Figure 5

The relative expression of bacterial control transcripts (n = 27) in Oryza sativa RNA samples as a function of the genomic DNA (gDNA) hybridisation intensity thresholds used to make a probe-mask file. RNA was extracted from O. sativa, labelled and hybridised to rice genome GeneChip arrays. Samples were spiked with bacterial control transcripts at either 1.8 pM or 3.6 pM (n = 3). Data were obtained from the PLEXdb database ([25, 30]; accession number OS1, submitted by T Close). Data were normalised to the median expression value of each gene across all six samples. Filled circles are scaled to the left-hand y-axis, and represent the ratio of bacterial control transcript mean signals in samples spiked at 3.6 pM and 1.8 pM, i.e. this number defines the slope of a linear regression with an expected value of 2. Unfilled circles are scaled to the right-hand y-axis and represent the bacterial control transcript mean signals in samples spiked at 3.6 pM, when the corresponding mean signal in samples spiked at 1.8 pM equals zero, i.e. this number defines the y-intercept term of a linear regression with an expected value of 0.