Standardized Phylogenetic Tree

Abstract

Table 1: Number of genes in Vertebrate taxonomic groups. (Genes whose sequences are determined for more than five species are counted.) Taxon No. of species Eutheria 289 Metatheria 42 Monotremata 15 Aves 107 Crocodylidae 9 Lepidosauria 98 Pleurodira 20 Cryptodira 5 Amphibia 50 Actinopterygii 107 Osteichthyes 0 Coelacanthiformes 1 Dipnoi 9 Chondrichthyes 22 Cephalaspidomorphi 7 Myxini 4 The evolutionary rate of a gene is constant while it is being neutral to selection. The molecular evolutionary models under such condition had been well studied and many genes are thought to and observed to behave in such way for most of the evolutionary time. There should, however, be time periods that a mutation of a gene is advantageous since it acquired or improved its function so that the individual who owned this mutant gene acquired better fitness to the environment among the population. Such cases are regarded to be positive Darwinian selection in which that mutant gene is selectively advantageous. From the prediction of the Neutral Theory of Molecular Evolution, such a gene that is operated by positive Darwinian selection is expected to have higher evolutionary rates, which can be observed as longer branch length when phylogenetic trees are drawn for the gene in question than the expectation from the neutral evolution because of the shorter fixation time. Therefore, those genes that are under positive natural selection is in principle can be figured out by testing whether branch length is significantly long compared to the standard branch length, where standard branch length reflects the branch length of the gene whose evolutionarily contributing mutants are neutral to selection. Although the principle is that simple, it is practically uneasy to test whether a gene is under positive natural selection mainly for two reasons: the absence of standard phylogenetic tree or sets of branch lengths, and the variations of evolutionary rates among genes. In other words, positive natural selection is testable given a standard reference tree and a method to convert any tree into what can be compared to that reference. Even though defining a standard is another problem, genome projects and other research projects had accumulated a huge number of sequence data, and now it is a time to compile data toward construction of standard molecular phylogenetic tree [1, 2, 3]. Here we present an attempt to develop such method as well as reference trees by using currently available data for levels of taxa and developing a method to compare trees of different genes. Those reference trees we designate standardized phylogenetic trees.