Wilkinson support calculated with exact probabilities: An example using Floricaula/LEAFY amino acid sequences that compares three hypotheses involving gene gain/loss in seed plants

This paper describes a method for quantifying the extent to which a character supports a hypothesized monophyletic group. The basic idea was first proposed by Wilkinson in 1998; hence, we call it Wilkinson support. A character provides Wilkinson support if it could have changed state on the branch leading to the hypothesized monophyletic group without requiring any extra steps in an evolutionary tree. We describe a method to determine the exact probability that a character would provide Wilkinson support for a random group of the same size as the hypothesized monophyletic group. A character's weight is defined as the negative natural log of this probability. The sum over all characters of these weights in a data set is a measure of total weighted support. We exemplify this method using 30 Floricaula/LEAFY amino acid sequences. One copy of this gene occurs in angiosperms, but two copies occur in the other four seed plant groups. Angiosperms could have been primitively single-copy or could have lost either of the two paralogs. These possibilities correspond to three hypotheses of monophyly. We use total weighted Wilkinson support to evaluate these three hypotheses, and all three are shown to be significantly different from random as individual hypothesized monophyletic groups. Comparing these three hypotheses for total weighted support reveals that one has much more support than do the other two. This hypothesis favors the "mostly-male" theory of flowering-plant origins.