Skip to main content

Table 1 Performance comparison between EDLMFC and other ncRPI prediction methods on RPI1807, NPInter v2.0, and RPI488

From: EDLMFC: an ensemble deep learning framework with multi-scale features combination for ncRNA–protein interaction prediction

Dataset Method ACC (%) TPR (%) TNR (%) PPV (%) F1 (%) MCC (%) AUC (%)
RPI1807 EDLMC 93.8 ± 0.3 96.9 ± 0.3 84.5 ± 0.9 94.9 ± 0.3 95.9 ± 0.2 83.3 ± 0.8 96.7 ± 0.3
  RPITER 93.5 ± 0.4 97.1 ± 0.4 82.7 ± 1.1 94.3 ± 0.3 95.7 ± 0.2 82.4 ± 1.0 97.7 ± 0.3
  IPMinter 93.5 ± 0.3 99.2 ± 0.4 76.8 ± 2.4 92.7 ± 0.7 95.8 ± 0.2 82.6 ± 0.9 88.0 ± 1.0
  CFRP 92.8 ± 0.4 97.6 ± 0.4 77.4 ± 0.6 92.7 ± 0.3 95.2 ± 0.3 79.7 ± 0.9 96.4 ± 0.1
NPInter v2.0 EDLMFC 89.7 ± 0.2 91.7 ± 0.4 87.7 ± 0.4 88.2 ± 0.3 89.9 ± 0.2 79.5 ± 0.4 95.9 ± 0.2
  RPITER 89.0 ± 0.6 91.6 ± 0.6 86.2 ± 0.1 87.0 ± 0.8 89.3 ± 0.6 78.1 ± 1.2 95.7 ± 0.4
  IPMinter 82.8 ± 1.0 84.3 ± 0.9 81.3 ± 2.6 81.3 ± 1.3 83.2 ± 0.9 65.6 ± 2.0 82.7 ± 1.0
  CFRP 82.1 ± 0.3 77.2 ± 0.5 86.9 ± 0.3 85.5 ± 0.3 81.1 ± 0.3 64.4 ± 0.5 88.4 ± 0.2
RPI488 EDLMC 86.1 ± 0.5 74.5 ± 0.8 96.7 ± 0.5 96.1 ± 0.4 82.9 ± 0.6 74.2 ± 0.9 89.9 ± 0.3
  RPITER 86.0 ± 1.0 75.1 ± 1.1 95.6 ± 1.9 95.3 ± 1.8 82.9 ± 1.1 74.0 ± 1.9 88.5 ± 0.7
  IPMinter 79.9 ± 0.8 84.6 ± 0.9 78.6 ± 1.9 79.4 ± 1.3 79.5 ± 0.9 63.7 ± 1.6 81.6 ± 0.9
  CFRP 79.9 ± 2.0 75.3 ± 1.5 85.3 ± 2.7 82.2 ± 2.8 77.1 ± 2.3 60.8 ± 4.3 86.0 ± 1.8
  1. The values in bold indicate this performance metric is the best among the four methods
  2. The mathematical notation (±) represents standard deviation
\