Skip to main content

Advertisement

Table 6 Comparison of TRL with baseline RL. AUCs when RL (baseline) and TRL are applied to build a classification rule model on three datasets, Petalidis (brain), KangA (IPF), and Lapointe (prostate). SRC means the source dataset (e.g., for target Petalidis, SRC1 is Freije, see Additional File 3). The mean and the standard error of the mean (SEM) for the AUC of a dataset was obtained by 10-fold cross-validation

From: Knowledge transfer via classification rules using functional mapping for integrative modeling of gene expression data

Dataset Petalidis KangA Lapointe
  AUC (SEM) AUC (SEM) AUC (SEM)
Baseline 0.83 (0.06) 0.86 (0.07) 0.93 (0.03)
SRC1 0.82 (0.05) 0.86 (0.07) 0.93 (0.03)
SRC2 0.88 (0.07) 0.86 (0.07) 0.89 (0.05)
SRC3 0.81 (0.07) 0.93 (0.05) 0.90 (0.03)
SRC4 0.78 (0.06) 0.86 (0.07) 0.93 (0.03)
SRC5 0.85 (0.05) 0.86 (0.07) 0.91 (0.04)
SRC6 0.81 (0.07) 0.86 (0.07) 0.91 (0.04)
  1. For each dataset, positive transfer is shown in bold font, while underlined AUCs denote negative transfer