Renée X. de Menezes

In the design of microarray or next-generation sequencing experiments it is crucial to choose the appropriate number of biological replicates. As often the number of differentially expressed genes and their effect sizes are small and too few replicates will lead to insufficient power to detect these. On the other hand, too many replicates unnecessary leads to high experimental costs. Power and sample size analysis can guide experimentalist in choosing the appropriate number of biological replicates. Several methods for power and sample size analysis have recently been proposed for microarray data. However, most of these are restricted to two group comparisons and require user-defined effect sizes. Here we propose a pilot-data based method for power and sample size analysis which can handle more general experimental designs and uses pilotdata to obtain estimates of the effect sizes. The method can also handle?2 distributed test statistics which enables power and sample size calculations for a much wider class of models, including high-dimensional generalized linear models which are used, e.g., for RNA-seq data analysis. The performance of the method is evaluated using simulated and experimental data from several microarray and next-generation sequencing experiments. Furthermore, we compare our proposed method for estimation of the density of effect sizes from pilot data with a recent proposed method specific for two group comparisons. © 2013 Walter de Gruyter GmbH, Berlin/Boston

Genes that play an important role in tumorigenesis are expected to show association between DNA copy number and RNA expression. Optimal power to find such associations can only be achieved if analysing copy number and gene expression jointly. Furthermore, some copy number changes extend over larger chromosomal regions affecting the expression levels of multiple resident genes.

Background: The comparability of gene expression data generated with different microarray platforms is still a matter of concern. Here we address the performance and the overlap in the detection of differentially expressed genes for five different microarray platforms in a challenging biological context where differences in gene expression are few and subtle.Results: Gene expression profiles in the hippocampus of five wild-type and five transgenic δC-doublecortin-like kinase mice were evaluated with five microarray platforms: Applied Biosystems, Affymetrix, Agilent, Illumina, LGTC home-spotted arrays. Using a fixed false discovery rate of 10% we detected surprising differences between the number of differentially expressed genes per platform. Four genes were selected by ABI, 130 by Affymetrix, 3,051 by Agilent, 54 by Illumina, and 13 by LGTC. Two genes were found significantly differentially expressed by all platforms and the four genes identified by the ABI platform were found by at least three other platforms. Quantitative RT-PCR analysis confirmed 20 out of 28 of the genes detected by two or more platforms and 8 out of 15 of the genes detected by Agilent only. We observed improved correlations between platforms when ranking the genes based on the significance level than with a fixed statistical cut-off. We demonstrate significant overlap in the affected gene sets identified by the different platforms, although biological processes were represented by only partially overlapping sets of genes. Aberrances in GABA-ergic signalling in the transgenic mice were consistently found by all platforms.Conclusion: The different microarray platforms give partially complementary views on biological processes affected. Our data indicate that when analyzing samples with only subtle differences in gene expression the use of two different platforms might be more attractive than increasing the number of replicates. Commercial two-color platforms seem to have higher power for finding differentially expressed genes between groups with small differences in expression.

Background: Functional analysis of data from genome-scale experiments, such as microarrays, requires an extensive selection of differentially expressed genes. Under many conditions, the proportion of differentially expressed genes is considerable, making the selection criteria a balance between the inclusion of false positives and the exclusion of false negatives. Results: We developed an analytical method to determine a p-value threshold from a microarray experiment that is dependent on the quality and design of the data set. To this aim, populations of p-values are modeled as mathematical functions in which the parameters to describe these functions are estimated in an unsupervised manner. The strength of the method is exemplified by its application to a published gene expression data set of sporadic and familial breast tumors with BRCA1 or BRCA2 mutations. Conclusion: We present an objective and unsupervised way to set thresholds adapted to the quality and design of the experiment. The resulting mathematical description of the data sets of genome-scale experiments enables a probabilistic approach in systems biology.

Background: In this study, we investigated the effect of genetic background on expression profiles. We analysed the transcriptome of mouse hindlimb muscle of five frequently used mouse inbred strains using spotted oligonucleotide microarrays. Results: Through ANOVA analysis with a false discovery rate of 10%, we show that 1.4% of the analysed genes is significantly differentially expressed between these mouse strains. Differential expression of several of these genes has been confirmed by quantitative RT-PCR. The number of genes affected by genetic background is approximately ten-fold lower than the number of differentially expressed genes caused by a dystrophic genetic defect. Conclusions: We conclude that evaluation of the effect of background on gene expression profiles in the tissue under study is an effective and sensible approach when comparing expression patterns in animal models with heterogeneous genetic backgrounds. Genes affected by the genetic background can be excluded in subsequent analyses of the disease-related changes in expression profiles. This is often a more effective strategy than backcrossing and inbreeding to obtain isogenic backgrounds.

In this paper, we model the epidemic course of a pathogen infection within a semi-closed group which generates clinical signals which do not necessarily permit its ready and certain identification. Typical examples of such a pathogen are influenza-type viruses. We allow for time-varying infectivity levels among individuals, and model the probability of infection per contact as a function of the clinical signals. In order to accomplish this, we introduce a modified chain-binomial Reed-Frost model. We obtain an expression for the basic reproduction ratio and determine conditions which guarantee that the epidemic does not survive in the long-term. These conditions being functions of the signal's distribution, they can be used to design and evaluate interventions, such as treatment protocols.

In statistical models of dependence, the effect of a categorical variable is typically described by contrasts among parameters. For reporting such effects, quasi‐variances provide an economical and intuitive method which permits approximate inference on any contrast by subsequent readers. Applications include generalised linear models, generalised additive models and hazard models. The present paper exposes the generality of quasi‐variances, emphasises the need to control relative errors of approximation, gives simple methods for obtaining quasi‐variances and bounds on the approximation error involved, and explores the domain of accuracy of the method. Conditions are identified under which the quasi‐variance approximation is exact, and numerical work indicates high accuracy in a variety of settings.

The analysis of differential gene expression in microarray experiments requires the development of adequate statistical tools. This article describes a simple statistical method for detecting differential expression between two conditions with a low number of replicates. When comparing two group means using a traditional t-test, gene-specific variance estimates are unstable and can lead to wrong conclusions. We construct a likelihood ratio test while modelling these variances hierarchically across all genes, and express it as a t-test statistic. By borrowing information across genes we can take advantage of their large numbers, and still yield a gene-specific test statistic. We show that this hierarchical t-test is more powerful than its traditional version and generates less false positives in a simulation study, especially with small sample sizes. This approach can be extended to cases where there are more than two groups.