I would like to conduct a statistical test to check if some character of interest is significantly different between two groups of species (let's call them groups A and B). Since A and B belong to different kingdoms, these two groups have a very ancient common ancestor. I have the phylogenies for groups A and B and would like to account for species relatedness when performing the test, for example using phylogenetic ANOVA. However, at least the R phytools implementation expects a single tree containing all species. I could generate a tree which connects these two clades, but such a tree will have two very long branches leading to each clade, while the rest would be very short. Is this the way to go, or is there a more appropriate way to do this? Maybe a dedicated test designed for situations like this?
