Randomization in Laboratory Procedure Is Key to Obtaining Reproducible Microarray Results

Abstract

The quality of gene expression microarray data has improved dramatically since the first arrays were introduced in the late 1990s. However, the reproducibility of data generated at multiple laboratory sites remains a matter of concern, especially for scientists who are attempting to combine and analyze data from public repositories. We have carried out a study in which a common set of RNA samples was assayed five times in four different laboratories using Affymetrix GeneChip arrays. We observed dramatic differences in the results across laboratories and identified batch effects in array processing as one of the primary causes for these differences. When batch processing of samples is confounded with experimental factors of interest it is not possible to separate their effects, and lists of differentially expressed genes may include many artifacts. This study demonstrates the substantial impact of sample processing on microarray analysis results and underscores the need for randomization in the laboratory as a means to avoid confounding of biological factors with procedural effects.

Host–Microbe Interaction Systems Biology: Lifecycle Transcriptomics and Comparative Genomics

“…During sample processing, samples were randomized in 96-well plates in order to avoid any confounding factors owing to well location, time point, or biological replicate batch effects and to minimize error owing to plate edge effects and fluid transfers. The need for randomization during sample preparation as a means to avoid, or reduce, the likelihood of confounding biological factors with processing or procedural effects has been described for microarray-based platforms [10–13]. The tight groupings of replicates and the clear separation or distance between groups of replicates, indicative of low noise in the data, was demonstrated by the published PCA plot [8] and is a testament to the solid experimental design.…”

Summarizing Probe Intensities of Affymetrix GeneChip 3' Expression Arrays Taking into Account Day-to-Day Variability

“…Replicates have to be accurately distributed in different days. Conversely, processing all the replicates in the same day will result in a systematic error, randomization being a key strategy to obtain reproducible results, as recently discussed in [12].…”

“…Therefore, an experimental design that allows to evaluate, through technical replicates and biological pooling, the experimental variability and to reduce its effects on the results by suitable statistical techniques is fundamental. In the last few years, some attempts have been reported in the literature [8], [9], [10], [11], [12].…”

Low-level maternal exposure to nicotine associates with significant metabolic perturbations in second-trimester amniotic fluid

“…Amniotic fluid samples were processed and analyzed separately from serum samples but under parallel protocols. We performed principal component analysis (PCA) to evaluate potential batch effects (Yang et al, 2008) and corrected for these effects, where necessary, using ComBat (Johnson, Li, & Rabinovic, 2007) and xMSanalyzer.…”