Figure 7.

Residual standard deviation against expression intensities. Standard deviation of the residuals observed for the regression fitted to the expression intensities are plotted against minimum (upper row) and mean (lower row) expression intensity of each probe. The blue line represents a loess-curve fitted to the values. bg_rma_log_quantile exhibits very high deviation of residuals in ranges of lower expression intensities, whereas noBg_vst_rsn shows homogeneous and low deviations of residuals over the whole range of expression intensities. Compared to noBg_vst_rsn, residual standard deviations tend to be a bit higher and less homogeneous in small ranges of expression intensities when noBg_cubicSpline is used for pre-processing. Thus, scores of -2, 0, and 2 are assigned for bg_rma_log_quantile, noBg_cubicSpline, and noBg_vst_rsn, respectively. For an overview over all methods, see Additional files 4 and 5.

Schmid et al. BMC Genomics 2010 11:349   doi:10.1186/1471-2164-11-349
Download authors' original image