University of Sussex
Gossman,_Toni_Ingolf.pdf (1.72 MB)

Investigating genome wide patterns of natural selection in eukaryotes

Download (1.72 MB)
posted on 2023-06-08, 14:04 authored by Toni Ingolf Gossmann
Mutations are the ultimate source of new genetic information and they can be neutral, harmful or beneficial. The ultimate fate of all mutations is either to be lost or to eventually become fixed in a population. In this thesis I investigate genome wide traces of natural selection in eukaryotes. I focus on the most common type of mutations, point mutations, in protein coding genes. I investigated whether there is adaptive evolution in 11 plant species comparisons by applying an extension of the McDonald Kreitman (MK) test and found little evidence of adaptive evolution. However, most of the investigated plant species have low effective population sizes (Ne) and the rate of adaptive evolution is thought to be correlated to Ne. I therefore extended my study using additional data from mammals, drosophilids and yeast to investigate the relationship between the rate of adaptive evolution and Ne. I found a highly significant correlation between the rate of adaptive evolution relative to the rate of neutral evolution (!a) and Ne. It has been proposed that evidence of adaptive evolution can be an artifact of fluctuating selection. I simulated a model of fluctuating selection, in which the average strength of selection acting upon mutations is zero. Under this model adaptive evolution is inferred using MK-type tests. However, the mutations which become fixed are on average positively selected. The signal of adaptive evolution is therefore genuine. Ne can not only vary between species but also across genomes. However, how much variation there is, and whether this affects the efficiency of natural selection, is unknown. I analysed 10 species and show that variation in Ne is widespread. However, this variation is limited, amounting to a few fold variation in Ne between most genomic regions. This is never-the-less sufficient to cause variation in the efficiency of selection.


File Version

  • Published version



Department affiliated with

  • Biology and Environmental Science Theses

Qualification level

  • doctoral

Qualification name

  • phd


  • eng


University of Sussex

Full text available

  • Yes

Legacy Posted Date


Usage metrics

    University of Sussex (Theses)


    No categories selected


    Ref. manager