Experimental comparison and evaluation of the Affymetrix exon and U133Plus2 GeneChip arrays

Abstract

Background: Affymetrix exon arrays offer scientists the only solution for exon-level expression profiling at the whole-genome scale on a single array. These arrays feature a new chip design with no mismatch probes and a radically new random primed protocol to generate sense DNA targets along the entire length of the transcript. In addition to these changes, a limited number of validating experiments and virtually no experimental data to rigorously address the comparability of all-exon arrays with conventional 3'-arrays result in a natural reluctance to replace conventional expression arrays with the new all-exon platform.

Methodology: Using commercially available Affymetrix arrays, we assess the performance of the Human Exon 1.0 ST (HuEx) and U133 Plus 2.0 (U133Plus2) platforms directly through a series of 'spike-in' hybridizations containing 25 transcripts in the presence of a fixed eukaryotic background. Specifically, we compare the measures of expression for HuEx and U133Plus2 arrays to evaluate the precision of these measures as well as the specificity and sensitivity of the measures' ability to detect differential expression.

Significance: This study presents an experimental comparison and systematic cross-validation of Affymetrix exon arrays and establishes high comparability of expression changes and probe performance characteristics between Affymetrix conventional and exon arrays. In addition, this study offers a reliable benchmark data set for the comparison of competing exon expression measures, the selection of methods suitable for mapping exon array measures to the wealth of previously generated microarray data, as well as the development of more advanced methods for exon- and transcript-level expression summarization.

Figure 1. A) Distribution of array intensities for all probesets in HuEx 1.0 [red] and U133Plus2 [blue].

B) Signal response to spike-in concentrations in HuEx 1.0 [red] and U133Plus2 [blue].

Figure 2. A) Model fit for HuEx [red] and U133Plus2 [blue].

After each probe has been fit to the model, described in Material and Methods, the rescaled variables overlay the classical Langmuir adsorption isotherm, shown in dotted black. B) Distribution of Initial Slopes for HuEx [red boxplot] and U133Plus2 [blue boxplot] spiked probes. C) Local slopes as a function of concentration for HuEx [median red line and inter-quartile range shaded red] vs U133Plus2 [median blue line and inter-quartile range shaded blue]. Local slopes represent the fitted log2 fold-change for probes with true log2 fold-change of 1 as a function of the nominal concentration. D) Distribution of detection threshold for HuEx 1.0 [red boxplot] and U133Plus2 [blue boxplot] spiked probes. Detection threshold is referred to probe concentration where signal-to-noise ratio exceeds 3.

Figure 3. A) Summarized expression measures response to log2 concentration in HuEx 1.0 [red] and U133Plus2 [blue].

B) Expression bias as a function of log2 concentration in HuEx 1.0 [red] and U133Plus2 [blue].