Identification of NBS-Encoding Genes

The NBS genes in the 13 genomes were identified as described in previous studies (Shao et al., 2014; Shao et al., 2015; Zhang et al., 2016). Briefly, a HMM search was first performed for protein sequences in each genome using hmmer3.0 (http://hmmer.org) using default parameter settings, with the NB-ARC domain (Pfam: PF00931) as query. The amino acid sequence of the NB-ARC domain was then used to run a BLASTp search against all protein sequences in each genome, with the threshold expectation value set to 1.0. All hits obtained using HMM or BLAST searches were then merged together, and the redundant hits were removed. The remaining sequences were further subjected to Pfam analysis (http://pfam.sanger.ac.uk/) to further exclude sequences that do not have a detectable NBS domain at an E value of 10⁻⁴. When two or more transcripts were annotated for a gene from alternative splicing, the longest form with an NBS domain was selected. Finally, the identified NBS domain-encoding genes were examined to see whether these encode the TIR, RPW8, CC, or LRR domains using Pfam and COILS analyses (Lupas et al., 1991). The exon position and intron phase of each gene were transformed from the gff3 file of the reference genome.