Modelling the formation of heterogeneous genomic landscapes

Thesis

The progression along the speciation continuum is driven by multiple evolutionary processes: selection, mutation, gene flow and drift. How these evolutionary forces, along with the genetic architecture e.g. recombination, affect the genomic landscape is a key question in evolutionary biology. The genomic landscape of divergence and introgression describe the heterogenous distribution of divergence and introgression along genomes, respectively. However, discerning which evolutionary processes shaped the evolution of a particular biological system is challenging. This is due to the fact that i) different strengths of the same evolutionary process can generate different progressions along the speciation continuum, and ii) through interaction between evolutionary processes, different evolutionary processes can produce that same pattern in the genomic landscape. Nevertheless, examining patterns in the genomic landscape can provide insights into the evolution of a particular biological system. In this thesis, I further develop an individual-based model to simulate the genomic landscape of divergence and introgression. In Chapter 2 and 3, I show the impact of additional layers of complexity on the genomic landscape of divergence. In Chapter 2, I explore the effect of i) different genotype-phenotype maps, by simulating genetic dominance and genotype-environment interaction; ii) frequency-dependent selection; and iii) varying the pattern of migration between populations. In Chapter 3, I model negative epistatic interactions in the form of Dobzhansky-Muller incompatibilities, and I introduce the simulation of multiple traits under selection. Through these developments, I emphasize that the genomic landscape of divergence is highly influenced by additional layers of complexity, which must be considered when interpreting genomic patterns that emerge between diverging lineages. Additionally, I highlight that under the same evolutionary parameters, different genomic landscapes of divergence can form. In Chapter 4, I develop the individual-based model to examine the genomic landscape of introgression. Here I demonstrate the genomic landscape of introgression is highly variable, and even under the same evolutionary conditions very different genomic landscapes of introgression can form. Additionally, I highlight that concurrent interactions between evolutionary processes and the genomic architecture can change the effect an evolutionary process has on the genomic landscape. Our model provides a set of tools for better understanding the effects of different evolutionary processes concurrently and independently. Here, we show that the same evolutionary parameters can form different patterns on both the genomic landscape of divergence and introgression. Thus, caution must be taken in interpreting the genomic landscapes, as specific patterns in the genomic landscape are not necessarily indicative of a particular set of evolutionary conditions.

Authors: Ali, HAA

• DOI: 10.5287/ora-6rnvzjqb1
• Type of award: DPhil
• Level of award: Doctoral
• Awarding institution: University of Oxford
• Language: English
• Keywords: introgression; genomic landscape of divergence; divergence; dobzhansky-muller incompatibilities; individual-based model (ibm)
• Subjects: evolutionary ecology; computational biology