Batch Effect Reduction of Microarray Data with Dependent Samples Using an Empirical Bayes Approach (BRIDGE)

• Activity Number: 244 - Statistical methods for microbiome data analysis and beyond
• Type: Contributed
• Date/Time: Wednesday, August 11, 2021 : 10:00 AM to 11:50 AM
• Sponsor: Section on Statistics in Genomics and Genetics
• Author(s): Qing Xia* and Jeffrey Thompson and Devin C Koestler
• Companies: Department of Biostatistics & Data Science, University of Kansas Medical Center and Department of Biostatistics & Data Science, University of Kansas Medical Center and University of Kansas Medical Center
• Keywords: longitudinal gene expression; temporal microarray data; batch effect correction; ComBat

Abstract

While there exist many methods for batch-effect correction of microarray data, most assume independence across samples; an assumption that is unlikely to hold in longitudinal microarray studies. We propose Batch effect Reduction of mIcroarray data with Dependent samples usinG empirical Bayes (BRIDGE), a three-step approach that leverages technical replicate samples profiled at multiple timepoints/batches, “bridge samples”, to inform batch-effect reduction/attenuation in longitudinal microarray studies. Extensive simulation studies and an analysis of a real biological data set were conducted to benchmark the performance of BRIDGE against ComBat. Our results demonstrate that while both BRIDGE and ComBat perform well in facilitating accurate estimates of time effects, BRIDGE outperforms ComBat in the removal of batch-effects, and as a result, was observed to have improved statistical power for detecting genes with a time effect. BRIDGE demonstrated competitive performance in batch effect reduction of confounded longitudinal microarray studies, in both simulated and real data, and may serve as a useful preprocessing method for researchers conducting longitudinal microarray studies.