Using edgeR software to analyze complex RNA-Seq data

RNA-Seq is now commonly used to identify differentially-expressed genes (DEGs) between one treatment and another. However, data analyses can be challenging when there are multiple genotypes and treatments. This study proposes a methodology to discriminate gene expression from a complex RNA-Seq experiment. Data were from an experiment comparing genotypes from a ‘Honeycrisp’ × MN1974 (Malus × domestica) cross whose fruit were either not crisp at harvest, or that retained or lost crispness after storage. ‘Honeycrisp’ fruit retain crispness after months of storage, while MN1974 fruit softened. Statistical analyses of gene expression were performed using edgeR software. To identify genes related to postharvest change in crispness, the expression abundance between the ‘retain’ and the ‘lose’ group, the ‘retain’ and the ‘non-crisp’ group, and the parents were compared. The DEGs found in all three comparisons with FDR<0.05 were further selected based on 1) the log2-fold change >1, and 2) the expression threshold (CPM >1). A total of 567 genes were identified that could be associated with crispness retention.