Combined Statistics for Differential Expression Analysis of RNA-Sequencing Data

Abstract

Nowadays, genome-wide expression differences between various experimental conditions are mainly monitored using RNA-sequencing. Albeit in active use for over a decade and great progress in RNA-Seq analytics, experts have not been yet able to eliminate its technical and systematic biases, inherent to every high-throughput experimental technique. The vast majority of the attempts made towards confronting RNA-sequencing data analysis challenges are primarily focusing on the development of new analysis methods. However, less effort has been devoted in combined statistical analysis approaches. Here, we present the latest developments in PANDORA, a p-value combination algorithm, implemented in the metaseqR Bioconductor package. PANDORA was proved to successfully combine results of differential expression analysis algorithms. Its power is further enhanced by more recent and powerful algorithms in order enhance clarity of the reported differentially expressed gene lists.