Whole genome duplications (WGDs) have frequently occurred throughout the tree of life and are thought to facilitate evolutionary success by providing raw genetic material and increasing the capacity for innovation and lineage diversification. This is no different in vertebrates; first formally suggested by Susumu Ohno in 1970, WGD has been attributed to much of the complexity and diverse characteristics of this lineage. The work described in this thesis aims to improve our understanding of the role played by WGD in vertebrate evolution and investigates the processes necessary for the associated novelties to arise. Whole genome duplications leave lasting traces in our genomes in the form of collinear blocks of paralogs. These long stretches of conserved gene order and content, termed 'microsynteny', are a distinctive feature of WGD and have been integral in reconstructing the history of ancestral duplication events. While gene order degrades quickly, gene content is often better conserved; recent work takes advantage of this to reconstruct ancestral pre-WGD and post-WGD chromosomes. However, these new methods are complicated and not well-documented. In Chapter 3, we develop an automated and user-friendly pipeline for reconstructing ancestral chromosomes before and after WGD, and use the conservation of gene content to infer chromosomal rearrangement events in this timeframe. We verify the efficacy of our tool by reconstructing the ancestral acipenseriform, a model system for vertebrate WGD and rediploidisation. Our pipeline should serve to make ancestral reconstruction more accessible and provide a solid foundation for future analysis. WGDs profoundly shaped early vertebrate evolution. The set of paralogs retained from these events – 'ohnologs' – have had lasting impacts on genome structure and function, including disease gene enrichment. Despite this, we still lack a gold standard ohnologs database. In Chapter 4, we have harnessed ancestral genome reconstructions together with phylogenetic and synteny information to produce a robust ohnologs dataset including details on the nature and quality of evidence, and a user-friendly interface at ohnologs.com. For the first time we have been able to resolve the 1R versus 2R origins of a large number of cases, as well as including previously hard-to-detect ohnologs. We find that 1R-retained ohnologs are involved in cellular transport, while those retained after both 1R and 2R are biased towards signalling functions, especially neuronal signalling. These findings suggest that after an initial adaptation to the cellular burden of polyploidy, WGD expanded the opportunities for intercellular communication. Rediploidisation is a necessary prerequisite to ohnolog sequence and functional divergence, and by extension, the evolutionary innovations attributed to WGDs. Despite this, we lack the understanding of this process after 1R; an event shared by all vertebrate genomes and associated with much of the early innovation in this lineage. In Chapter 5, we have harnessed high-quality ohnolog datasets together with careful phylogenetic analysis to provide evidence for delayed and independent rediploidisation in jawed and jawless vertebrates, both in the Hox clusters and in other genomic blocks. We address the systematic biases in previous work and verify that our observation is unlikely to be the result of phylogenetic error. This finding has implications for inferring the evolutionary pressures and conditions under which 1R occurred and under which crown vertebrates evolved, providing new insights into early vertebrate evolution. WGD has been central to vertebrate evolution. The work here builds a strong foundation to improve our understanding of this process and its enduring impact.

APPROVED