Genes for Phylogenies: Identifying Genes with Intrinsically Good Properties for Phylogenetic Estimation

Iowa State University, Ames IA

Investigators

Abstract

The study of phylogenetic relationships among species of vertebrate animals relies heavily at present on analyses of mitochondrial gene sequences, and less so on analyses of nuclear genes. Dr. Gavin Naylor of Iowa State University is exploring several candidate nuclear genes, presumed to be single-copy or from small families (duplications) of genes, in order to find sources of reliable characters for phylogenetic analysis of vertebrates. From initial surveys of over 7000 genes in gene-sequence databases, ten or so candidates will be selected, amplification primers constructed to extract and amplify the segments, and the genes then sequenced for several model taxa of fishes, birds, and mammals. In comparing mutational differences between gene sequences from different organisms, such properties as nucleotide base compositional evenness, the proportion of sites free to vary, the among-site rate variation, and codon usage will be evaluated in tests using a known phylogenetic tree for the model taxa. Not all genes are equally likely to yield informative results in phylogenetic analyses, but there has been little explicit testing of properties of candidate genes, to determine beforehand useful gene sources. The work to be carried out will identify genes with properties that predispose them to be phylogenetically useful and will provide empirical tests of their utility. As sequencing technology improves and as more genomes are sequenced in their entirety, bioinformatics will turn to comparative genomics to mine information that is buried in the comparison of whle-genomes. This endeavor will be greatly enhanced with an accurate understanding of the phylogenetic relationships among the genomes being compared, and in turn robust, reliable phylogenies will depend upon data drawn from several nuclear as well as mitochondrial genes.