Figure 5

Correlation of log2 expression ratios from the OneRA and TwoRA and that of Δlog2IN from the two compared samples. A&C, scatter plots of log2 expression ratios from the OneRA and the TwoRA for the (SA,SN), (DRG,SN) pairs respectively. For instance, the log2 OneRA expression ratio for the (DRG,SN) pair is log2 OneRA DRG – log2 OneRA SN. For the (SN, SA) pair, the most significant differences in gene expression (ATF3 and SPRR1A) are consistent in both protocol groups. For the (DRG,SN), the ratios are more varying, the regression line (shown in blue) appears to be shifted from the diagonal towards smaller values of expression ratios in The TwoRA. B&D, scatter plots of Δlog2IN values for the (DRG,SN) pair and the (SA,SN) pair respectively. Δlog2IN were calculated by subtracting the log2 OneRA intensity from the log2 TwoRA intensity. Points in red in 'C' are probesets where the intensity in one sample could not be shifted as much as in the other sample because the intensity cannot lie outside the dynamic range of the scanner. These are referred to as FCE (floor & ceiling effect) probesets and have varying expression ratios following TwoRA. The FCE probesets also show the most varying DRG and SN Δlog2IN values (points in red, D). Though, for these same probesets, the Δlog2IN values in the SA and the SN are fairly consistent (points in red, B).