Resolving tricky nodes in the tree of life through amino acid recoding

Genomic data allowed for a detailed resolution of the tree of life. Yet, tricky nodes such as the root of the animal, plants, eukaryotes, bacterial and archaeal trees remain unresolved. Genomic datasets are heterogeneous as genes and species evolve under different selective pressures, impending the efficacy of evolutionary analyses. Amino acid recodings were developed to reduce heterogeneity, but clear evidence to justify their use is missing. We use simulated genomic-scale datasets and show that recodings can substantially improve phylogenetic accuracy when tackling tricky nodes. We apply our findings to address the root of the animal tree where the debate centers on whether sponges (Porifera) or comb jellies (Ctenophora) branched out first. We show that results from real data follow predictions from simulated data and indicate that a placement of the ctenophores as the first branching animal lineage is most likely artifactual.