Figure 2

Phylogenetic tree of latrotoxin protein sequences. Previously published sequences labeled with NCBI accession numbers and newly assembled transcript sequences from L. hesperus with a predicted open reading frame of at least 500 amino acids from this study (in bold). The midpoint-rooted tree is a 50% majority-rule consensus of 3002 trees sampled in Bayesian analysis. Values at nodes show posterior probabilities ≥ 0.95, followed after the slash by ML bootstrap values when >= 70%. Shaded boxes indicate clades of known latrotoxin subtypes associated with specific phyletic targets with representative targets shown to the right; illustrations by Emily Damstra and used here with her permission. An asterisk symbol (*) after the name of the sequence indicates exclusive expression in the venom gland (zero eCPM in other tissues) otherwise the minimum fold difference in expression between the venom gland and the other two tissues is indicated. Underlined sequences vary in placement between the Bayesian and ML trees, as described in the text.