Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.731443
Title: Some spatial statistical techniques with applications to cellular imaging data
Author: Honnor, Thomas R.
ISNI:       0000 0004 6496 7739
Awarding Body: University of Warwick
Current Institution: University of Warwick
Date of Award: 2017
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Thesis embargoed until 10 Nov 2019
Access from Institution:
Abstract:
The aim of this thesis is to provide techniques for the analysis of a variety of types of spatial data, each corresponding to one of three biological questions on the function of the protein TACC3 during mitosis. A starting point in each investigation is the interpretation of the biological question and understanding of the form of the available data, from which a mathematical representation of data and corresponding statistical problem are developed. The thesis begins with description of a methodology for application to two collections of (marked) point patterns to determine the significance of differences in their structure, achieved through comparison of summary statistics and quantification of the significance of such differences by permutation tests. A methodology is then proposed for application to a pair spatio-temporal processes to estimate their individual temporal evolutions, including ideas from optimal transportation theory, and a test of dependence between such estimators. The thesis concludes with a proposed model for line data, designed to approximate the mitotic spindle structure using trajectories on the surface of spheroids, and a comparison score to compare model t between models and/or observations. The results of methodologies when applied to simulated data are presented as part of investigations into their validity and power. Application to biological data indicates that TACC3 influences microtubule structure during mitosis at a range of scales, supporting and extending previous investigations. Each of the methodologies is designed to require minimal assumptions and numbers of parameters, resulting in techniques which may be applied more widely to similar biological data from additional experiments or data arising from other fields.
Supervisor: Not available Sponsor: Engineering and Physical Sciences Research Council
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.731443  DOI: Not available
Keywords: TK Electrical engineering. Electronics Nuclear engineering
Share: