*2.4. Di*ff*erential Expression Analysis*

*DESeq2* software (version 1.14.1) in R was used to identify DEGs in pairwise comparisons [56]. Two series of DE analysis using the following contrast arguments was performed with the reference assembly approach in flag-leaf and panicle samples: contrast 1—the condition effect for each genotype, in other words, IL\_DvsIL\_C, and SWA\_DvsSWA\_C, and contrast 2—the genotypic effect for each condition, in other words, IL\_CvsSWA\_C and IL\_DvsSWA\_D. Only genes that have at least ten reads in total were used for DE analysis. Differentially expressed genes (DEGs) were defined as those presenting an absolute fold change (FC) ≥ 2 or ≤ 0.5 and a False Discovery Rate (FDR) adjusted *p-*value ≤ 0.05 in any pairwise comparison. DEGs were then subjected to enrichment analysis of gene ontology (GO) terms, KEGG, and other metabolic pathways defined by a hypergeometric and Fisher's exact test using agriGO (version 2.0), KOBAS (version 3.0), and STRING database (version 10.5). The MapMan tool (http://MapMan.gabipd.org) was used to visualize the involvement of the DEGs in pathways of interest.
