BACKGROUND: In genetic transcription research, gene expression is typically reported in a test sample relative to a reference sample. Laboratory assays that measure gene expression levels, from Q-RT-PCR to microarrays to RNA-Seq experiments, will compare two samples to the same genetic sequence of interest. Standard practice is to use the log(2)-ratio as the measure of relative expression. There are drawbacks to using this measurement, including unstable ratios when the denominator is small. This paper suggests an alternative estimate based on a proportion that is just as simple to calculate, just as intuitive, with the added benefit of greater numerical stability. RESULTS: Analysis of two groups of mice measured with 16 cDNA microarrays found similar results between the previously used methods and our proposed methods. In a study of liver and kidney samples measured with RNA-Seq, we found that proportion statistics could detect additional differentially expressed genes usually classified as missing by ratio statistics. Additionally, simulations demonstrated that one of our proposed proportion-based test statistics was robust to deviations from distributional assumptions where all other methods examined were not. CONCLUSIONS: To measure relative expression between two samples, the proportion estimates that we propose yield equivalent results to the log(2)-ratio under most circumstances and better results than the log(2)-ratio when expression values are close to zero.

Concerns about the reliability of expression data from microarrays inspire ongoing research into measurement error in these experiments. Error arises at both the technical level within the laboratory and the experimental level. In this paper, we will focus on estimating the spot-specific error, as there are few currently available models. This paper outlines two different approaches to quantify the reliability of spot-specific intensity estimates. In both cases, the spatial correlation between pixels and its impact on spot quality is accounted for. The first method is a straightforward parametric estimate of within-spot variance that assumes a Gaussian distribution and accounts for spatial correlation via an overdispersion factor. The second method employs a nonparametric quality estimate referred to throughout as the mean square prediction error (MSPE). The MSPE first smoothes a pixel region and then measures the difference between actual pixel values and the smoother. Both methods herein are compared for real and simulated data to assess numerical characteristics and the ability to describe poor spot quality. We conclude that both approaches capture noise in the microarray platform and highlight situations where one method or the other is superior.

Abstract Summary: Microarray technology is now routinely used to monitor genome-wide expression profiles. However, current microarray imaging and analysis packages typically require manual intervention and assumptions on alignments. Unfortunately, limitations and assumptions are typically undisclosed and methods are not published. To facilitate exploration of image data, we developed SignalViewer. This paper presents a description of the application. Availability: SignalViewer is available at http://qge.fhcrc.org/signalviewer/ Supplementary information: Screenshots are available at the above location, along with downloads for Windows 2000 and Linux (Redhat 7.2).