Phylogenomic workflow with SIFTER. The workflow first runs an iterative Blast search against the RefSeq database (Pruitt et al. 2007) (only fully sequenced organisms) and filters the results to get putative orthologous and in-paralogous proteins. These sequences are used to build a multiple alignment with MAFFT (Katoh et al. 2005). After filtering out alignment columns with more than 60% gaps, a phylogenetic tree is build. If there are less than 20 proteins in the alignment Phyml (Guindon an...