Content area
Конспект
The current trends in molecular phylogenetics are towards assembling large data matrices from many independent loci and employing realistic probabilistic models. Large genome-scale data sets shall reduce the sampling error, whereas complex models accommodating heterogeneity among sites and along the phylogenetic tree can decrease systematic errors. The theme of this dissertation project is using both bioinformatic and experimental approaches to develop genome-scale nuclear gene markers and applying them in studies of phylogeny of ray-finned fish (Actinopterygii) and systematics of clupeiforms. Bioinformatic tools and computer programs were developed to search for conserved single-copy nuclear genes with long exons. By comparing within and between genomes of zebrafish and pufferfish, I have found 138 candidate markers. Ten of fifteen candidates tested were found as good phylogenetic markers, showing similar performance as the popular nuclear marker, recombination activating gene 1 (RAG1). Using the ten newly developed nuclear markers, I conducted a phylogenetic analysis on 52 taxa representing 41 of 44 ray-finned fish orders along with four tetrapods as outgroups. The effects of different data partitioning methods were also tested. Some classic hypotheses about phylogenetic interrelationships of ray-finned fish based on morphological characters were rediscovered in this study, such as the “Holostei” group. In the last two chapters, I present the results of phylogenetic analyses of clupeiforms based on mitochondrial 12S and 16S ribosomal RNA genes, RAG1, RAG2 and six new nuclear loci. Clupeiforms include herrings, anchovies, etc. They have worldwide distribution and important commercial values. The most significant result of the study on clupeiforms is that Clupeidae is not monophyletic. Finally, the last chapter showed that adding sequences from the six new loci significantly improved the resolution and suggested a different relationship at the basal clupeiods.
