Figure 5 | Scientific Reports

Figure 5

From: How to do quantile normalization correctly for gene expression data analyses

Figure 5

Feature-selection performance based on combining datasets that have been normalized separately given increasing class effect proportion (from 0 to 0.8) (A) distribution of F-scores Best performing method in the combination scenario is marked with a red *. (B) Feature-selection performance (F-Score) and Batch effect correction (Delta) based on the data combination scenario across 3 different class effect proportion (CEP) levels Performance values are summarized as ranks, where 1 is the best (highlighted in red or green), and 5 is the worst (c.f. Fig. 3).

Back to article page