Figure 3 : MedianCD and SVCD normalization allowed to detect much larger numbers of differentially expressed genes (DEGs) in the datasets with differential gene expression.

From: Variation-preserving normalization unveils blind spots in gene expression profiling

Figure 3

Panels show results for the real dataset (a), synthetic dataset with differential gene expression (b), and synthetic dataset without differential gene expression (c). They display the number of DEGs for each treatment compared to the corresponding control, obtained after applying the four normalization methods (empty black circles, Median normalization; empty red up triangles, Quantile normalization; filled green circles, MedianCD normalization; filled blue up triangles, SVCD normalization). For the synthetic dataset with differential gene expression (b), the numbers of treatment positives are also shown, as empty black down triangles. In each panel, treatments are ordered according to the number of DEGs identified with SVCD normalization, increasing from left to right (see Supplementary Table 2, for real dataset). Differential gene expression was analyzed with R/Bioconductor package limma. Supplementary Fig. S2 shows results obtained with t-tests, qualitatively similar but with much lower detection of differential gene expression.