puma: a Bioconductor package for propagating uncertainty in microarray analysis

Abstract

BackgroundMost analyses of microarray data are based on point estimates of expression levels and ignore the uncertainty of such estimates. By determining uncertainties from Affymetrix GeneChip data and propagating these uncertainties to downstream analyses it has been shown that we can improve results of differential expression detection, principal component analysis and clustering. Previously, implementations of these uncertainty propagation methods have only been available as separate packages, written in different languages. Previous implementations have also suffered from being very costly to compute, and in the case of differential expression detection, have been limited in the experimental designs to which they can be applied.Resultspuma is a Bioconductor package incorporating a suite of analysis methods for use on Affymetrix GeneChip data. puma extends the differential expression detection methods of previous work from the 2-class case to the multi-factorial case. puma can be used to automatically create design and contrast matrices for typical experimental designs, which can be used both within the package itself but also in other Bioconductor packages. The implementation of differential expression detection methods has been parallelised leading to significant decreases in processing time on a range of computer architectures. puma incorporates the first R implementation of an uncertainty propagation version of principal component analysis, and an implementation of a clustering method based on uncertainty propagation. All of these techniques are brought together in a single, easy-to-use package with clear, task-based documentation.ConclusionFor the first time, the puma package makes a suite of uncertainty propagation methods available to a general audience. These methods can be used to improve results from more traditional analyses of microarray data. puma also offers improvements in terms of scope and speed of execution over previously available methods. puma is recommended for anyone working with the Affymetrix GeneChip platform for gene expression analysis and can also be applied more generally.

Transcriptomic Analysis of Triclosan-Susceptible and -Tolerant Escherichia coli O157:H19 in Response to Triclosan Exposure

“…Microarray data were preprocessed using the mmgMOS normalization method 25 using the default settings, and differential expression (DE) was calculated using the puma DE method, both of which are implemented in the Bioconductor package ''puma.'' 25 The puma method uses a Bayesian hierarchical model to calculate the probability of positive likelihood ratio (PPLR). The PPLR associates probability values of genes being differentially expressed, which is a measure of the false-positive detection of DE, to each ratio and generates lists of genes ranked by the probability of DE.…”

Dietary n-3 polyunsaturated fatty acid supplementation alters the expression of genes involved in the control of fertility in the bovine uterine endometrium

“…Data were quality assessed before and after normalization using a number of in-built quality control methods implemented in the Bioconductor affycoretools and associated packages to identify problems if they existed with microarray hybridization, RNA degradation, and data normalization. Microarray data from all 14 arrays were preprocessed by the mmg-MOS normalization method (40,61) employing the default settings and differential expression (DE) was analyzed by the puma-DE method both implemented in the Bioconductor package "puma" (44,60,61,67). The puma method uses a Bayesian hierarchical model to calculate the probability of positive likelihood ratio (PPLR) of differential expression.…”

Model-based method for transcription factor target identification with limited data

“…Drosophila developmental microarray time series from ref. 21 were reprocessed using mmgMOS from the puma package (17). The means of the log-scale expression values were equalized across chips.…”

“…ijm , where the gene and condition specific measurement variance parameters σ 2 if and σ 2 ijm are obtained from a probe-level analysis of the microarray data (16,17) (see Materials and Methods). The log likelihood for the model parameters θ can then be calculated exactly using standard Gaussian process regression techniques to integrate out the functions m and f (9).…”