.34 0.89 0.63
  pairwise, coarse grained 0.72 0.16 0.68 0.23 0.74 0.24 0.72 0.16 0.68 0.22 0.78 0.30
  pairwise, w. coarse grained 0.76 0.23 0.71 0.28 0.81 0.41 0.75 0.15 0.71 0.23 0.82 0.35
  pairwise, MiGaL-Layer 0 0.62 0.07 0.61 0.07 0.67 0.06 0.62 0.07 0.60 0.06 0.66 0.04
  pairwise, MiGaL-Layer 1 0.74 0.27 0.68 0.24 0.77 0.33 0.74 0.27 0.68 0.24 0.76 0.33
  pairwise, MiGaL-Layer 2 0.74 0.23 0.70 0.29 0.82 0.42 0.73 0.22 0.69 0.27 0.78 0.37
  pairwise, MiGaL-Layer 3 0.76 0.27 0.71 0.30 0.84 0.49 0.75 0.26 0.71 0.29 0.82 0.47
Ensemble distance consensus 0.64 0.32 0.61 0.15 0.72 0.25 0.63 0.31 0.60 0.14 0.70 0.24
  pairwise 0.65 0.42 0.61 0.15 0.72 0.26 0.65 0.32 0.61 0.30 0.72 0.31
Mountain metric using base-pair probabilities consensus 0.48 0.17 0.58 0.27 0.65 0.40 0.50 0.18 0.56 0.24 0.61 0.28
  pairwise 0.78 0.32 0.75 0.34 0.76 0.31 0.79 0.32 0.72 0.30 0.74 0.31
RNApdist-like consensus 0.76 0.28 0.79 0.37 0.89 0.44 0.76 0.27 0.73 0.30 0.74 0.25
  pairwise 0.75 0.25 0.78 0.36 0.86 0.45 0.75 0.24 0.73 0.28 0.78 0.30
  1. Low, Medium and High refer to the same information content categories as shown in Fig. 2. Each column consists of two numbers: Left: average AUC, Right: sensitivity at 5% false positive rate. Bold numbers indicate the most accurate method in each category.