Skip to main content

Table 3 The performance of CoreBoost classifiers

From: Histone modification profiles are predictive for tissue/cell-type specific expression of both protein-coding and microRNA genes

    Sensitivity PPV F-Score
HSMM Promoter CpG 0.929 ± 0.098 0.467 ± 0.012 0.619 ± 0.033
   non-CpG 0.764 ± 0.214 0.617 ± 0.043 0.662 ± 0.123
  Body CpG 0.919 ± 0.135 0.473 ± 0.022 0.618 ± 0.043
   non-CpG 0.860 ± 0.183 0.608 ± 0.025 0.700 ± 0.093
miRNA Promoter Set I 0.417 ± 0.037 0.441 ± 0.036 0.426 ± 0.011
   Set II 0.365 ± 0.073† 0.521 ± 0.085 0.422 ± 0.079†
  Body Set I 0.255 ± 0.080 0.220 ± 0.068† 0.234 ± 0.068†
   Set II 0.125 ± 0.057* 0.099 ± 0.069* 0.102 ± 0.056*
  1. In the "HSMM" rows, the classifiers were trained based on eight HMV features in CD4+ T cells, in the proximal promoter and gene body region, respectively. In the "miRNA" rows, the classifiers were trained on full Set I or Set II HMV features. Averages and errors are given as the mean and standard deviation from 100 replicates. The significances of comparison between the performance of CoreBoost trained on features in those regions and control regions are indicated by symbols next to each number. No symbol indicates p-value < 1e-5; * indicates p-value < 1e-2 and > = 1e-5; † indicates p-value > 1e-2.