Use this URL to cite or link to this record in EThOS:
Title: Integrative analysis of ChIP-chip datasets in Saccharomyces cerevisiae
Author: Bennett, Mark
ISNI:       0000 0004 2733 7826
Awarding Body: Cardiff University
Current Institution: Cardiff University
Date of Award: 2012
Availability of Full Text:
Access from EThOS:
Access from Institution:
ChIP-chip is a technology originally developed to determine the binding sites of proteins in chromatin on a genome wide scale. Its uses have since been expanded to analyse other genome features, such as epigenetic modifications and, in our laboratory, DNA damage. Datasets comprise many thousands of data points and therefore require bioinformatic tools for their analysis. Currently available tools are limited in their applications and lack the ability to normalise data so as to allow relative comparisons between different datasets. This has limited the analyses of multiple ChIP-chip datasets from different experimental conditions. The first part of the study presented here is bioinformatic, presenting a selection of tools written in R for ChIP-chip data analysis, including a novel normalisation procedure which allows datasets from different conditions to be analysed together, permitting comparisons of values between different experiments and opening up a new dimension of analysis of these datasets. A novel enrichment detection procedure is presented, suited to many formats of data, including protein binding (which forms peaks) and epigenetic modifications (which can form extended regions of enrichment). Graphical tools are also presented, to facilitate the analysis of these large datasets. A method of predicting the output of a ChIP-chip dataset is presented, which has been used to show that ChIP-chip is capable of detecting sequence dependent damage events. All functions work together, using a common data format, and are effcient and easy to use. The second part of this study applies these bioinformatic tools in a biological context. An analysis of Abf1 protein binding datasets has been undertaken, revealing many more binding sites than had previously been identified. Analysis of the sequences at these binding sites identifed the previously determined consensus binding motif in only a subset, with no novel motif identifiable in the remainder, suggesting binding may be in uenced by factors other than sequence.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: QH426 Genetics ; RC0254 Neoplasms. Tumors. Oncology (including Cancer)