Skip to main content

Advertisement

Table 2 Results on the BaCelLo sets

From: SCLpredT: Ab initio and homology-based prediction of subcellular localization by N-to-1 neural networks

  SCLpred   BaCelLo   MSA   MSA+HOM  
     Fungi     
  Spec Sens Spec Sens Spec Sens Spec Sens
cyto 0.46 0.39 0.39 0.60 0.74 0.30 0.72 0.34
mito 0.72 0.78 0.72 0.81 0.79 0.87 0.75 0.92
nucl 0.83 0.82 0.85 0.67 0.80 0.94 0.83 0.94
secr/extr 0.86 0.85 0.85 0.94 0.99 0.72 0.99 0.62
GC 0.67   0.66   0.69 (0.01) 0.68 (0.014)
Q 0.75   0.70   0.80 (0.008) 0.81 (0.007)
     Animals     
  Spec Sens Spec Sens Spec Sens Spec Sens
cyto 0.58 0.54 0.41 0.65 0.66 0.33 0.72 0.45
mito 0.77 0.74 0.66 0.76 0.68 0.93 0.69 0.98
nucl 0.83 0.85 0.85 0.65 0.77 0.92 0.83 0.92
secr/extr 0.93 0.93 0.91 0.91 0.97 0.86 0.95 0.91
GC 0.72   0.67   0.76 (0.007) 0.81 (0.006)
Q 0.82   0.74   0.80 (0.006) 0.84 (0.006)
     Plants     
  Spec Sens Spec Sens Spec Sens Spec Sens
chlo 0.68 0.79 0.76 0.73 0.83 0.56 0.82 0.61
cyto 0.39 0.36 0.47 0.52 0.40 0.76 0.47 0.78
mito 0.49 0.34 0.54 0.51 0.56 0.50 0.60 0.55
nucl 0.83 0.76 0.76 0.72 0.64 0.77 0.73 0.90
secr/extr 0.89 0.85 0.65 0.85 0.67 0.40 0.76 0.41
GC 0.63   0.59   0.57 (0.022) 0.66 (0.019)
Q 0.68   0.68   0.62 (0.021) 0.67 (0.019)
  1. Results on Fungi, Animal and Plant by the MSA and MSA+HOM architectures compared against BaCelLo, measured in 10-fold cross-validation on the BaCelLo sets (see text for details). SCLpred and BaCelLo results are from (Mooney et al. 2011), Table 3. Best results are in bold. Standard deviations for MSA and MSA+HOM global results are in brackets.