2.1 Evolutionary bioinformatics
The protein sequences used in this study were obtained from the National Center for Biotechnology Information (NCBI) (http://www.ncbi.nlm.nih.gov), Ensembl Plants (http://plants.ensembl.org/index.html), the OneKP database (http://www.onekp.com). Candidate protein sequences that satisfied the criteria of E-value<10-5 were selected by BLASTP searches (Cai et al., 2017; Chen et al., 2017; Zhao et al., 2019). The similarity among protein sequences and families) was calculated according to the BLASTP results using Arabidopsis as a reference species, and the heatmap was drawn using Excel (Microsoft, USA). The alignment of amino acid sequence was performed using MAFFT. For amino acid alignment, conserved domains of aligned sequences were identified using Gblocks v0.91b (http://phylogeny.lirmm.fr). The best-fitting model of amino acid substitution was tested using IQ-Tree v1.6.8 (http://www.iqtree.org), and LG+G4 was identified as best-fitting model under the Akaike Information Criterion. The maximum likelihood (ML) tree was constructed using FastTree v2.1 (http://www.microbesonline.org/fasttree) and was displayed using iTOL v3.0 (https://itol.embl.de).