Global lability, regional resolution, and majority-rule consensus bias

Abstract

Evolutionary interpretation of paleontological patterns requires a hypothesis of phylogeny, but our phylogenetic hypotheses may not perfectly mirror organismal phylogeny. Tree summary methods less conservative than strict consensus may increase resolution, but these methods may present a biased summary of the full set of most parsimonious trees. When we fail to acknowledge all equally optimal topologies, we risk disregarding trees that are closer to the correct phylogeny. We discuss a case where two subsets of trees were recovered in the set of most parsimonious trees, each with a profoundly different interpretation of character evolution near the root of Echinodermata. This was caused by the presence of a bimodally labile taxon in the matrix with two different topological subsets, each equally parsimonious but differing in the number of consistent trees. Majority-rule consensus favors the subset with the largest number of trees consistent with the placement of the rogue taxon. This bias favors clusters not because of the biological implications of the tree, but on the basis of great inequality in the sizes of the islands of parsimony. We thus recommend that majority-rule consensus trees not be used to summarize the results of a phylogenetic analysis.

Higher-order phylogeny of modern birds (Theropoda, Aves: Neornithes) based on comparative anatomy. II. Analysis and discussion

“…A strict consensus tree of the MPTs ( Figs 10 – 18 ) was completely resolved for the Neornithes with the exception of six polytomies (mostly trichotomies, some nested, discussed below), uncertainties sufficiently limited so as to obviate a majority-rule consensus tree for the primary solutions set, or to delimit ambiguity where one or more ‘rogue taxa’ may be influential ( Sumrall, Brochu & Merck, 2001 ). The strict consensus tree for the 97 MPTs shared the following summary statistics: (i) component information, 173; (ii) Nelson–Platnick term information, 4367; (iii) Nelson–Platnick total information, 4540; and (iv) Mickevich consensus information, 0.168.…”

A new southern Laramidian ankylosaurid, Akainacephalus johnsoni gen. et sp. nov., from the upper Campanian Kaiparowits Formation of southern Utah, USA

“…These authors were only able to make further interpretations of ankylosaurian phylogeny by calculating a 50% majority rule consensus tree and a maximum agreement subtree. The former method, although useful for statistical exploration of the relative frequency of branching relationships in a set of trees, should not be used to make phylogenetic or evolutionary interpretations, because it is a biased summary of parsimonious trees, and thus can lead to potentially false precision, as elegantly discussed by Sumrall, Brochu & Merck, (2001) . Maximum agreement subtrees do report phylogenetic structure common to all most parsimonious trees, but do so at the expense of many of the taxa in the original analysis.…”

Reassessment of the phylogenetic interrelationships of basal turtles (Testudinata)

“…The strict consensus is often deemed too strict, or insensitive, especially when taxa with a significant amount of missing data (e.g., fossils) are considered (e.g., Wilkinson 1994Wilkinson , 2003. Concerning the majority-rule consensus, it has been demonstrated that when ambiguity is rampant within a data matrix, especially because of missing data, biases in the method may drive the consensus tree toward the most ambiguous set of topologies (Sharkey & Leathers 2001;Sumrall et al 2001). Finally, the interpretation of the Adams consensus if often ambiguous because it does not represent common components/clades as strict and majority-rule consensus do, but common nestings (see Wilkinson 1994 andKitching et al 1998 for an introduction to the literature relating to consensus trees).…”