Gene order comparisons for phylogenetic inference: evolution of the mitochondrial genome.

Abstract

Detailed knowledge of gene maps or even complete nucleotide sequences for small genomes leads to the feasibility of evolutionary inference based on the macrostructure of entire genomes, rather than on the traditional comparison of homologous versions of a single gene in different organisms. The mathematical modeling of evolution at the genomic level, however, and the associated inferential apparatus are qualitatively different from the usual sequence comparison theory developed to study evolution at the level of individual gene sequences. We describe the construction of a database of 16 mitochondrial gene orders from fungi and other eukaryotes by using complete or nearly complete genomic sequences; propose a measure of gene order rearrangement based on the minimal set of chromosomal inversions, transpositions, insertions, and deletions necessary to convert the order in one genome to that of the other; report on algorithm design and the development of the DERANGE software for the calculation of this measure; and present the results of analyzing the mitochondrial data with the aid of this tool.Evolutionary inference based on DNA sequences traditionally compares homologous versions of a single gene in different organisms. These comparisons are generally reliable indicators of phylogenetic relationships, even for very divergent organisms, but are limited in being based on point mutations only. In particular, homology between related mitochondrial genes may become difficult to distinguish from noise levels due to rapid nucleotide substitution (1), and this is not the only context in which the degree of sequence homology between genes having common origin is not a useful measure. Availability ofcomplete nucleotide sequence for organellar genomes suggests the possibility of inferring phylogenetic distances from their gene orders instead offrom sequences of individual genes (2). Analyses ofevolution at the genome level necessarily differ from sequence comparisons of individual genes. Though the processes of insertion and deletion of sequence elements have direct counterparts at the genomic level, the predominant process, nucleotide substitution, does not, whereas other processes assume major importance, such as the transposition of a segment from one region of a chromosome to another or the inversion of a chromosomal segment. Here we propose a quantitative analysis of transposition, inversion, and insertion/deletion, leading to the reconstruction of a mitochondrial phylogeny.Though the inference of evolutionary history through genomic rearrangements is well-established (3-5), it has been the goal of our work to define a general edit distance that combines a variety of order-disrupting events, to devise and implement a combinatorial algorithm capable of estimating this distance, and to apply these tools in a uniform way across a wide spectrum of eukaryotic organisms to generate input suitable for phylogenetic tree construction methods. Our results generally agree with evolutionary relationships inferred from gene...

Analysis of Gene Order Evolution Beyond Single-Copy Genes

“…In comparative genomics, gene orders were originally modeled as unsigned [120] or signed [99] permutations, in order to analyze data on organellar or prokaryote genomes. This provided an alternative to the classical method of comparing the DNA sequence of single genes.…”

Genome-Wide EST Data Mining Approaches to Resolving Incongruence of Molecular Phylogenies

“…Kluge believed that phylogenetic analysis should always be performed using all the evidences [14], but Miyamoto and Fitch argued that partitions (including genes) should not be combined [15]. Concatenated alignments of a couple of genes improved supports [3][4][5][6][7][8][9][10][11][12][13][14][15][16]. Concatenating alignments into one from genome-scale 106 genes based on 7 yeast genomes [17] for phylogenetic analysis proclaimed ending incongruence [18].…”

Edit distance for genome comparison based on non-local operations

“…Nevertheless, since these processes are widespread and accumulate over time, it is interesting to ask whether the comparison of genomes, i.e. of their gene orders, can provide us a statistical basis for assessing the similarity or distance or evolutionary divergence between organisms, in analogy with nucleotide sequence comparison [1,4,5,6,8]. In this paper, we motivate and describe an algorithm capable of inferring the minimal set of inversions and transpositions necessary to convert one genome order into another.…”

“…We have sampled 11 mitochondrial genomes from the database of [8]. As listed in Table 1, this contains four multicellular organisms, extending from nematodes to humans, and seven fungi, of which three are budding yeasts, one is a fission yeast, and three are filamentous fungi.…”