Skip to main content
Figure 4 | BMC Bioinformatics

Figure 4

From: A power law global error model for the identification of differentially expressed genes in microarray data

Figure 4

Correlation between the value of PLGEM-STN and the nominal concentration variation in comparison to competing statistics. Exp01 of the Latin Square data set was taken as the baseline to which the remaining 13 experiments were compared. The observed values of the indicated statistics are plotted against the nominal log ratio, deduced from the known spiked-in concentrations (left panels). A nominal log ratio of 0 is assumed for the remaining transcripts and a box-plot of their corresponding values of the indicated statistics is superimposed to the plot. For those cases where one of the two known spiked-in concentrations is 0, the value of the statistic is instead plotted against the non-null concentration (right panels). Red and black dots represent transcripts that are present in Exp01 or in the remaining 13 experiments, respectively.

Back to article page