orming
one-way analysis of variation (ANOVA) for control
and drought samples, respectively. After multiple test correction
using the fdrtool method, genes with significantly
different expressions between developmental stages were
analyzed further to determine where the significances
arise, using the Tukey’s Honestly Significant Difference
method (the TukeyHSD function in the R package ‘stats’).
Genes with significant expression level changes between
consecutive floret stages were considered as developmentally
regulated.