( elation coefficients for other Signal c...)The New Affymetrix GeneChip Scanner 3000

Figure 1 shows scatter plots generated for these analyses. The high correlation of Signal values generated by the GeneChip Scanner 3000 and the GeneArray 2500 at the same and different external sites is illustrated by plots A and B, respectively.

Percent Present Comparisons

The percent Present metric can be a useful measure of assay and instrument performance. It describes the total number of transcripts detected (called Present) out of the total number of probe sets on a probe array.

Figure 2 illustrates the average percent Present for 12 HG-U133A arrays run on each scanner type at each external site. Error bars represent one standard deviation.

The GeneChip Scanner 3000 average percent Present is statistically different from the GeneArray 2500, with the GeneChip Scanner 3000 yielding a higher average percent Present call for probe arrays run at all sites (Figure 2, *). For probe arrays run at Site 4, the GeneArray 2500 yielded a noticeably lower percent Present when compared to the GeneChip Scanner 3000 at the same site and GeneArray 2500 scanners at other sites (Figure 2, *). While the percent Present metric of the GeneArray 2500 at Site 4 appears to be different than other GeneArray 2500 scanners, it is within Affymetrix specifications and is representative of the range of GeneArray 2500 scanner performance across Affymetrix customer sites.

False Change Comparisons

False Change is a metric used to characterize reproducibility of results when scanning the same target sample hybridized to different probe arrays of the same design. False Change is defined as the frequency of the sum of Increa
'"/>

Source: