Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.560099
Title: Modelling transcriptional networks in plant senescence
Author: McHattie, Stuart
Awarding Body: University of Warwick
Current Institution: University of Warwick
Date of Award: 2011
Availability of Full Text:
Access through EThOS:
Access through Institution:
Abstract:
Senescence is a highly regulated developmental process in plants in which nutrients are remobilised from organs which are no longer required or are stressed so that they may be used by organs which are only just developing. Whilst much is known about the causes of and the resulting outcomes of this process, very little is known about the genetic machinery which link them. Some genes have been identified as having a regulatory role in the senescence process, but many of these have been determined by a forward genetics approach whereby mutants are randomly screened for phenotypical e↵ects. A much better approach, where possible, is the reverse genetics approach whereby mutants are sought for testing as they are suspected to demonstrate a phenotypical e↵ect. It was the purpose of this study to find novel ways of identifying those genes which may be highly regulatory of the senescence process and to determine how they are able to lead from several di↵erent known causes of senescence through to the senescence response itself. It was hypothesised that, using measurements of the expression levels of a very large number of genes throughout the senescence process, theoretical models of regulation between those genes could be determined and that these theoretical models would allow specific interactions to be identified and explained using biological validation techniques. A large microarray experiment, performed prior to the start of this project, measured the expression of over 30,000 genes in the Arabidopsis thaliana genome over a period of 21 days during natural senescence. By cleaning this data and fitting an ANOVA driven model to the resulting intensity measurements, it has been possible to separate e↵ects leading to observed expression changes. The levels of each of these e↵ects were tested by F-tests and this has allowed the identification of 8,878 genes which are significantly di↵erentially expressed during senescence. By first using theoretical models to find genes amongst the set of 8,878 which demonstrate highly robust regulatory behaviour on other genes in the set, 118 genes were able to be isolated for further study. A senescence phenotype screen was developed to assess reduced-expression mutants of many of those 118 genes and 8 were shown to have a significantly altered timing of senescence when compared with wild-type plants. The surrounding networks of each of those 8 genes were formed by applying theoretical regulatory network modelling in another novel manner similar to a Metropolis-Hastings approach which identified a set of 75 genes providing a testable regulatory network model. The resulting network model has been tested biologically to establish the accuracy of the predictions. Whilst many of the predictions were not confirmed, a vast network has been identified surrounding two of the highly regulatory genes indicating a junction of two separate pathways leading to the senescence response and providing a network structure which could be used in another round of theory and validation. Additionally, these results introduce new interesting questions about how the senescence network may have evolved to respond to so many inputs.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.560099  DOI: Not available
Keywords: QK Botany
Share: