A pan-genomics approach linking genotype and phenotype in the mega-pest, Helicoverpa armigera

Abstract

The Cotton Bollworm (Helicoverpa armigera) is a globally distributed polyphagous pest with a profound economic and environmental impact. Human intervention has left a distinct imprint on the genetic and phenotypic landscape of this organism from decades of attempts to control it, initially with synthetic pesticides and more recently with Bacillus thuringiensis (Bt) derived bio-toxins. Understanding the genetic mechanisms that drive pesticide resistance is crucial for effective monitoring of resistance but remains challenging. Recent advances in sequencing technologies, specifically long-read sequencing, have dramatically improved the resolution of the genetic information able to be gained from individuals. This information enables the application of pan-genomic approaches (which are focused on assessing the large-scale variation within a group) to organisms with large or complex genomes.

In this thesis I describe my work on a pan-genomic approach to catalogue significant genetic variation within Cotton Bollworm, with a particular emphasis on its two primary sub-species, H. armigera armigera and H. armigera conferta. Utilizing both short-read and long-read sequencing methods, distinct patterns emerge in the variation of genes associated with pesticide resistance, such as cytochrome P450s and serine endo-peptidases, between these two sub-species. These findings bear implications for understanding the variable susceptibilities of these sub-species in response to pesticides.

Furthermore, I show evidence from a linkage experiment for the novel genetic mechanism of resistance of Cotton Bollworm to a specific class of Bt bio-toxin (Vip3Aa). The putative causal mutation (a transposable element insertion) is missed by traditional short-read approaches but can be identified using long-read sequencing. This highlights the improvement of resolution of genetic variation of long-read sequencing in comparison to short-read technologies.

Finally, I assess short-read and long-read based molecular approaches for identification of Bt resistance alleles from pooled populations and an individual, respectively. The use of molecular monitoring approaches has long been hoped to improve identification and reduce requirements for labour intensive phenotype testing. I find that due to the large structural variation that are often causal of Bt resistance, long-read technologies have superior performance at identification of resistance alleles, and with the use of targeted sequencing have the potential to provide rapid results for the assessed known Bt resistance genotypes.

In conclusion, I demonstrate the utility of pan-genome approaches, particularly the insights from long-reads, in improving the ability to understand genetic mechanisms which drive phenotypes of interest. The insight from this thesis inform future strategies for managing pesticide resistance in this, and other, insect pests, contributing to more sustainable pest management practices. Show more