Unravelling the Origins and Evolution of the Animal Kingdom using Genomics

There are ~35 classified phyla/sub-phyla within the Animal Kingdom; some of which haveunresolved relationships. The advent of genomics has made it possible to study new aspects of animalevolution, including comparative genomics (e.g., gene loss/gain, non-coding regions, synteny, etc),gene family evolution, and their evolutionary relationships using genome-wide data. No study to date has compared all the wealth of genomic data available to understand theevolution of the Animal Kingdom. Using a core bioinformatics pipeline and dataset to infer HomologyGroups (HGs), the losses and novelties of these HGs were proven integral to the diversification of theanimal kingdom. The same core pipeline was used to extract homeobox gene HGs, a key family usedto understand origin and diversification in animals. Gene trees were inferred from the core datasetHGs to determine the evolution of a gene family iconic in the study of animal body plans. Conservedanimal genes were also mined using the same pipeline and dataset. Animal phylogenomics is one ofthe most controversial areas in modern evolutionary science. Whilst many new methods have beendeveloped, no study to date has tried to assess the impact of gene age in the reconstruction ofevolutionary trees. The phyla with the largest count of HG losses also had the highest counts of HG novelties. Notall of these were strictly de novo, but the numbers suggest a re-manufacturing of the genetic materialfrom the genes reduced to those that were more recently diverged. A comprehensive classification of all the diversity of animal homeobox genes is lacking. Thegene trees showed complex patterns, with similar homeobox expansions between more distant species,and interlapping homeobox families. The highly conserved HGs recovered, for the animal phylogenies, well-established relationshipsbetween some phyla using maximum likelihood and Bayesian inference methods. Ctenophora wasconsistently recovered as sister to all other animals, and interesting relationships between ecdysozoansand lophotrochozoans. However, it was proven that it takes more than a highly conserved set of genesto infer a stable and correct phylogeny. Each of the additional methods used to extend the core bioinformatics pipeline revealed apattern of correlation, particularly among the fast evolving species, such as platyhelminthes, nematodes and tardigrades. These HG losers, and gainers also had lineage specific homeobox clades,and caused artefactual problems in the phylogenies.