Skip to main content

Table 5 Comparison of miPred and miR-KDE in terms of the feature set and the classification mechanism.

From: Using a kernel density estimation based classifier to predict species-specific microRNA precursors

  Without the four stem-loop features1 With the four stem-loop features2
  %SE %SP %ACC %Fm %MCC %SE %SP %ACC %Fm %MCC
HU2163
   SVM 88.0% 88.0% 88.0% 88.0% 75.9% 90.7% 90.7% 90.7% 90.7% 81.5%
   RVKDE 85.2% 90.7% 88.0% 87.6% 76.0% 88.9% 92.6% 90.7% 90.6% 81.5%
NH33504
   SVM 96.7% 90.4% 93.6% 93.7% 87.3% 97.3% 91.3% 94.3% 94.4% 88.7%
   RVKDE 94.8% 93.4% 94.1% 94.1% 88.2% 95.8% 93.5% 94.7% 94.7% 89.3%
  1. The best performance among each dataset is highlighted with bold font. 1Using the 29 features in miPred. 2Using the 33 features in miR-KDE, i.e., the 29 features derived from miPred and the four stem-loop features. 3Using the HU400 dataset to predict the HU216 dataset. 4Using the HU400 dataset to predict the NH3350 dataset.