Conservation genomic management of two critically endangered New Zealand birds.
Degree GrantorUniversity of Canterbury
Degree NameDoctor of Philosophy
In order to conserve global biodiversity, a multifaceted approach is needed to address complex conservation issues. One valuable tool in this approach is the use of genetic data to inform management (i.e., conservation genetics). For intensively managed threatened populations, genetic diversity can be managed through a conservation breeding approach, where relatively unrelated individuals are paired together to minimise inbreeding and maximise diversity in an effort to maximise evolutionary potential. For many, the pedigree has been the tool of choice for making pairing recommendations in conservation breeding programmes, as it relies on available ancestry data to estimate kinship —a measure of coancestry or pairwise relatedness— between individuals. However, threatened species recovery programmes often struggle to use this approach when pedigrees are shallow or incomplete. While genetic data (i.e., microsatellites) can measure relatedness for pairing recommendations, emerging evidence indicates this approach lacks precision in genetically depauperate species and more precise measures may be obtained from genomic data (i.e., thousands of single nucleotide polymorphisms, or SNPs). The field of conservation genetics is currently transitioning from using relatively few genetic markers to using thousands of genome-wide SNPs using high throughput sequencing (HTS) technologies. While the emerging field of conservation genomics promises greater precision for population diversity measures, relatively few studies to date have used these technologies, and exemplars are needed to demonstrate how to effectively and efficiently navigate from genetic to genomic technologies for use in conservation genetic management.
This thesis serves as one such exemplar, using two critically endangered birds as Proof-of-Concept: the kakī/black stilt (Himantopus novaezelandiae) and kākāriki karaka/orange-fronted parakeet (Cyanoramphus malherbi). Both species are endemic to Aotearoa New Zealand and part of their management includes conservation breeding programmes, where individuals are bred in captivity with their offspring translocated to predator-controlled wild habitats. Pairing recommendations for captive kakī and kākāriki karaka have been based loosely on visualised pedigree diagrams, but no studies to date have formally analysed either pedigree. In order to establish the capabilities and limitations of existing tools for genetic/genomic management, in Chapter 2 I developed multigenerational pedigrees for both species to investigate founder representation, relatedness, and mean kinship. This chapter highlights limitations of pedigrees for species with conservation breeding programmes that are routinely augmented by individuals of unknown ancestry, and underscores the value in incorporating empirical data (i.e., genetics and genomics) into management.
In the form of a Molecular Ecology opinion piece lead by me, Chapter 3 provides an overview of the gap between the availability of genomic tools and their use for conservation (i.e., the ‘conservation genomics gap’) and provides a pathway for people to transition and upskill in bioinformatic capacity. This piece describes how interdisciplinary relationships are enabling advances in both conservation genomics and primary industry research (e.g., agriculture, fisheries, forestry and horticulture), given the shared goals and applied nature of both disciplines. While conservation geneticists can learn about genomic approaches for aligned questions from primary industry, conservation geneticists can lend biodiversity expertise to primary industry for improved primary production output.
In Chapter 4, in an invited submission for the Genes “Conservation Genetics and Genomics” Special Issue, my co-authors (including co-first author Natalie Forsdick) and I explore the capacity for using readily available closely-related reference genomes for conservation management. In this chapter, we compare diversity estimates (i.e., nucleotide diversity, individual heterozygosity, and relatedness) derived from SNPs discovered using genotyping-by-sequencing and whole genome resequencing reads mapped to conordinal (killdeer, Charadrius vociferus), confamilial (pied avocet, Recurvirostra avosetta), congeneric (pied stilt, Himantopus himantopus) and conspecific reference genomes. Results indicate that diversity and individual heterozygosity estimates calculated from SNPs discovered using closely related reference genomes correlate significantly with estimates calculated from SNPs discovered using a conspecific genome, with congeneric and confamilial references provide higher correlations and more similar measures. While conspecific genomes may be necessary to address other questions in conservation, SNP discovery in birds using high-quality reference genomes of closely related species is a cost-effective approach for estimating diversity measures in threatened species.
In Chapter 5, in a manuscript submitted to Evolutionary Applications, my co- authors and I compare relatedness measures using pedigree, genetic, and genomic approaches for making pairing decisions in two critically endangered birds from Aotearoa with conservation breeding programmes: kakī and kākāriki karaka. This study uses family groups (i.e., parents, offspring, and siblings) to assess methods of estimating relatedness, as first order relationships between parents & offspring and siblings in these conservation breeding programmes are known. Our findings indicate genetic measures of relatedness are indeed the least precise when assessing known parent-offspring and sibling relationships, with SNPs providing more precision. Our results also show that pairing recommendations are most similar when using pedigrees and SNPs. Overall, these results indicate that in lieu of robust pedigrees, SNPs are the most effective measure of relatedness, which has exciting implications for poorly pedigreed populations worldwide.
Beyond using putatively independent SNPs for estimating relatedness, many researchers are looking to discover the genomic basis underlying maladaptive traits in small populations (e.g., inbreeding depression). While outside the scope of this thesis, Chapter 6 discusses new avenues for research given rich genomic and pedigree data sets now available for both kakī and kākāriki karaka. We anticipate that population genomic management simulations that balance selection for genome-wide diversity while penalising individuals for carrying maladaptive traits will allow researchers the ability to assess whether this approach enhances recovery in threatened populations. Overall, these combined chapters provide a toolbox for conservation geneticists who are transitioning to genomic technologies, especially for conservation breeding programmes. While this research project uses kakī and kākāriki karaka as focal species, it sits under the umbrella of a forward-thinking conservation genomics initiative that seeks to maximise the genetic diversity of a wide range of threatened species and enhance recovery efforts for species worldwide.