Novel approaches for large-scale phylogenetics and applications in the context of the amphibian tree of life

During this thesis, I addressed some problems associated with large-scale phylogenetic analyses by tackling issues related to missing data and careful handling and addition of novel data in large-scale reconstructions, presenting an application of this approach in the context of amphibian phylogenetics. I developed a method (called “Concatabominations”) building on the original Safe Taxonomic Reduction method (Wilkinson 1995) as an alternative approach to the issue of identifying rogue taxa. The safe removal of rogue taxa due to missing data can potentially reduce the terraces in tree space search and improve resolution in the final consensus tree. In a pragmatic point of view, the new method can help in targeting taxa that require further sampling during a research design. Novel sequence data for the rediscovered Ericabatrachus baleensis allowed to explore its placement in the Amphibian tree of life. I tested the inclusion of novel data using a backbone alignment from a previous work (de novo analysis) and a backbone phylogenetic tree (constrained analysis), after careful curation of gene partitions to include in an analysis. I found that the use of a constrained phylogenetic inference using a previous accepted tree seems to be a practical solution to the rapid phylogenetic placement of a taxon in cases of well-supported relationships. However, a de novo analysis might ensure an optimal alignment and avoid risks introduced when adding new data. Finally, I investigated the evolutionary relationships of the three lineages of the extant amphibians (Anura, Caudata and Gymnophiona) using an independent source of evidence: miRNAs, recently used to help resolve difficult phylogenetic problems. The analyses yielded a high number of shared miRNAs using the Xenopus tropicalis genome, contrasting with a lower number of miRNAs discovered using the Axolotl transcriptome. This suggests that not using genomic data is not ideal to validate miRNAs. Nevertheless, in spite of the limitations, I was able to find two potential novel miRNAs: one supporting the monophyly of Lissamphibia, and another supporting the Batrachia hypothesis. Overall, I hope the work developed in this thesis contributes with new insights into large-scale phylogenetics and in particular to amphibian phylogenetics.

Name dc:type.qualificationname

phd

Level dc:type.qualificationlevel

doctoral

Grantor dc:publisher.institution

National University of Ireland Maynooth

Year dc:date.issued

2014

Author dc:creator

• Siu Ting Salvatierra, Karen

• Biology

Language dc:language

en