Abstract
Genetic diversity is the raw material for evolutionary adaptation, providing the variation necessary for populations to respond to changing environments and anthropogenic pressures. Examining patterns of genetic diversity allows us to understand how populations have responded to historical and contemporary environmental challenges, and how demography has shaped current population structure and genetic variation. Advances in omics technologies now enable the assessment of genetic diversity at broader scales, including genomics, epigenomics, and microbiome composition, and the interactions among these layers, offering insights into the genomic basis of adaptive variation and population responses to emerging threats such as introduced pathogens, human-induced bottlenecks, and climate change.
Hector’s dolphins (Cephalorhynchus hectori hectori) and Māui dolphins (C. h. maui), two subspecies endemic to the coastal waters of Aotearoa New Zealand, are among the most vulnerable marine mammals globally. Threats include their small population sizes, inshore distribution, low reproductive rates, and exposure to anthropogenic impacts, including fisheries bycatch, pollution, vessel traffic, and emerging pathogens. Understanding their evolutionary history, genomic health, and adaptive potential in the face of such threats is therefore critical for effective conservation management. To achieve this, I developed an integrative conservation genomics framework to investigate evolutionary divergence, population genomic health, epigenomic differentiation, and disease vulnerability of Hector’s and Māui dolphins. I first generated high-quality chromosome-level reference genomes for both subspecies using Oxford nanopore long-read and 10× Genomics linked-read sequencing (Chapter 2). By leveraging the high chromosomal synteny characteristic of cetaceans, I developed a reference-guided scaffolding strategy that produced assemblies comparable to the best available cetacean genomes. These genomic resources, published in Molecular Ecology Resources, provide a robust foundation for downstream analyses, minimising the potential for reference bias.
Then, utilising low-coverage whole-genome resequencing of 48 individuals sampled across the species range, I reconstructed demographic history and assessed genomic diversity (Chapter 3). Results support a climate-driven divergence during the Last Glacial Maximum approximately 15,000 years ago. The South Coast Hector’s dolphin population exhibits elevated inbreeding and genetic load comparable to that observed in the critically endangered Māui dolphin, raising conservation concern for this understudied population. Additionally, structural variants, including chromosomal inversions, were identified as potential contributors to reproductive isolation despite occasional overlap between the subspecies.
Building upon these genomic analyses, I examined epigenomic differentiation between the subspecies using an existing methylation array dataset (Chapter 4). Despite technical constraints inherent to cross-species array hybridisation, significantly differentially methylated regions were detected between subspecies and environmental contexts. Genes associated with thermoregulation, reproduction, immunity, and development showed subspecies-specific methylation patterns, while environmentally associated signals were enriched in immune and metabolic pathways. These findings demonstrate that epigenetic clock datasets can reveal adaptive and evolutionary signals in natural populations and highlight both biological and technical drivers of methylation variation.
Finally, I integrated host whole-genome data with skin microbiome profiles from 30 individuals to evaluate predictors of infectious disease susceptibility using explainable machine learning models (Chapter 5). Māui dolphins exhibited reduced microbial alpha diversity relative to Hector’s populations, and individuals that died from infectious disease showed signatures of dysbiosis. Increased inbreeding, reduced genome-wide heterozygosity, altered microbiome diversity, and female sex were jointly associated with elevated pathogen mortality risk. This multi-omics framework identifies the skin microbiome–host genetics relationship as a dynamic, minimally-invasive biomarker for monitoring health in endangered cetacean populations.
Collectively, this thesis provides high-resolution genomic and multi-omics insights into the evolutionary history, genetic diversity, adaptive divergence, and health vulnerability of Hector’s and Māui dolphins. Beyond advancing knowledge of these endangered subspecies, it demonstrates how integrative genomic approaches can inform conservation strategies for small, isolated populations facing accelerating environmental change.