Comprehensive Plant Protein Kinase Analysis

The PKs were classified into groups and subfamilies according to the HMMs of each family built from four plant model species (Arabidopsis thaliana, Chlamydomonas reinhardtii, Oryza sativa, and Physcomitrella patens) and supported among 21 other plant species (Lehti-Shiu and Shiu, 2012 (link)). The classification was further validated through phylogenetic analyses. The domain sequences from all PKs were aligned using Muscle v.8.31 (Edgar, 2004 (link)), and a phylogenetic tree was constructed for each kinase dataset using the maximum likelihood approach in FastTree v.2.1.10 software (Price et al., 2010 (link)) with 1,000 bootstraps and default parameters through the CIPRES gateway (Miller et al., 2011 ). The resulting dendrograms were visualized and plotted using the statistical software R (Ihaka and Gentleman, 1996 ) together with the ggtree (Yu et al., 2017 (link)) and ggplot2 (Wickham, 2009 (link)) packages.
For each PK, we obtained the following characteristics: (a) gene location and intron number, according to the GFF annotation files; (b) molecular weight and isoelectric point with ExPASy (Gasteiger et al., 2003 (link)); (c) subcellular localization prediction using CELLO v.2.5 (Yu et al., 2006 (link)) and LOCALIZER v.1.0.4 (Sperschneider et al., 2017 (link)) software; (d) the presence of transmembrane domains using TMHMM Server v.2.0 (Krogh et al., 2001 (link)); (e) the presence of N-terminal signal peptides with SignalP Server v.5.0 (Armenteros et al., 2019 (link)); and (f) gene ontology (GO) term IDs using Blast2GO software (Conesa and Götz, 2008 (link)) with the SwissProt Viridiplantae protein dataset (Consortium, 2019 (link)).