Genome evolution, speciation and molecular population genetics of giant weta

Abstract

Weta are among the largest extant insects in the world and are an iconic part of the New Zealand fauna. The decline observed in a number of species has resulted from a combination of habitat destruction and predation by introduced mammalian predators. A significant proportion of species are now classified as threatened, and these are the focus of a variety of translocations and breeding programs. The development of genomic resources, coupled with a better understanding of their biology and evolution, is essential to ensure the preservation and survival of these iconic New Zealand insects. The aim of this thesis was to employ various next-generation sequencing approaches to investigate aspects of weta biology and evolution. Three applications, transcriptome sequencing, genome sequencing and genotyping-by-sequencing, are highlighted with the resulting data being used to investigate the molecular evolution, determine underlying phylogenetic relationships, assemble and annotate a draft genome, develop a panel of genome-wide single-nucleotide polymorphisms (SNPs) and assess the levels of diversity and inbreeding in species of conservation interest. New Zealand tree (Hemideina) and giant (Deinacrida) weta are an ideal group to study selection patterns at the molecular level as they encompass a variety of habitats, including alpine adapted species which show elevated metabolic rates relative to lowland species. RNA-seq was used to generate transcriptomes for all 18 species. Orthologue identification using a bidirectional best hit approach identified 755 orthologues, which encompass a diverse range of Gene Ontology (GO) terms. A total of 83 orthologues were inferred as being under positive selection for at least one codon. Enzymes involved in oxidative phosphorylation, melanin synthesis and free-radical scavenging are represented within the 83 transcripts, providing a possible insight into the mechanisms by which metabolic rate variation and alpine adaption have evolved. Investigations into patterns of selection along branches leading to alpine lineages suggest that each alpine lineage has its own unique set of adaptations to the alpine environment. A total of 77 genes were inferred as being under positive selection only on alpine lineages and these represent interesting candidates underlying species specific alpine adaption within New Zealand weta. Sequencing a genome is an important first step to understanding many aspects of a species’ biology and evolutionary history. The Poor Knights giant weta (D. fallai) genome is large, with flow cytometry estimating the male haploid genome size at 8.5 Gbp. The draft assembly presented here represents, to the best of our knowledge, the first genome available for Anostostomatidae and the second for the insect order Orthoptera. The final assembly, comprises 5.38 Gbp over 711,650 scaffolds, with an N50 scaffold size of 516 kbp. Over half (58%) of the genome is comprised of repetitive elements, many of which are potentially unique to weta because they lack similarity to known repeat families. The difference between the flow cytometry estimate and assembly size is shown to result from a combination of repeat compression and low sequencing coverage. Although the resulting draft genome assembly is fragmented, it represents a valuable first step and will help aid our understanding and support further research into the genetic basis of weta biology and conservation efforts for these endemic species. Three Deinacrida species (D. fallai, D. heteracantha and D. mahoenui) were used to develop a genome-wide panel of SNPs from genotyping-by-sequencing. All three species are of conservation interest, and D. heteracantha and D. mahoenui are the focus of a variety of translocation and breeding programs. Although all three species have different population histories, all have undergone bottleneck events due to factors such as predation, habitat loss and/or translocation. An understanding of the underlying genetic variation and inbreeding levels is essential for the effective management of these species. This study reveals genome-wide SNP heterozygosity to be equivalent to the levels observed in wild outbred taxa, and that minimal inbreeding is occurring. A total of 32 putative outlier loci are inferred as being under diversifying selection, indicative of the potential for adaptive variation among the species. The application of these genomic resources to threatened species will help inform conservation management, ensuring the survival of these iconic insects.