Skip to main content

Table 2 Overall performance comparison of 5 gene lists in the GB, BC and OC validation datasets

From: Splitting random forest (SRF) for determining compact sets of genes that distinguish between cancer subtypes

Cancer Methods No. of Genes Prediction Accuracy Change in Prediction Accuracy+ Multi-Class AUC++ Area Covered by Radar Chart+++
GB SRF50 36 80.1%   0.87 1.68
  Single RF 88 77.8% −2.3% 0.86 1.63
  Verhaak et al. 833 86.0% 5.9% 0.92 1.91
  ANOVA 8670 84.1% 4.0% 0.9 1.83
  Top 50 ANOVA 50 77.2% −2.9% 0.87 1.71
BC SRF50 48 84.0%   0.91 1.85
  Single RF 46 74.1% −9.9% 0.87 1.69
  Parker et al. 979 89.0% 5.0% 0.89 1.78
  ANOVA 4976 85.2% 1.2% 0.86 1.65
  Top 50 ANOVA 50 82.7% −1.3% 0.87 1.72
OC SRF50 189 89.8%   0.96 2.06
  Single RF 245 88.9% −0.9% 0.96 2.06
  Tothill et al. 2106 91.7% 1.9% 0.97 2.11
  ANOVA 7144 90.7% 0.9% 0.97 2.09
  Top 50 ANOVA 50 87.0% −2.8% 0.95 2.01
  1. +All changes of prediction accuracy were calculated using SRF50 as the referent; ++ Multi-class AUC values were calculated by taking average of the all pairwise AUC values; +++ Area covered by radar chart (ACRC) [29].