About the Project
array of <50 m pixels across a biological tissue section. The resulting spectra mostly comprise lipid-derived ions and can
be used a characteristic fingerprints to identify tissue phenotypes. Tissue images can be produced by selecting a single ion
from the spectrum that corresponds to each pixel. Multivariate statistical tools can be used to seek spectral correlations
between pixels and to base images on sets of common spectral features rather than single ions, potentially revealing a
number of distinct tissue phenotypes. Hierarchical cluster analysis and principal component analysis (PCA) have been used
in this context, but whilst they are computationally straightforward, they sacrifice a fraction of the available information.
PCA, for example, requires that the different components are orthogonal and provides only a qualitative segmentation of
the image. Independent component analysis does not limit component spectra in this way and allows for a degree of
commonality in the spectral features of components. To allow quantitative treatment of the data, we will use an
assumption of Poisson variability in the image signals that better describes the data than the more widely used Gaussian
assumption. Combining Linear Poisson Models of the data with Independent Component Analysis allows the tissue
phenotypes (components) identified to be quantified and for the measurement errors to be determined. We will apply this
powerful quantitative tool to the analysis of tissue samples of relevance to a range of tumour types in order to investigate
tumour heterogeneity through the identification of their component tissue phenotypes.
For international students we also offer a unique 4 year PhD programme that gives you the opportunity to undertake an accredited Teaching Certificate whilst carrying out an independent research project across a range of biological, medical and health sciences. For more information please visit www.internationalphd.manchester.ac.uk.
related area / subject. Candidates with experience in mass spectrometry and/ or with an interest in multivariate data
analysis especially as applied to image analysis are encouraged to apply.
This project has a Band 2 fee. Details of our different fee bands can be found on our website (https://www.bmh.manchester.ac.uk/study/research/fees/). For information on how to apply for this project, please visit the Faculty of Biology, Medicine and Health Doctoral Academy website (https://www.bmh.manchester.ac.uk/study/research/apply/).
Informal enquiries made directly to the primary supervisor.
Mass Spectrometry 428: 62-70.
Deepaisarn, S., et al. (2018). "Quantifying biological samples using Linear Poisson Independent Component Analysis for MALDIToF
mass spectra." Bioinformatics 34(6): 1001-1008.
Henderson, F., et al. (2018). "Multi-modal imaging of long-term recovery post-stroke by positron emission tomography and
matrix-assisted laser desorption/ionisation mass spectrometry." Rapid Communications in Mass Spectrometry 32(9): 721-729.
Based on your current searches we recommend the following search filters.
Based on your current search criteria we thought you might be interested in these.
Mathematical Modelling of Neurovascular function: Developing and refining brain tissue models to enable world leading research into neurovascular coupling in health and disease
University of Sheffield