Skip to main content
Figure 2 | BMC Bioinformatics

Figure 2

From: Iterative rank-order normalization of gene expression microarray data

Figure 2

Effect of background subtraction method on probeset distributions. Color denotes spatial density (red: high, blue: low). Raw unprocessed log10 probe signal intensities are plotted in (A). MAS5 probesets (B) exhibit a markedly different distribution than the underlying probe-level data, with a large increase in variance at low intensities due to MM subtraction. dChip probesets (C) are similar to those of MAS5, despite the large difference in probeset summarization methodology, due to similar use of mis-match probe subtraction. IRON probesets (D), due to the use of RMA background subtraction, exhibit a distribution similar to that of the underlying probe-level data. RMA probesets (data not shown) produce a distribution highly similar to that of IRON, due to shared background subtraction methodology.

Back to article page