Meta-analysis methods for combining multiple expression profiles: comparisons, statistical characterization and an application guideline

Abstract

BackgroundAs high-throughput genomic technologies become accurate and affordable, an increasing number of data sets have been accumulated in the public domain and genomic information integration and meta-analysis have become routine in biomedical research. In this paper, we focus on microarray meta-analysis, where multiple microarray studies with relevant biological hypotheses are combined in order to improve candidate marker detection. Many methods have been developed and applied in the literature, but their performance and properties have only been minimally investigated. There is currently no clear conclusion or guideline as to the proper choice of a meta-analysis method given an application; the decision essentially requires both statistical and biological considerations.ResultsWe performed 12 microarray meta-analysis methods for combining multiple simulated expression profiles, and such methods can be categorized for different hypothesis setting purposes: (1) HS A : DE genes with non-zero effect sizes in all studies, (2) HS B : DE genes with non-zero effect sizes in one or more studies and (3) HS r : DE gene with non-zero effect in "majority" of studies. We then performed a comprehensive comparative analysis through six large-scale real applications using four quantitative statistical evaluation criteria: detection capability, biological association, stability and robustness. We elucidated hypothesis settings behind the methods and further apply multi-dimensional scaling (MDS) and an entropy measure to characterize the meta-analysis methods and data structure, respectively.ConclusionsThe aggregated results from the simulation study categorized the 12 methods into three hypothesis settings (HS A , HS B , and HS r ). Evaluation in real data and results from MDS and entropy analyses provided an insightful and practical guideline to the choice of the most suitable method in a given application. All source files for simulation and real data are available on the author’s publication website.

Molecular and Genetic Characterization of Depression: Overlap with Other Psychiatric Disorders and Aging

“…Many microarray meta-analytical methods are available and the best method choice for an application depends on the data structure and biological goal [28, 29]. In this study, we combine two approaches, based on their complement biological assumptions: The rth ordered p-value with one-sided correction (rOP-OC) method [28] detected genes that are differentially expressed in at least 5 out of 8 studies and with consistent direction of changes (up-regulated or down-regulated), and the random effects model (REM) detected genes by combining effect sizes across all studies.…”

CX3CR1 is dysregulated in blood and brain from schizophrenia patients

“…With increasing numbers of analyzed samples, it is important to apply normalization procedures that will balance effects that may arise from the heterogeneity in tissue regions, microarray platform, and sample quality that could collectively deteriorate the meta-analysis. Indeed, different methods have been proposed and discussed to conduct meta-analysis (Chang et al, 2013;Chen et al, 2011;Conlon et al, 2007;Lopez et al, 2008;Phan et al, 2012;Schurmann et al, 2012;Seita et al, 2012;Stevens and Doerge, 2005;Tian and Suarez-Farinas, 2013;Warnat et al, 2005).…”

Effects of aging on circadian patterns of gene expression in the human prefrontal cortex

“…The empirical P value was estimated from a null distribution of R 2 generated from 1,000 TOD-randomized expression datasets for BA11 and BA47 separately. Adaptive weighted Fisher method was adopted to combine multiple P values across brain regions (26). The q value was estimated using R package qvalue (27).…”