%0 Journal Article
%T Effects of Taxon Sampling in Reconstructions of Intron Evolution
%A Mikhail A. Nikitin
%A Vladimir V. Aleoshin
%J International Journal of Genomics
%D 2013
%I Hindawi Publishing Corporation
%R 10.1155/2013/671316
%X Introns comprise a considerable portion of eukaryotic genomes; however, their evolution is understudied. Numerous works of the last years largely disagree on many aspects of intron evolution. Interpretation of these differences is hindered because different algorithms and taxon sampling strategies were used. Here, we present the first attempt of a systematic evaluation of the effects of taxon sampling on popular intron evolution estimation algorithms. Using the ¡°taxon jackknife¡± method, we compared the effect of taxon sampling on the behavior of intron evolution inferring algorithms. We show that taxon sampling can dramatically affect the inferences and identify conditions where algorithms are prone to systematic errors. Presence or absence of some key species is often more important than the taxon sampling size alone. Criteria of representativeness of the taxonomic sampling for reliable reconstructions are outlined. Presence of the deep-branching species with relatively high intron density is more important than sheer number of species. According to these criteria, currently available genomic databases are representative enough to provide reliable inferences of the intron evolution in animals, land plants, and fungi, but they underrepresent many groups of unicellular eukaryotes, including the well-studied Alveolata. 1. Introduction Introns are noncoding sequences inside many eukaryotic genes. Their abundance may vary several orders of magnitude, from hundreds of thousands in mammalian genomes to less than 100 in the genome of Saccharomyces cerevisiae. The origin and evolution of introns is a highly controversial topic despite 25 years of research. The variation of intron content between different lineages suggests a high variation in the rate of intron gain and loss, which may relate to differences in population size, absence or presence of the sexual process, activity of transposable elements, properties of the splicing mechanism, and many other characteristics of genomes, organisms, and populations. Intron sequences evolve at a high rate, and the negative selection typically stabilizes only few nucleotide positions in splicing sites and the branch point. Interestingly, a significant portion of introns occupy the same positions in the same genes in species that diverged billions years ago (20% of common introns in mammals and Arabidopsis thaliana for the set of 684 conservative genes [1]). These introns were interpreted either as ancestral, originating before the divergence of major eukaryotic lineages, or as convergently inserted in the same
%U http://www.hindawi.com/journals/ijg/2013/671316/