The construction of transcription factor networks through natural selection.
Doctoral thesis, UCL (University College London).
Transcription regulation plays a key role in determining cellular function, response to external stimuli and development. Regulatory proteins orchestrate gene expression through thousands of interactions resulting in large, complex networks. Understanding the principles on which these networks are constructed can provide insight into the way the expression patterns of different genes co-evolve. One method by which this question can be addressed is to focus on the evolution of the structure of transcription factor networks (TFNs). In order to do this, a model for their evolution through cis mutation, trans mutation, gene duplication and gene deletion is constructed. This model is used to determine the circumstances under which the asymmetrical in and out degree distributions observed in real networks are reproduced. In this way it is possible to draw conclusions about the contributions of these different evolutionary processes to the evolution of TFNs. Conclusions are also drawn on the way rates of evolution vary with the position of gene in the network. Following this, the contributions of cis mutations, which occur in the promoters of regulated genes, and trans mutations, which occur in the coding reign of transcription factors, to the evolution of TFNs are investigated. A space of neutral genotypes is constructed, and the evolution of TFNs through cis and trans mutations in this space is characterised. The results are then used to account for large scale rewiring observed in the yeast sex determination network. Finally the principles governing the evolution of autoregulatory motifs are investigated. It is shown that negative autoregulation, which functions as a noise reduction mechanism in haploid TFNs, is not evolvable in diploid TFNs. This is attributed to the effects of dominance in diploid TFNs. The fate of duplicates of autoregulating genes in haploid networks is also investigated. It is shown that such duplicates are especially prone to loss of function mutations. This is used to account for the lack of observed autoregulatory duplicates participating in network motifs. From this work, it is concluded that the relative rates of different evolutionary processes are responsible for shaping the global statistical properties of TFN structure. However, the more detailed TFN structure, such as network motif distribution, is strongly influenced by the population genetic details of the system being considered. In addition, extensive neutral evolution is shown to be possible in TFNs. However, the effects of neutral evolution on network structure are shown to depend strongly on the structure of the space on neutral genotypes in which the TFN is evolving.
|Title:||The construction of transcription factor networks through natural selection|
|Open access status:||An open access version is available from UCL Discovery|
|UCL classification:||UCL > School of BEAMS > Faculty of Maths and Physical Sciences > CoMPLEX - Maths and Physics in the Life Sciences and Experimental Biology|
Archive Staff Only