Transcriptomic analysis
Transcriptomic analysis was performed comparing the tillers that flowered in the next season (17Hot) and tillers that remained vegetative (17Control) as described in Samarth et al. (2019). 150 bp generated paired-end reads were assembled into a single reference transcriptome using the Trinity pipeline (Haas et al., 2013). Reads from each of the samples were aligned to the reference assembly to generate a count matrix using Bowtie2.0 (Langmead & Salzberg, 2012) followed by the differential expression analysis using the DESeq2 package in R. To accurately identify the significantly differentially expressed genes (adj. p-value<0.05), stringent parameters including an E-value of 10-5, 70 % query coverage and 50% protein identity were used for annotation based on orthologous proteins identified in B. distachyon using blastp. Gene ontology and pathway enrichment analysis were performed based on the PANTHER server (Mi et al., 2016). Validation of the RNA-seq results was performed using RT-qPCR with a random selection of four genes from the transcriptomic data (Appendix S2).