Figure 2.

Phylogenetic tree of SIT proteins, including full‐length SIT sequences determined from available diatom genomes, previously published sequences available in GenBank, and the MMETSP data set of marine microbial eukaryotes. Sequence fragments were included in the tree if they encoded at least seven predicted TM domains. Sequences with more than 10 TM domains were segmented and aligned in 10 TM segments, the names of these genera are labeled along with proposed domain duplication events I–VI. The tree is rooted by a clade containing SIT proteins encoded only by non‐diatom organisms, but non‐diatom SIT proteins are also found on other branches of the tree. Branches within a clade with >50% bootstrap support are collapsed for simplicity. Node scores are bootstrap values from 100 bootstrapped trees.