Cooke, Emma J. (2013) Transcriptional analysis of the interaction between botrytis cinerea and a host arabidopsis thaliana using high-throughput data. PhD thesis, University of Warwick.

Research output not available from this repository.

Request-a-Copy directly from author or use local Library Get it For Me service.

Request Changes to record.

Abstract

Botrytis cinerea is an economically important necrotrophic pathogen which causes
disease in hundreds of species of plants during pre- and post-harvest conditions. This
thesis investigates the transcriptional responses of B. cinerea and the host Arabidopsis
thaliana during the infection through the analysis of microarray and RNA-seq data
sets. This work develops techniques for clustering time series expression profiles and
identifying direct gene targets using microarray data; and techniques for identifying
differentially expressed and differentially spliced genes using RNA-seq data.
A clustering algorithm which uses Gaussian process regression to capture the time
series structure of microarray data was developed and analysed. Features which are
not considered by standard clustering algorithms were added, specifically the ability
to include replicate data by using replicate information to inform a prior distribution
for noise, and the ability to consider outlier values by using a mixture model likelihood.
This algorithm is shown to produce more coherent and biologically meaningful
clusters than standard algorithms when applied to publicly available time series data.
This algorithm was also used to cluster A. thaliana transcription factors with similar
expression profiles during B. cinerea infection.
The transcription factors CAMTA3 and MYB108 are known to play a role in the A.
thaliana defence response to B. cinerea. Mutant A. thaliana plants were generated
which constitutively express the CAMTA3 gene, and these are shown to be more
susceptible to B. cinerea infection than wild-type plants. Microarray data sets from
mutant CAMTA3 and MYB108 A. thaliana plants were generated and used together
with a time series data set of A. thaliana infected with B. cinerea to identify the most
likely direct targets for these two transcription factors. Possible regulatory motifs to
which these transcription factors bind were also identified.
RNA-seq data sets of A. thaliana infected and mock-infected with B. cinerea at
three key infection stages were generated. 2,081 novel splice junctions were identi
fied for A. thaliana from the data. Differentially expressed genes for A. thaliana
and B. cinerea were identified between the key infection stages using existing methods,
however these methods are limited to pairwise testing. An improved method
using generalised linear models was developed to enable the incorporation of both
time and infection stage factors, which identified 12,940 A. thaliana genes differentially
expressed due to B. cinerea infection. Different isoforms of A. thaliana genes
were identified at a transcript level, at an event level and at a splice junction level.
Generalised linear models were then used with the multinomial distribution, which
considered both time and infection stage factors, to identify 928 A. thaliana genes
which are likely to be differentially spliced due to B. cinerea infection.

Item Type:	Thesis (PhD)
Subjects:	Q Science > QK Botany S Agriculture > SB Plant culture
Library of Congress Subject Headings (LCSH):	Botrytis cinerea, Arabidopsis thaliana, Transcription factors
Official Date:	January 2013
Dates:	Date Event January 2013 Submitted
Institution:	University of Warwick
Theses Department:	Department of Chemistry
Thesis Type:	PhD
Publication Status:	Unpublished
Supervisor(s)/Advisor:	Denby, Katherine J.; Wild, David L.
Extent:	xvi, 240 leaves : illustrations, charts.
Language:	eng

Request changes or add full text files to a record

Repository staff actions (login required)

View Item