Phylogenetic Information Content of Five Nuclear Gene Sequences in Animals: Initial Assessment of Character Sets from Concordance and Divergence Studies

Abstract

The phylogenetic information content of five nuclear genes encoding amino acid sequences was assessed on the basis of their character support of well-established phy? logenies and their ability to recover these phylogenies using parsimony methods. The five genes encode dopa decarboxylase, phosphoenolpyruvate carboxykinase, RNA polymerase ll (largest subunit), elongation factor-2, and elongation factor-la. The analyses employ the unambiguously alignable portions of all available animal sequences for these genes, representing a range of divergence times from about 10 to >550 million years ago. Separate analyses, including character mappings on the test phylogenies and, when different, on the most-parsimonious trees, were performed with nucleotides separated by codon position and with amino acids. Pairwise sequence divergences were calculated for all comparisons and are related to character information content. Amino acid and most nucleotide character sets recovered the test phylogenies, although with varying degrees of support, homoplasy, and resolution, confirming that these genes contain substantial phylogenetic information. An estimate is provided of the time span over which each character set in each gene will prove phylogenetically most informative. These genes should prove widely useful for systematics. (Nuclear genes; animals; character sets; sequence diver? gences; concordance studies.)